Telmar.com Telmar.com eTelmar.net Home Page The Advertising Media Internet Center

Telmar Home Page Telmar.com

 

Media Guru

Guru Search Results: 146 matches were found

Thursday, January 31, 2002 #5039
calculate reach and frequency

The Media Guru Answers(Thursday, January 31, 2002 ):
You need a computer with software such as that offered by Telmar.

The calculation is extremely complex. As input, you need average issue audience, duplication between issues of the same publication and duplication between each possible pair of different publications. These must be combined using a complex formula such as the Beta-binomial function. There are variants of this formula, which might be preferred, depending on media type and other variables


Wednesday, January 23, 2002 #5028
How do you calculate commercial wearout?

The Media Guru Answers(Thursday, January 31, 2002 ):
Wear out is not a matter of calculation, although some develop rules based on their brands' experience..

Click here to see further Guru comment on wear out.


Tuesday, January 22, 2002 #5026
Now that Outdoor can be mixed with other media, what is your thinking on how do the number of uses effect the frequency distribution? Should we be transferring GRP's, number of days or number of boards times days? How does this effect the frequency of the programs?

The Media Guru Answers(Thursday, January 31, 2002 ):
Outdoor could always be mixed with other media, so the Guru presumes you mean that the media software you use now has the ability to calculate reach and frequency for the combined media. Your question is probably answered in the software's manual.


Tuesday, January 22, 2002 #5025
What do I need to calculate reach for print and what is the formula?

The Media Guru Answers(Wednesday, January 23, 2002 ):
You need a computer with software such as that offered by Telmar.

The calculation is extremely complex. As input, you need average issue audience, duplication between issues of the same publication and duplication between each possible pair of different publications. These must be combined using a complex formula such as the Beta-binomial function. There are variants of this formula, which might be preferred, depending on media type and other variables


Sunday, January 06, 2002 #4981
I am trying to compare cpm between print and radio. The demo is Adults 18-54. I have the readership for the demo for the MSA for the print. I have the gross impressions for the radio and have calculated a cpm for all (both the print and radio). Is this correct? Can I compare readership against GI's?

The Media Guru Answers(Sunday, January 06, 2002 ):
Yes, readership is equivalent to impressions.


Thursday, January 03, 2002 #4971
Dear Guru - Was there ever a "chart" that enabled media buyers to calculate reach/freq, gross impressions etc for broadcast television planning. I have been explaining to someone that we use programs for this kind of thing, but this person seems to remember using a chart and thinks i should be able to do this manually if he could. I've never heard of it, have you? He would have been planning around 1975. Thanks.

The Media Guru Answers(Thursday, January 03, 2002 ):
Yes, before computers became common in the 80's, when there were just 3 networks, with 90%+ share, no cable, and few independent stations, R&F tables were the way it was done. Every few years, using Nielsen cume studies of actual scehdules, average reaches for various GRP levels were calculated. There might be variables for the number of programs or episodes used. In this way all possibilities for a daypart could be displayed on a single, typed page.

Today, with computers on every desk, 6 broadcast networks amassing only 50% share, dozens of cable options and hundreds of independent stations, accuracy requires computer systems. Such crude tables could be still constructed, but why bother when computers and software are so readily available?

The Guru who was using the charts in the 60's, is happy with his computer today.


Wednesday, December 05, 2001 #4928
Guru, how do I calculate a full page ad price if the directory only gives a column inch price? Thank you

The Media Guru Answers(Wednesday, December 05, 2001 ):
The direcory should also give number of column inches or inches and columns per page. Then you can mulitply, but pages are usually priced at some discount vs their column-inch multiple.

It sounds as if you are looking at classified rates, when you want to buy display advertising.


Tuesday, November 27, 2001 #4912
Dear Guru, Could you please clarify this for me: 1) is the ad production cost included in the 15% agency commision or it it charged on top of that? 2) the usual 5% for media buying, is it paid out of the 15%agency commission or in addition to that? 3) Is there any approximate scale to calculate how agency commission increases or decreases depending on the overall media cost volume? Thank you very much

The Media Guru Answers(Tuesday, November 27, 2001 ):
The traditional 15% is the agency's commission on any outlay made on the clients' behalf, whether media purchase or production cost.

One difference is that media costs are usually quoted by the media at "gross" which includes 15% commission, while the agency needs to "gross up" production, by adding 17.65%, creating a number that is then 85% cost and 15% commission.

That is, when media has a gross cost of $1000, the agency gets $150 commission and the media vendor nets $850. When there is a production expense of $850, the agency grosses it up by 17.65%, or $150, for a total of $1000.

While 5% is a common talking number for media services, it is not a standard. It is part of the 15%.

Fees other than the traditional 15% are a matter of negotiation. They may be more than 15 on low-billing accounts and considerably less on large accounts. They may be something completely different than percentage-based. The American Association of Advertising Agencies is likely to have a reoprt on averages and ratios.


Wednesday, November 21, 2001 #4902
We are trying to figure out a gross cpm for a client who we have given a net cpm to. She is looking to compare the two. (I am not speaking in terms of gross/net rate -agency 15%) Here is a scenario that I am trying to figure out: By publication - 439,000 target audience $51,000 FP rate = $116 net cpm. How would we go about getting the gross cpm to compare to this? From a total plan standpoint, we also presented net cpms and the client is looking for a gross cpm for comparison. How exactly would I go about figuring this out?

The Media Guru Answers(Wednesday, November 21, 2001 ):
The Guru does not understand your question. The common definition of the terms gross rate / net rate referes to the difference between a rate including agency commission (traditionally 15%) and a rate not including this commission. This definition would yield a $135.29 gross cpm for $116 net. The only reason the gross would be different would be if there was some other understanding about agency commission / compensation.

The formula to calculate gross cpm, when commission is 15% is:
net cpm ÷ (1 - 0.15).


Tuesday, November 13, 2001 #4883
hi, i've been dealing with an ad agency and fighting for a concept event with a big cost, part of the event is a promotion to require a proof of purchase in order to join. They say they need to know the media implications of this, they need to measure if it will increase sales, and if the media planning is effective. I'm just curious, how do i make an effective media plan to justify that the cost of the event is just minimal also plus the media mileages they get from posters, streamers, and merchandizing.

The Media Guru Answers(Wednesday, November 14, 2001 ):
The simplest to calculate and for them to understand is to determine the number of exposures of all the elements and then put a value on these exposures using a cpm the agency experiences in some cmparable medium, such as out-of-home.


Thursday, November 01, 2001 #4857
how would i calculate US media spill into canada beyond paid circulation

The Media Guru Answers(Thursday, November 01, 2001 ):
"Spill" is not a calculation, it is a measurement. The various media measurement and auditing organizations may or may not publish audience in Canada for the U.S. media they cover. Visit Nielsen, Arbitron, MRI, etc.


Wednesday, August 15, 2001 #4658
Guru, Could use some help framing questions for my agency relating to the effectiveness of a media campaign. We recently ran a test cell for a new campaign (our first)in which the agency provided information on total TRPs, total reach and total frequency over the life of the test. I need to determine how the frequency builds over time. Are there any formulas/rules of thumb for calculating build over time? If not, what specifically should I ask them for? Thanks.

The Media Guru Answers(Wednesday, August 15, 2001 ):
Reach relates to GRP in a curvilinear function. Frequency relates to GRP in a straight line. This doesn't mean that each week adds the same amount of frequency, merely that it's fairly easy to work with.

The easiest thing however, is probably to ask the agency to calculate cumulative reach and frequency, week by week, over the course of the campaign.


Saturday, August 11, 2001 #4650
can you give me the formulae to calculate the OTS viv a vis competition in the print media.

The Media Guru Answers(Saturday, August 11, 2001 ):
The question isn't quite clear: If you mean the formula to determine what your own OTC goalshould be vis-a-vis competition, it is very simple:
If Y= your goal OTS and
C = competitor's OTS then
Y > C

If you mean a formula to calculate your competitors current OTS, you need to have quantities from reports before you can determine what formulae to apply. For example, do you know spending ($)and average cost per OTS (O)? Then the formula is
$ ÷ OTS


Tuesday, July 31, 2001 #4621
Hello Media Guru Is there software available that will have reach and frequency information for Trade publications. If not what is the best way to calculate this information?

The Media Guru Answers(Tuesday, July 31, 2001 ):
Programs like Telmar's print planning systems can process Intelliquest (computer and tech trades), as well as some others which exist in the medical and other fields. The software can also estimate R&F for other, unmeasured trade titles if you have circulation and reader-per-copy estimates.


Monday, July 23, 2001 #4598
Dear Guru, I need to compare the W25-54 gross rating points of national consumer magazines to the W25-54 gross rating points of a network television schedule. I am concerned about keeping “apples to apples,” when making this comparison. In doing so, it seems that it would be necessary to calculate GRPs based on the same universe (for both magazines and television). The television universe that I am using is provided by Nielsen and is based on W25-54 who live in a household with a television, which I think is roughly 98% of the population. My question is: Is it appropriate for my print universe to be based on W25-54 living in a TV HH (98%), OR should my print universe be based on 100% of the W25-54 population? Also, can I calculate the rating points of magazines individually and then add up my schedule to determine GRPs OR must I calculate magazine rating points based on a schedule in order to prevent duplication of W25-54? I am doing these calculations by hand and do not have access to MRI – outside of requesting information, such as a magazine’s W25-54 audience, from a magazine sales rep. Any help would be greatly appreciated! Thank you.

The Media Guru Answers(Tuesday, July 24, 2001 ):
It would be simpler to compare impressions and not worry about the slight universe differences. If you must use GRP, adjust TV grp downward according to the TV penetration factor.

GRPs disregard duplication, so simple addition is fine.


Monday, July 09, 2001 #4562
Dear Guru, Our client has asked me to produce a "payout scenario" that they would be able to expect in sales based on a national plan that I have done for them. I don't know what the creative will look like because it is done in house. How can I calculate the sales potential of a media plan? Thanks again.

The Media Guru Answers(Wednesday, July 11, 2001 ):
Except in direct reposnse scenarios with considerable past history, this can only be very approximate. Try the following steps:
  • What percent of the target will buy the product over the period of the plan if there is no plan?
  • What percent of persons will you reach at effective levels during the plan?
  • What percent of non-target persons will buy the product over the period of the plan if there is no plan?
  • What percent of non-target persons will you reach at effective levels during the plan?
  • Can you assume that current non-users will be moved to purchase by exposure to the plan?

Of course a lot depends on whether the advertising is aimed at new-user trial, increasing use among currentusers, using up, etc.

Once you have all these assumptions organized, then comparing the value of projected added sales to the cost of advertising leads to payout estimates.


Friday, July 06, 2001 #4555
Our agency handles a lot of business to business accounts how would one go about calcualting reach and frequency for each particular business sector ex. one account makes catheters. How would you calculate in various value-added opportunities into reach and frequency like links on a site, direct mail lists etc. Thanks for the great service.

The Media Guru Answers(Friday, July 06, 2001 ):
To calculate reach and frequency two data points are necessary:
Unduplicated audience within the target (sector, in your case) and total population for the target. The media type or unit size are not relevant; reach is pure arithmetic; relative impact and other creative judgements are separate.

It should be basic to estimate the numbers of audience for any media vehicle, site or mailing in your plan. You certainly must have an idea of the size of the sectors you are targeting. The tricky part would be estimating the duplication between advertisements. In the medical field, possibly PERQ has some useful estimates.

Once you add the gross audiecne of all your ads and eloiminate the estimated duplication, you divide by the population to determine reach.


Tuesday, July 03, 2001 #4544
how do I calculate an index score for a marketing survey I've completed?

The Media Guru Answers(Thursday, July 05, 2001 ):
An index score compares measured behavior to a norm. The behavior and the norm are typically percentages of defined groups and the index is a relative percentage, expressed without decimals or percent signs.

For example, if your survey shows that 30% of persons 21-34 prefer Brand X beer, while 20% of everyone over age 21 prefers Brand X, then the 21-34 index of preference is

30 ÷ 20
or
1.50
Or
150%
or
an index of 150


Friday, June 29, 2001 #4538
Hello again, I have two questions about calculating reach and frequency that I have been unable to find in the archives of past responses. Perhaps you can help? 1. I normally use the formula (a+b)-(.a*b) to determine combined reach of two mediums, such as radio and print. How do I calculate the combined reach of more than two? The plan I am working on includes spot TV, spot radio and local newspaper. 2. Is it possible to determine a combined reach for more than one market or should each market be reported separately? In the past, I have provided separate delivery for each market in the same plan with a total number of gross impressions for the whole plan. Is this correct? Thanks in advance!

The Media Guru Answers(Friday, June 29, 2001 ):
1. This common formula is based on an assumption that different media duplicate their audiences according to random probability. Therefore if you follow this assumption, media may be added to combinations of media in a "chain" of the same formula. So, once you have combined TV and Radio, you can use this combination as your "a" and then combine it with newspaper as "b."

2. You can combine reaches across markets by doing a weighted average. Multiply the reach in each market by the percent of U.S. in each market. Add all the products and divide by the sum of the % U.S.


Wednesday, June 27, 2001 #4526
Please tell me what is the way for calculating the media commission for a client for media planning, media buying... How are calculated the 2,5%, the 5% etc.? Thank you

The Media Guru Answers(Wednesday, June 27, 2001 ):
Usually the percentage is based on gross media cost. Gross media cost is the "commissionable" rate established by the media vendor to be charged to advertising agencies. This rate is structured so that 15% of it is the commission and 85% is the "net" to the medium. If the gross rate is $100, then 5% commission for media buying is $5 or 0.05 x $100.


Wednesday, June 20, 2001 #4506
I am reading your SQAD Report on "Costs per TV Household Rating Point". I am a novice at this. What does this mean and how do I read this report? Thank you, Michael Jordan

The Media Guru Answers(Saturday, June 23, 2001 ):
A "rating point" is equivalent to one percent of the population in the specifed universe. So, if a commerical has an audience of 1000 in a market with 100,000 housesehold, it has a rating of 1. If the audeince is five times as large it has a rating of 5. If you have a schedule consisting of five commericals like this, you would have 25 Gross Rating Points (GRPs).

If you divide the cost of the schedule by the GRPs, the resulkt is the Cost per TV Household rating point. SQAD collects the actual buying records of several agencies and other media buyers to calculate the marketplace averages.


Sunday, June 17, 2001 #4492
Hi Guru - I have a few cable TV questions. 1. Can reach/frequency estimates be done for cable schedules. My rep keeps giving me everything but. 2. Does Nielen measure all cable stations. 3. Why can't I get FX numbers on Telmar, just ESPN. 4. If I can get R/F for cable what are some of the major differences from Network numbers. Thanks.

The Media Guru Answers(Sunday, June 17, 2001 ):
  1. Yes they can. Some smaller networks may not have the facilities to calculate R&F, but that doesn't seem likely.
  2. Yes, but not all cable networks have enough measured audience to be considered reportable by Nielsen
  3. Telmar systems which use your own Nielsen tape data will allow you to examine any reportable network. Systems like Market Maestro, which use established generalized data can only incorporate networks old enough and large enough at the time of the system update to have establish patterns, but not all the 100+ imaginable networks
  4. Because cable ratings are a fraction of broadcast ratings, and turnover is less, cable reaches cume lower in relation to GRPs. SInce cable universes are smaller than broadcast, reach potential is lower as well


Monday, May 21, 2001 #4418
I am just starting a job dealing with Direct Response Television Advertising. Seeing that traditional TV media rates are valued based on ratings points. How do stations develop their rates for Direct Response Television? Is Cost per Thousand a measure which can be used in Direct Response Television analysis or is the measure just Cost per Order? I have a client which swears that I need to give him in depth Cost per Thousand numbers but our agency owner says that is ludicrous in DRTV. Thanks for any info.

The Media Guru Answers(Tuesday, May 22, 2001 ):
Cost per thousand (cpm) is mereley the cost of a spot divided by its audience. All station inventory is essentially priced at least partially based on a cost/audience ratio, which determines the value of the spot to the station.

However, the value to a DR advertiser is based on cost per response, whether that is measured in inquiries or orders. Most DR practicioners have learned that there is little relationship between audience size and response. It is not unusual to get more orders from a low-rated late fringe program than a prime time program.

So, while it is possible to calculate cpm in a DR buy, and inherently harmless to report cpms, it would be wrong to judge a buy on this standard. If your owner objects to the waste of your time and the client's, he is justified in his objections.


Friday, May 04, 2001 #4368
Media Guru, please help. How do I calculate reach and frequency for a two-week, two-newspaper buy? We are placing 4 ads per week (total of 8 ads for the schedule) on Newspaper #1, which has a maximum reach of 9% of our target. Newspaper #2 will carry 2 ads per week (4 ads for the schedule) with a maximum reach of 23% of our target. Please advise. Thanks!

The Media Guru Answers(Sunday, May 06, 2001 ):
Find some example newspaper R&Fs at The Newspaper Advertsing Associations Marketscope site.

In very general terms, you can estimate some parameters. If newspaper A has a 9% maximum reach, it probably has a single copy reach of around 7%.

If B has a maximum of 23%, then it likely has single copy reach around 20%. So the outside bounds of reach for your schedule are a minimum of 20, but more likely closer to 25, the random combination of the two papers' single copies. The outside maximum is 32 ( the 9% plus the 23% maxima), but more likely closer to 30 (random again).

A solid estimate of 25-30 reach for your schedule should be good enough, but you could use the eTelmar pay-per-use system for a specific calculation.

Frequency, of course, will be the sum of the single copy audeinces of all insertions (GRP) ÷ the reach estimate.


Tuesday, March 20, 2001 #4273
Dear Guru, Can you guess at an estimated reach for a 4-week radio schedule at 150 TRP/WK (W2554)? Also, do you happen to know the formula for manually figuring reach/frequency? Thanks for your help!

The Media Guru Answers(Tuesday, March 20, 2001 ):
The formulas for radio reach are enormously complex and must take into account number of stations, daypart mix, average rating and other variables.

Your 600 GRP schedule might generate (roughly) from 35 to 70 reach, depending on these variables.

Click here to see past Guru responses abourt calculating reach.


Tuesday, March 20, 2001 #4272
How do you calculate CPM?

The Media Guru Answers(Tuesday, March 20, 2001 ):
Cost ÷ audience impressions in thousands


Tuesday, March 13, 2001 #4251
Hello Media Guru -- Hopefully you can help me. We are currently in planning and we are analyzing our competitiors When - In Avg Weekly and Avg 4 wk deliveries by Quarter and Full Year. We have pulled their GRPs by week for Network by daypart, Syndicated and Cable so we have the National TV deliveries. We have pulled their Print schedules off of Stradegy and now want to come up w/the same deliveries (When-in Avg weekly and Avg 4 wk by Quarter and Full Year)for Print in order to combine the TV and Print deliveries. Is there a method that you suggest? Thank you in advance-- Bridget

The Media Guru Answers(Tuesday, March 13, 2001 ):
It isn't clear which steps cause your problems.

If you're being strict, for average 4-week a simple arithmetic adjustment from average month to average 28 days will suffice. For average print week, you could take the year's schedule and divide it into 52 roughly equal groups, then average the R&F of all 52. If print is flighted, then you should calculate for active periods and average with zero weight for as much time as there is hiatus.


Saturday, February 17, 2001 #4192
What is brand visibilty index ? How do you calculate Brand Visibilty Index from Media Planning perspective ? Please state examples ?

The Media Guru Answers(Monday, February 19, 2001 ):
"Brand Visibilty Index" is not a standardized media term. It might be a term invented by one agency or advertsing school to indicate a specific concept they use in describing some situation. It might be an index of Brand GRP versus category average GRP. Or it might be something else based on awareness, clutter, etc.


Monday, January 08, 2001 #4087
Guru, First off, just wanted to let you know that I find this to be one of the most usefull sites on the web - as a management consultant in need of a crash course on media planning, the information found in these pages has proven invaluable...Now, on to my question: I am working on the launch of a branded consumer services play (auto related), and am trying to build a marketing budget from the bottom up, rather than as a strict % of sales. I have modeled an overly simplified media plan, and am looking for guidance on placeholders to use for weights (TRP) for TV and Radio, # of weekly inserts for newspaper, and showing level for outdoor. I know there are numerous factors and considerations I am leaving out (I know the GURU doesn't like sweeping generalizations), but I need a place to start. Goal: generate "substantial awareness" (think Midas, Maaco). Thanks for your insights.

The Media Guru Answers(Thursday, January 11, 2001 ):
The Guru thanks you for the compliments.

Keep in mind that while "substantial awareness" may be a snappy phrase for discussion of plans, you need to quantify such a term in order to quantify the building blocks of getting there.

Let's suppose we decide the goal is 80% ad awareness among the target within a given campaign period. Therefore, your advertising must reach at least 80% of the target in that period, with enough frequency for the message to penetrate and stick, let's say at least three times.

Now, you can calculate that generating that reach in TV will call for a certain number of TRP (you can use the media software at eTelmar for calculations). Or you can examine getting that reach with radio or a combination of TV and radio.

Outdoor will generate high reach more efficiently than either, with a #25 showing, but outdoor's necessary simplicity of message may not stand alone in filling your needs.

Newspaper has its own contribution and you need to judge from a marketing perspective whther you need a small store-locator ad every day, a full page branding message once a week, or some other approach, if any.


Thursday, December 07, 2000 #4026
Dear Guru, If there are 2 different target groups need to be sufficient in one TV schedule, the weight of rating for Target A is 40%, for Target B is 60%. how to calculate/evaluate the performance? and does it have a standard formula? Thanks for your answer in advance. Sara Shiung

The Media Guru Answers(Friday, December 08, 2000 ):
If the Guru correctly understands your query, it seems like a simple weighted average would work.


Monday, November 06, 2000 #3945
How and when do you do a weight average?

The Media Guru Answers(Tuesday, November 07, 2000 ):
Weighted averages are used to calculate the overall average when you have percentages of various groups that are parts of the larger whole. Click here to see past Guru responses using weighted averages.


Tuesday, September 12, 2000 #3790
Hello, How can I find out what companies market to teens and what percent of their marketing budget is allocated to the teen segment? I am also interested in how to find customer acquisition costs per company and per industry. Some companies publish this information in press releases (such as Amazon.com's $19 cost per new customer) or in the annual report (for public companies only). Is there a source where I can find these costs? Thanks.

The Media Guru Answers(Saturday, September 16, 2000 ):
The easiest way to find companies marketing to teens is to skim through some teen-oriented magazines at your library. Of course, this won't be complete, so the next step is to use a tool like CMR (Competitive Media Reports) to track advertisers in other teen oriented media. Most companies would treat the info you want as confidential, so only public or monitored behavior such as the Guru has mentioned will be available in most cases.

Cusotmer acquisrition costs are only relevant for certain industries and businesses like Amazon whose direct sales methods allow tracking. Otherwise you will need to find marketing expenditure and customer # estimates from sources like MRI to calculate this yourself.

In some industries, the trade media may have reported on this.


Tuesday, August 15, 2000 #3705
A retail client of mine is planning a short media campaign to support a 4-day event (Thu-Sun). I'd like to show a reach curve or the like to illustrate the reach built by a 7-day media support (Mon-Sun) and a 4 1/2 day support (Wed - Sun Noon). Vehicles will be radio, TV, and newspaper. How can I do this?

The Media Guru Answers(Friday, August 18, 2000 ):
In total, there is not likely to be much demonstrable difference between identical quantities of media delivered over a Monday-Sunday period.

If you want to illustrate reach accumulation day by day, then you can calculate the reach of the schedule that runs on the first day of the campaign in each case, the schedule that runs on the first two days of each campagn, etc. Then you can plot the two curves, with reach on the Y axis and days on the X axis, using the charting tools of Excel, Powerpoint, Corel, or whatever you might have.

It might look something like the chart below (not actual reach data).


Wednesday, August 02, 2000 #3666
Ref. question 3663 Thanx for answering my question. I buy slots with high eff. index when my objective is to accumulate GRP's and drill my message into my consumers mind. This is the secondary stage where after creating the initial reach i focus on accumulating greatest total number of impressions (Funnel Treatment). As for the decay factor it reflects the decrease in the recall leval when advertising is reduced or stoped. I normally use 10% decay level in IMphase(IM horizontal planning technologies) The question that i want to ask you is what is the better way of flighting. There is a 70's 3+ eff frequency model by Prof. MacDonald which says that brusting is a better flighting patteren.On the other hand there is more recent Recency concept championed by Prof. JP Jones of Syracuse university of NY which says that as far as FMCG goods are concerned people are in the market every week and infect only needs one OTS to stimulate purchase.Please comment MY second question is how do you calculate Eff Frequency. Normally i use Eff frequency model where i calculate the eff frequency by applying judgement and common sence in a disciplined manner using Marketing, Advertising and competitive factors Thanx Sarwar Khan Media Manager R-Lintas Lahore,Pakistan

The Media Guru Answers(Sunday, August 06, 2000 ):
1. In regard to 3+ effective frequency versus recency, the Guru tends to favor recency for "Fast Moving Consumer Goods." Recency is not really a contrast to the 3+ frequency theory, but an extension. As championed by Erwin Ephron, a core concept of recency is that once the third exposure is delivered, all additional exposures are at 3+.

2. Once again, there seems to be a semantic issue when you say "calculate" effective frequency. If you mean setting the frequency level to be considered effective, then your "judgment and common sence in a disciplined manner using Marketing, Advertising and competitive factors are the right approach, and the Ostrow Model will be helpful.

If instead, you mean to calculate the effective frequency delivered by your schedule, this has absolutely nothing to do with the subjective factors you have listed. A reach model determines how many persons are exposed to each discrete number of ad units in the schedule. That is if your reach is 75%, that means, explicitly, that 75% of the target has experienced one or more ad exposures. Within this, perhaps 70% of the target has been exposed to 2 or more, 66% to 3 or more, etc, up to the full number of units in the schedule. Reach models allow for expressing all of these levels. "Effective reach" mean those reached at least the minimum number of times established as effective, most typically 3.


Saturday, July 29, 2000 #3663
Dear Media Guru I am a media planner from Pakistan.I need to ask what are the possible comparison tools that we can use while planning for different programs on television.At the moment while planning i calculate cost index, rating index, efficiency index, Avg GRP's, Maximum reach, and avg.viewing miniutes for each time slot. Normally i advertise in time slots with high effeciency index, is this a good comparrison tool for planning or not. Normally the decay factor that i take is 10% is this OK or not. What are the different possible ways to break the adverising clutter on television and increase the possibility of high ad exposure. Thax in anticipation Sarwar Khan Media Manager R-Lintas (Pvt.)ltd. Lahore Pakistan

The Media Guru Answers(Saturday, July 29, 2000 ):
It always fascinates the Guru that countries sharing a common language can use it quite differently when applying it to the jargon of a particular business or interest.

What you are describing as "planning" seems to the Guru to be what he would regard as a buyer's selecting a schedule after a plan has been approved. You haven't mentioned what goals you are pursuing with your schedules. Selecting spots with the best efficiency index (audience versus cost) will get you the greatest total number of impressions, but possibly not the greatest net reach. The best rating is more often likely to lead to high reach, but perhaps not without due regard to efficiency and duplication.

"Decay factor" is an unfamiliar term to the Guru. "Maximum reach" and "average viewing minutes" don't seem relevant to assessing individual spots as the Guru understands the terms.

Overall, the Guru believes you should be comparing possible schedules, rather than individual spots to accomplish planning goals.

Optimizers serve this purpose, but running reach analyses of several schedules can get you there, as well.


Wednesday, July 26, 2000 #3654
Please provide formula to manually calculate Reach & Frequency for press. Thanks

The Media Guru Answers(Saturday, July 29, 2000 ):
This calculation is very complicated. If you don't have detailed tables of duplication factors between different publications and between various numbers of multiple issues of the same publication, only fairly crude formulae are available.

Click here to see past Guru responses about reach calculation formulae.


Tuesday, June 13, 2000 #3548
I am in the process of evaluating a print proposal submitted by a business to business annual register with company listings/profiles, accessible by category. In addition to receiving a P4C ad, my client wil also receive 8 bold type listings with descriptive information and 4 bold type listings(company name and phone # only) throughout various sections of the register. At first glance the package looks like a great idea. The circulation is nearly 100% targeted, the CPM (based on the P4C alone) is very low, and there are additional merchandising perks that will expose my client to their target for one full year. The problem is, I must put a "value" on each component of the package. Do you have any ideas on how to place a value on the "bold type listings" described above?

The Media Guru Answers(Thursday, June 15, 2000 ):
Your situation is analagous to evaluating reach versus GRPs or a full commerical in a special versus billboards.

Since the deal seems efficient and effective simplay based on the P4C, any value you give to the other elements can be arbitrary and will be just for the sake of dicsussion. Why not calculate the impressions of all the other elements and price them at 25% of the P4C cpm?


Monday, June 12, 2000 #3547
I am buying radio in two different markets - one is a large market which is measured by Arbitron. The other is a small market where I get the ratings through Arbitron county measuring. The two cities are only 45 miles apart and there is a large amount of radio overlap. Is there any way to figure an accurate combined reach & frequency? Thanks.

The Media Guru Answers(Monday, June 12, 2000 ):
First, define "market." If these radio markets are both in the same DMA, and you want DMA R&F, add the two stations' reach in thousands and divide by DMA universe. If they are in two different Metros, calculate reach within each and do a weighted average of the two:
  • Metro "A" target population = 100,000
  • Metro "B" target population = 20,000
  • Metro "A" target reach = 40% (40,000)
  • Metro "B" target reach = 55% (11,000)
  • Combined, total coverage area reach = 40,000 + 11,000 ÷ 100,000 + 20,000, or 42.5%


Monday, June 05, 2000 #3532
At what market penetration level does buying TV on a national level become more efficient than spot buying on a local level? Is it the same for Radio? How about Newspaper? And secondly, is there a way to calculate this in general?

The Media Guru Answers(Sunday, June 11, 2000 ):
In TV the variables are demographic and daypart. Some demographics have a greater differential in spot vs network CPP. One daypart / demographic scenario may become more efficient in network after 25 markets, another one at 75.

For example, in one recent cost guide which the Guru has on hand, the daytime HH CPP for network was equivalent to daytime spot CPP for the the top 68 markets. In Prime, the Network HH CPP was equal to top 22 markets' spot.

For other demographics and other media the breakeven will be different still. There is no rule of thumb beyond experience. You need to compile spot costs and determine where they break even versus national.


Friday, May 26, 2000 #3499
What is common way to calculate tv programm loyalty?

The Media Guru Answers(Monday, May 29, 2000 ):
The usual basis is "how many episodes out of 4 does the viewer watch."


Wednesday, May 24, 2000 #3494
Hi Guru, I have a few, related r/f questions: Can you please explain the technical differences between r/f's calculated using the Metheringham, Mellow and Prince methods? Are there cases where one methodology would be more favorable than the others and what are the cases. And, finally, is there any value to runnning a yearly r/f? THANKS !!

The Media Guru Answers(Wednesday, May 24, 2000 ):
You appear to be referring to magazine r/f in particular.

Metheringham is one of the oldest print r/f methods, based on duplication between titles and duplication between issues.

Mello is an extension of Metheringham created by MRI. The Guru confesses to being unfamiliar with the "Prince" method.

"Most favorable" would be a case-by-case judgement call. Assumong you have all three systems available, you can do the same schedule 3 ways and compare. The Guru doesn't think there are general cases that can be cited where one is superior to another. Most likely teh differences will be caused by multi title versus few titles/multi-insertion schedules, or weekly versus monthly.

Yearly r/f, when schedules are reasonably substantial, is not likely to be useful in comparing scehdules. Some advertisers might use yearly data when looking for box-car numbers for trade promotion or sales meetings.


Wednesday, May 24, 2000 #3492
Dear Guru, How the results of TV sposnsorship can be measured and evaluated? Is it possible to compare it somehow with ordinary spots advertising? In Russia we have monitoring of TV channels. All events - programmes, breaks & commercials (all types - including spots, sponsorship etc) are fixed in the database. Combining it with the PeopleMeter ratings theoretically we can calculate everything. The main problem is that I do not know how can I calculate rating for sponsorship. Should it be rating of the whole program or rating of the real moments of sponsorship or something else? Maybe you can advise some literature about the subject? how it is done in other countries? Thank you in advance Ksenia

The Media Guru Answers(Monday, May 29, 2000 ):
Generally, when considering the total commerical audience of a sponsored program, the total program audience rating is used. But if your people meter data allow you to accumulate the net of all commercials and sponsor mentions, use that.

Evaluating the total benefit of sponsorship goes beyond these data. The Advertising Research Foundation InfoCenter. For details about the InfoCenter, call 212-751-5656, extension 230. and ESOMAR, the European Survey, Opinion and Market Research Organization would have relevant research.


Tuesday, May 16, 2000 #3479
Are there parameters (highs and lows) for effective reach and frequency? In other words, is there a particular reach and a particular frequency that are considered "average" as they relate to broadcast media? How would one determine whether an advertiser is spending adequate funds to meet these "averages" when airing a broadcast schedule on a Mon-Sun basis?

The Media Guru Answers(Friday, May 19, 2000 ):
The Guru finds the concept of average irrelevant in this context.Such measures are relevant in relation to competition and one's own communications goals. What does it benefit an auto brand if the "average" advertiser has a reach of 50% at 3+ frequency when all automotive competitors are delivering 75% at 3+?

As to turning spending into effective reach and frequency, that's typically part of a media plan. Budget gets expressed as schedules of TV, radio, print, etc. Reach and frequency are calculated by available software for these GRPs. Effective reach / frequency is an inherent part of the calculation.


Friday, May 12, 2000 #3464
I am trying to figure out an answer to a complcated problem. Say you have widgets. you have identified 4 markets that might buy widgets. you then have a number X for each market that shows the possible number of sites that could have widgets - how can you create a weighting scheme to balance the data by segment? total # Market possible sites % have widgets A 10,000 55% B 25,000 25% C 5,000 5% D 10,000 90% please reply to sjedwards@rcn.com.

The Media Guru Answers(Friday, May 19, 2000 ):
With the data you have supplied the only approach would be to multiply sites by percents and calculate a new distribution. So:
  • A: 10,000 x 55% = 5,500
  • B: 25,000 x 25% = 6,250
  • C: 5,000 x 5% = 250
  • D: 10,000 x 90% = 9,000
  • 5,500
    6,250
    250
    +9,000
    21,000
  • and thus
    A=26%
    B= 30%
    C=1% and
    D= 43%


Tuesday, May 02, 2000 #3439
Regarding effective reach and effective frequency, are there general accepted boundaries of these measurements as they relate to radio and television? How do you compute effective reach and frequency?

The Media Guru Answers(Thursday, May 04, 2000 ):
The Guru has seen effective frequencies from 2 to 9 used in plans. Most often, 3 is the "bogie" but 4 and 5 are not uncommon.

In the Guru's opinion, the effective levels make sense when applied to a majority of the target, that is, 50%+.

As far as computing effective R&F, the capability is typically built into reach and frequency calculators. As part of calculating reach, the frequency distribution is calculated. This is a calculation of the discreet number of persons reached by each ad in the schedule. Thus one can compile the number (or %) of target persons reached "at least" the set number of times.


Tuesday, May 02, 2000 #3436
I need to calculate target ratings for a particular cable network. I've got HH VPVH and the target VPVH. Now what?

The Media Guru Answers(Tuesday, May 02, 2000 ):
HH VPVH would mean "Household Viewers Per Viewing Household," which makes no sense.

You need viewing households and Target Viewers per Viewing Household. Multiply these two and divide by target universe to get target rating.


Monday, May 01, 2000 #3434
I am trying to determine how best to manually calculate reach and frequency for Out of Home Media. Would you be able to help and provide me with reach curves and turnover ratios for OOH media. Thank you.

The Media Guru Answers(Tuesday, May 02, 2000 ):
Out-of-home (outdoor poster media) is usually bought in #25, #50 or #100 "showings." These are based on daily effective circulation, or traffic, equal to 25, 50 or 100 GRP per day, respectively.

Within the state of the art, in rough terms, these levels usually mean 4-week reach and frequencies of approximately

  • 80 / 8.8 / 700
  • 87 / 16.1 / 1400 and
  • 92 / 30.4 / 2800.

As should be apparent, there is not much room for fine tuning, nor much reason for considering other GRP levels.


Monday, April 17, 2000 #3401
Dear Guru, I need to develop a cost estimate and approx. reach/freq. for a US television buy in the top 40 markets. Here's what I have and what I still need to know: I have the markets and approx. CPP per daypart from SQAD. I need to know how to calculate a rough estimate of reach & freq for 1 week to 1 year based on 200 points per week in each market. Can a network (CBS etc)place the entire buy, or do I have to do this per market. I'm one person and can't spend too much time executing this (if it happens). Any advice would be great. Thanks

The Media Guru Answers(Monday, April 17, 2000 ):
No, networks don't place spot buys. You can use spot reps or media buying services. Find these in Standard Rate and Data Service (SRDS) or The Standard Directories of Advertising Agencies and Advertisers ('The Redbook')

Either one can help you with reach and frequency, or eTelmar.com offers an inexpensive, online reach calculator.

If you are buying 200 points per week for a year in the top 40 markets, you are spending in the 10's of millions, at least. This is ample to hire a buying service or at least some experienced free-lance help. Either one would save you far, far more in media costs than the expense of their fees.


Wednesday, April 12, 2000 #3392
Guru, I've never used a planning program as most of my planning has been national print and outdoor, local broadcast, and things I've felt I can handle on my own.I've seen so many planning programs and websites for planning it's hard to tell the good from the bad. Have you ever evaluated planning programs and, if you have, can you recommened one or two? Thanks

The Media Guru Answers(Wednesday, April 12, 2000 ):
To the Guru, the term "planning program" means programs like Telmar's AdPlus or Telmar's full set of individual media analysis programs or the eTelmar online suite of media programs.

Such programs calculate reach, frequency, effective reach, frequency distribution, and quintiles for individual media plus combinations of media as well as cross-tabulations and rankers from media audience databases. Flow charting is also a typical option.

These programs don't actually create media plans, that is determine how much budget to invest in each medium, ad units to use, and scheduling. There are such programs on the drawing board, but require that the planner quantify and factor those concepts which would be subjective judgements.


Friday, April 07, 2000 #3375
How to calculate Affinity Index when both percents are equal, that is: target = 0,01 and universe = 0,01? I learned to calculate target divided to universe x 100, but in some cases, the result is very high and something seems to be wrong on my calculation.

The Media Guru Answers(Sunday, April 09, 2000 ):
The Guru is unclear about your question. "Universe" is, by definition, 100% of something.


Thursday, April 06, 2000 #3370
dear guru, iam developing a marketing plan for a portal in india and of indian origin.in that context i need to know what are the number of advertising exposures required on the net(say on sites hotmail and yahoo) to induce a person to visit the portal.by knowing that iam trying to calculate the effective exposure index.thanks in advance.awaiting ur reply.

The Media Guru Answers(Thursday, April 06, 2000 ):
Put this way, you make it a very complex question. Some individuals will never visit, some will respond to the third exposure they see. And of course, it depends on the creative.

As a general rule, though, current average banner click rate is about 0.5%. On that basis, 200 impressions produce one visit.


Thursday, March 16, 2000 #3326
Dear Guru: I would like to know if there is any equation to calculate media mix reach?

The Media Guru Answers(Thursday, March 16, 2000 ):
There are several, equivalent ways to express the arithmetic to combines media according to random probability, which has been found generally adequate for the purpose of multimedia combination.

Here's an easy one:

  1. Work with two reaches at a time
  2. Treat the reach of each medium as a decimal (50 reach is 0.5)
  3. Add reach of medium A and medium B
  4. Multiply reach of medium A by Reach of medium B
  5. Subtract the product of the multiplication from the sum of the addition

Example:

  • Reach of medium A = 40, reach of medium B = 55
  • 0.4 + 0.55 = 0.95
  • 0.40 x 0.55 = 0.22
  • 0.95 - 0.22 = 0.73
  • Combined reach is 73

To add additional media, treat the combination as medium A and the next medium as B.

In some cases, a planner may have access to research which shows that an adjustment should be made for actual, measured, duplication between different media, rather than use the "random probability" formula above. In that case, more sophisticated reach calculating software packages, such as those from Telmar allow you to make the calculation and build in known adjustments.


Tuesday, March 07, 2000 #3291
Is there a formula which calculates effective reach and frequency? I know that reach x frequency=grp's, but how can I determine what the effective reach and frequency would be for 100 grp's or 150 grp's?

The Media Guru Answers(Friday, March 10, 2000 ):
Of course there's a formula, but it can be immensely complicated. In fact, media planners rarely, if ever, considered effective frequency before computers became a part of everyday reach and frequency calculation in the 70's.

Your "reach x frequency=grp's" is not a formula, but merely the arithmetical relationship of these quantities as they are defined.

GRPs are the convenient weights and mesures we use in media buying. They are simple statistical measurements, whereas reach and frequency are more complex statistical models In some cases, there are relatively simple reach formulae derived from compiling the actual, measured reaches of actual schedules with known GRPs. The formula is non-linear.

To find the effective reach of a schedule, you first determine level of frequency to consider "effective" and then examine the frequency distribution of the schedule to see how many people have been reached that number of times The frequency distribution shows exactly how many people have been exposed to each integral number of announcements in a schedule.

The math is based on non-linear functions. For any given reach and GRP set, the frequency distribution can vary considerably depending on the media combined and the dayparts within the media.


Wednesday, February 23, 2000 #3239
Dear Guru: I need to analyze a cpm tv negotiation on w1849. I would like to transfer the negotiated w1849 cpm to w2549. Can you please explain the formula to transfer Thanks.

The Media Guru Answers(Thursday, February 24, 2000 ):
Step 1: calculate the vph ratio (or impressions ratio) of w25-49 versus w18-49.
Step 2: divide the w18-49 cpm by this ratio.

Example:
Suppose w18-49 CPM = $10.00
Suppose the w18-49 vph is .700 and w25-49 is .600
then the ratio is
.600 ÷ .700 or
.857
Then $10.00 ÷ 0.857 = $11.66

It's best if you have the actual vph or impressions of your buy, but you can make an estimate from the daypart averages.


Thursday, January 20, 2000 #3136
Is there a simplified reach and frequency calculation formula that allows for the number of stations (TV or radio) as well as the target audience size?

The Media Guru Answers(Saturday, January 22, 2000 ):
Reach and frequency calculations have become quite complex today and are typically done by computer. Because reach is curvilinear, the formula can be quite complex, even without this issue. A different algorithm is needed for each dispersion scenario.

A good system will account for number of stations, at least in radio; AMIC's sister company, Telmar has such a system.

Since reach calculations are typically done with percentages of universe, like rating and percent reach, target audience size is not specifically relevant. Different curves will have been deduced for different targets, based on their accumulation patterns, which may not exhibit a direct correlataion to size. If reach in thousands is needed. it is simple to calculate by multiplying perent reach against target population.


Wednesday, January 05, 2000 #3096
Oh Great Guru -- I need to calculate GRPs, but I don't have reach or frequency on some tv buys. I do have CPM, total impressions and impressions/week and the total population of the demographic. Can you supply a formula for calculating GRP based on what I have?

The Media Guru Answers(Wednesday, January 05, 2000 ):
(Impressions divided by population) x 100 = GRP.

For example,
if impressions are 2 million and population is 1 million, GRPs = 200.


Monday, December 20, 1999 #3061
Dear Media Guru, we are currently in the process of conducting a dip stick media survey to be integrated in a media reviews presentation for an FMCG. Based on your experience and references what would be the minimum acceptable sample size to use in order to insure the research findings are viable and reliable ? Note that the total population in our market is close to eighteen million, and the specified target group is about ten million. Thanks.

The Media Guru Answers(Monday, December 20, 1999 ):
The population is not really relevant to determining the reliability of a sample.

The key consideration really is: what size are the answers you expect and what potential error can you stand in your decision making?

The formula to calculate one standard error (68% confidence) is:
The square root of ((P times Q)divided by N),
where

  • P is the percentage of the sample offering the response to test (treated as a decimal fraction)
  • Q is the remaining percentage of the sample, and
  • N is the sample size
For 90% confidence, the above formula is multiplied by 1.645.

When you hear that results of a political poll are +/- 3%, this is the range of error around a 50% answer, usually at "90% confidence," meaning that if the poll was repeated with the same sample size, 90 times out of 100 the same question would have a response between 47 and 53%.

"+/-3%" on an answer of 50% means 6% relative error.

A sample size of 750 would bring you +/- 3% on a 50% answer:

100% - 50% = 50%

P=0.50, Q=0.50 and N= 750

0.50 x 0.50 = 0.25

0.25 / 750 = 0.000333333

Square root of 0.000333333 = 1.8275

1.8275 x 1.645 = 3.003

That's fine for political polls, where responses tend to hover around 50%. But in media studies, 1% or 2% may be exposed to a given ad placement. The same 750 sample would give reliability of +/- 0.84 on a 2% response, which is a relative error of 42%.

So, use this formula with your anticipated answer sizes and the level of relative error with which you can comfortably make decisions to determine a suitable sample size.


Tuesday, December 14, 1999 #3051
How do I calculate a radio station's turnover rate?

The Media Guru Answers(Tuesday, December 14, 1999 ):
Divide weekly cume by AQH (Average Quarter Hour) audience.


Tuesday, November 02, 1999 #2927
I am basing a media plan on the recency theory and wanted to know how to calculate cost per reach and/or cost per reach point for my broadcast buys?

The Media Guru Answers(Tuesday, November 02, 1999 ):
The Guru has discussed this previously.

Click here to see past Guru responses


Friday, October 29, 1999 #2919
Hi Guru! I'm trying to calculate magazine CPMs. Should I use the ABC Audited numbers, or the total audience numbers (some magazines have high "hand-off" rates)?

The Media Guru Answers(Friday, October 29, 1999 ):
Consumer magazine CPMs are ususally compared based on total audience. High pass-along readership is good up to a point. You're most likley trying to reach people not necessarily owners of copies of magazines.


Thursday, October 28, 1999 #2916
Hi Guru, How is the CPM rate calculated by the web publishers?. what are the criteria adapted by them to arrive on their CPM rate?(say $30 or $40 for every 1000 impression).

The Media Guru Answers(Thursday, October 28, 1999 ):
The CPM = Ad Cost ÷ Ad impressions. Many sites quote ad rates in terms of CPM. That is you can order the number of impression you want your banner to receive, and get exactly that number priced at "x" CPM.

Several issues are taken into consideration in setting CPM prices:

  • Competitive pricing - a site can't successfully charge double the CPM of another with similar audience and content.
  • Traffic - up to a point, more size is considered to have a premium value. Then there will be econimies o scale
  • Unique audience- hard to reach demographics are more valuable


Wednesday, October 20, 1999 #2892
Dear Guru, I am attempting to do a publicty report for the coverage (unpaid) our organization receives on a quarterly basis. I have been calculating the advertising value of publicity, based on what each medium would typically charge for that size ad. However, I am looking for a formula to calculate the publicity value of such coverage. I understand that the National PRSA has come up with a formula to calculate this (something like X 3 for television and X 6 for print), but I haven't been able to substantiate this. Can you help?

The Media Guru Answers(Wednesday, October 20, 1999 ):
The Guru doesn't deal with publicity, but if PRSA has a standard, the advantage is that it is a standard.


Tuesday, October 19, 1999 #2880
Guru, I’m trying to put together a print ”insert” plan for a magazine and I have a little problem with calculating the duplication for the plan. The magzine circulation is 100.000 ex, target audience reach 40% (1 insert), target selectivity/profile 80%. By adding a second ”round” of inserts in the same magazine the total reach ad up to 48% (same selectivity). How do I calculate the no ”extra” inserts distributed by the second ”round” of inserts? BR CD

The Media Guru Answers(Tuesday, October 19, 1999 ):
If you reach 40% of your target with one insertion and a net of 48% of your target with two, then the duplication is 32% of the target:

Two insertions has a gross exposure of 80% (40 + 40) and if the net is 48, the duplicated is 80 - 48.


Thursday, October 14, 1999 #2873
Where can I get information on pricing of advertisements in a web site? What are different methodologies for pricing and types of advertisements in a web site? HOw can Return on Investment ROI be calculated?

The Media Guru Answers(Sunday, October 17, 1999 ):
Many web sites have on line rate cards. Go to the Guru Archives Search Engine. Use " rates" as your search term, to see what we have on the topic. Industry information sites, like The Industry Standard, CyberAtlas and NUA Internet Surveys may also be informative.

ROI is simply a comparison of money spent to money coming in. What you plug in to each side of the equation is up to your judgement and your business model.


Friday, September 24, 1999 #2823
As a client not well versed in media and media measurement tools I was asked some questions that I don't quite know the meaning of or the terminoligy in how it is used. These terms are specific and deal with planning strategies. Please help where you can. -A & U or segmentation studies (audience probably) A & U ? -Volumetric analysis? -What does RDI, RPI stand for/mean (indexes for investment strategies?) -Econometric modeling? Thank you, Guru.

The Media Guru Answers(Saturday, September 25, 1999 ):
These are all consumer behavior measurements, not media measurements, but they are used in forming media strategies.

  • A&U is Attitude and Usage study (or sometimes Awareness and Usage study). This is a survey of consumers concerning their knowledge of the product and/or the advertising, feelings about the product and category and ways and amounts of usage. It can be used to define and segment the target.
  • Volumetric analysis goes beyond defining who is using your product and segments users according to the quantity (volume) consumed. The classic example is: Men 18-34 are 20% of beer drinkers, but consume 80% of all beer.
  • RDI and RPI are not entirely familiar to the Guru. The _DI and _PI forms in these contexts are usually (something Development Index and (something) Potential Index. The "somethings" are most typically "Brand" and "Category." Brand Potential Index can be equivalent to Category DevelopmentIndex.

    BDI is a comparison of the sales in a specific market versus the markets portion of national sales. So, a market where 3% of all product sales occur but only 2% of the target population lives has a BDI of 150 (3 ÷ 2). CDI is calculated the same way, but using the entire product category's sales.

  • Econometric modeling is a broad, general term, taking into account all the above and other measures of consumer behavior.


Wednesday, September 22, 1999 #2815
Can you please refresh my memory and tell me how to calculate multi-week reach and frequency across television and radio? Thank you

The Media Guru Answers(Thursday, September 23, 1999 ):
If you mean combining these media, the formula has been addressed. Click here to see past Guru responses.

If you mean how to get multiweek reaches for either medium, you need reach curves or software, the extension formulae are tow complex for casual use.


Wednesday, September 22, 1999 #2814
Hi Guru The ad agency I work for has a theory that cable GRP's and radio GRP's effectivenesss are significantly less than network and spot television. On our flow charts we only calculate 1/2 half of these points. I have heard this theory before but I've never seen a plan that cuts the GRP's in half. What do you think?

The Media Guru Answers(Thursday, September 23, 1999 ):
The Guru has been aware of theories that use effectiveness factors in comparing media. Sometimes GRP are adjusted on the flow chart, but since the flow chart often serves as the buying control document, more often the adjustments are shown in reach and frequency comparisons.

There can certainly be an argument that radio has less effectiveness than TV, commercial exposure versus commercial exposure, all else being equal. But, the argument doesn't seem to be rationale for cable TV. The commercial is the same, the presentation is the same. Unless there are objective measures of attentiveness or clutter or recall used, why is cable less effective? Individual commerical audience size is not relevent to message effectiveness of the medium; one consumer is not aware of how may others are watching the same program.


Tuesday, September 14, 1999 #2795
Dear Guru, I am writing to you from the Middle East. First of all I am very excited to discover the AMIC site. I have recently been exposed to various documentation on the recency theory. Alongwith the documentation I have seen something called reach curves. The reach curves I have seen are typically for 1+, 2+, and 3+ levels for all adults and all women audiences. I understand it is an easy way to translate Effective Reach goals into GRP goals e.g. X GRPs will get you Y% 3+ reach against the target. It also clearly depicts the point of diminishing return. I am eager to know how I can develop reach curves for my market. Can this be done by us in the media department or do we need to approach some company which specializes in this area. What sort of data is required? Just to give you a background, we are not a metered market. TV audience measurement is conducted thrice a year using face-to-face interviews with a representative sample. Viewership is typically available by 15 minute time segments for all channels across various demos. Thanks in advance.

The Media Guru Answers(Tuesday, September 14, 1999 ):
Reach curves have been in use since long before computers were used in media departments and long before metered measurement.

Curves are created by using the reach of actual schedules. For example, in the U.S., Nielsen would report the actual reach of specific brands' schedules, based on examining the net unduplicated viewers in their reasearch data who viewed the program schedules used by the brand's commercials.

Once you have several schedules ( 8 or so will do) with actual reaches and frequencies for various GRP levels, you can use the regression analysis data function in a spreadsheet, like MS Excel or Lotus 1-2-3, to calculate a formula which describes the curve. This formula can literally draw the curve on a graph, or let you build a table of GRP / Reach pairs. By the way, it is the frequency and GRPs which are used in building this regression, because while reach is a curve, frequency is a straight line.


Tuesday, September 14, 1999 #2792
What can you tell me about reach-based planning? Thank you in advance.

The Media Guru Answers(Tuesday, September 14, 1999 ):
> The usual assumption is that print and broadcast duplicate with random probability, there is no special, greater or lesser likelihood that persons in the audience of the print schedule will also be or not be in the audience of the broadcast schedule.

Mechanically. the combination may be calculated in a few equivalent ways. The Guru finds it easiest to consider the reaches as decimals (50% reach = 0.50).

Subtract the reach of print from 1 and multiply this by 1minus the reach of broadcast. Suppose print has a 40% reach and broadcast has 55%.

By subtracting 0.4 from 1 (1 - 0.4 = 0.6), you have the probabilty of the target not being exposed to print. Subtract 0.55 from 1 to get the probability of not being exposed to broadcast (1 - 0.55 = 0.45)

Multiply these two together (0.6 * 0.45 = 0.27) and you have determined there is a 27% probability of people not being exposed to either of the combined media, or a 73% reach.

This formula is typically used in media software to combine different media.

Certainly there are cases where there is a somewhat better than random probabilty of media duplication, such as TV Guide combining with a TV schedule, but that's the exception, calling for judgement.


Monday, August 30, 1999 #2752
Respectable Guru, How shell one evaluate an editorial where there is sponsor mentioned several times? Is such article's value same as the one of an adequate sized ad? What if there is additional 4C photo with logo visible? Is there any methodology on how to evaluate and calculate the value of an PR editiorial? Are you perhaps aware of any books or other sources on this topic - sponsorship evaluation? Thank you for your answers.

The Media Guru Answers(Wednesday, September 01, 1999 ):
This is a public relations question more than a media question, however . . .

Typically, PR mentions are evaluated simply based on the value of the equivalent space purchsed as media. But an editorial mention really has more impact, seeming to be a disinterested, third party endorsement.


Monday, August 30, 1999 #2750
Let me elaborate further on the question posed by Ajay (Question sent from India , which was answered on 8d August). In India, the data collection and hence reporting of the peoplemeter data is on a weekly basis, unlike the daily collection and reporting in most other markets. Since we follow a weekly collection, the sample is determined for each of the seven days. (after rejecting viewing which does not satisfy the threshold levels of various criteria that the viewing data is supposed to fulfill). As is obvious, this effective sample could be different across the days. Hence, we actually could end up having 7 different samples for each of the seven days. The question now arises as to which of these seven figures to use for projection to the universe. This is the part where the difference in the reach and rating calculations occur. A rating figure is calculated based on the sample for each day. Hence , on Monday, if the effective sample is 95, then this 95 is projected to the universe figures. On Tuesday, the effective sample could be 96 - then this 96 is projected to the universe figures. And so on. Hence the actual weights attached to the sample could vary, though the universe figures remain the same. Once the sample figures have been projected, the ratings are calculated. These rating figures can then be averaged across days , if desired, since a rating figure can be averaged across time periods. On the other hand, a reach figure cannot be averaged. Hence, if the sample is different across each of the days, the dilemma is as to which of the effective sample to use for the projection purpose. Hence they designate one day as REFERENCE DAY. The effective sample on the reference day is the one which is used for projection purposes and hence for all further calculations for reach figures. The reference day changes depending on the period chosen. In India, the research agency has fixed the reference day to be the last day of the period chosen. So, if I vary my period of analysis, the reference day changes and hence my reach figures change. This is where the confusion occurs ! Since a rating calculation does not have a reference day, the ratings don't change, irrespective of the period chosen. So please let us know if this is the norm followed across countries ? Is the concept of reference day valid ? How do other countries deal with this ?

The Media Guru Answers(Thursday, September 02, 1999 ):
The Guru is not aware of this method in use elsewhere. It does not seem that it would have significant effect unless there are substantial daily variations .


Saturday, August 28, 1999 #2748
Hello, Guru! What is media inflation and how is it calculated ? Thanks.

The Media Guru Answers(Saturday, August 28, 1999 ):
Like any other economic inflation, this simply refers to the price increase from one time to another.


Monday, August 23, 1999 #2734
Dear Guru, in regards to broadcast, my company advertises on national cable networks only. Our media buying company submitted a post-buy analysis for 2Q, but did not include reach/frequency info. When I asked for this information, they said "it's not standard to give cable r/f" is this true and if so, why? Thank you.

The Media Guru Answers(Monday, August 23, 1999 ):
The Guru agrees that it is not "standard" to include delivered R&F in a post analysis. It is probably not relevant, if the buy was built around a planned R&F and the post shows that the buy delivered as estimated.

However, what is standard, is for a service to respond to a client's question. If the buy delivered out of line with the estimate, the service should, at minimum recalculate the R&F. If the issue is running an actual R&F of the schedule, based on spot by spot use of the Nielsen cume system, significant expense might be involved, and this could be open to negotiation.


Thursday, August 19, 1999 #2729
Dear Guru, 1- Please let me know SQARE model that SQAD use to calculate CPP for TV and Radio. Please let me know the detail or any link I can find more information or books... 2- Do you know any model for reach vs GRPs? Our client ask us to show the data like that. The problem that we try to find the suitable daypart mix, station mix, medium mix that is good for our advertising strategy but we don't have any optimiser programs. We have only ratings data like Telescope and Prinscope of ACNielsen. Do you know any example to solve this kind of problem? 3- Our client also want to have a model to set advertising budget to get for example 80+ reach but we can not know until it happen. How to solve this issue? warmest regards, Thai Vang

The Media Guru Answers(Thursday, August 19, 1999 ):
A general explanation of SQAD's model is available from SQAD. They will give you the same information they would give the Guru. But the essence is manipulating actual buying data in real situations, provided confidentially by actual media buyers.

GRP's and reach do not have any standard realtionship, except within given media and population parameters. You are writing from Viet Nam, where Televison audience cume patterns are likely to be quite different than in the U.S. Even within the U.S., Hispanic TV reach curves are very, very differerent than the General Market TV reach curves.

The way to build a model, to oversimplify, is to collect a great number of actual reaches of real schedules, and then plot their frequency against reach in a regresssion analysis, which gives you the formula for the "curve." Frequency is plotted, rather than reach, because frequency is a straight line while reach is a curve. The curve formula then allows you to create a model with a reach solution for any GRP input. The more variables you use to build different curves, the more sophisticated your model can be.


Thursday, August 05, 1999 #2690
Dear Guru, I am a media planner in India. We have a research agency which provides us data on television viewership. The data is collected by a peoplemeter which has a picture matching technology. The problem I am facing is that the TRPs Or TVRs as they call them are calculated on the basis of the sample on that particular day, whereas Reach for a programme/ spot is calculated based on the sample on the sunday of the last week of your analysis. To give an example, if I have a spot on the 1st of June and I select my period of analysis as 31/5/99 (Monday) to 13/6/99 (Sunday)a period of 2 weeks. The TRP for my spot would be calculated based on the sample of the 1st of June, but reach would be calculated on the basis of the sample on the 13th of June. This gives me two major problems. The 1st being that my TRP and Reach figures have little relation. The 2nd being that the reach figure given for the given spot on the 1st of June would vary depending on the last week of my analysis. This is a problem that manifests itself when I try to plot reach curves. If I state that my brand has achieved 50% reach by June, I could be in trouble the next month where the reach figure might actually drop purely because of a change in sample size. I would like to ask you if you face the same problem in your country. Or is there a better system to report data. My research agency says that this is the best method, I refuse to agree. Please do enlighten me. Regards Ajay

The Media Guru Answers(Sunday, August 08, 1999 ):
The system you describe does not make any sense to the Guru. Ratings and reach should be based on the projection to the population represented by the sample, so changes in daily sample size would not be a factor in the base. Usually, samples across days can be added to increas the sample for a period of time.


Monday, August 02, 1999 #2678
Hi. Are there any clear rules for calculating the level of spending increase needed to achieve an X% increase in brand awareness? Thanks

The Media Guru Answers(Thursday, August 05, 1999 ):
Certainly, no "clear" rules. For one thing, it changes depending on the current level. I.e. a 100% increase in awarenes, from 10% to 20% requires a different spending change than a 50% increase from 60% to 90%.

The choice of added media and base media are also key factors.

The Guru speculates that the closest you can get, to calculating a percent increase in awareness through advertising spending is to calculate the cost of making the same increase in reach, sustained over a time frame comaprable to the base period.


Wednesday, July 28, 1999 #2666
How are weighted averages calculated?

The Media Guru Answers(Saturday, July 31, 1999 ):
Each value in a series is multiplied by its appropriate weight and the sum of the products is divided by the sum of the weights. Example:
  • The ratings of a TV program are 5.0 for teens, 4.0 for Adults 18-24 and 3.0 for Adults 25-49, so what is the 12-49 rating?
  • It's not the straight average of 5.0, 4.0 and 3.0, or 4.0; we must account for the different sizes of the populations in these three age groups.
  • Let's suppose the teen population is 22.5 million, A18-24 is 24.7 million and A25-49 is 103.3 million. We multiply each rating by its population base:
    5.0 x 22.5 = 112.5
    4.0 x 24.7 = 98.8 and
    3.0 x 103.3 = 309.9
  • Sum the products: 112.5 + 98.8 + 309.9 = 521.2
  • Sum the weights:
    22.5 + 24.7 + 103.3 = 150.5
  • Divide the sum of products by the sum of weights
    521.2 ÷ 103.3 = 3.5 correct (weighted) average rating.


Monday, July 19, 1999 #2643
Dear Guru! I've got the following question. Our client has a product to advertise. He has set advertising goals for the ad campaign. We defined the level of effective frequency needed to reach these goals. 1. What is the range of effective reach? For example, 30%

The Media Guru Answers(Friday, July 23, 1999 ):
Media plan communications goals should specify a level of effective reach along with specifying the effective level of frequency.

Basic, as well as more advanced media software, calculates reach and frequency, frequency distribution and reach at various (effective) frequency levels. Input is typically GRPs.

Setting an effective reach goal can be based on gut, such as reaching the majority of the target at effective frequency levels in 4 weeks, or based on sales predictions. For example, this might be an estimate that 10% of those reached efectively will buy and X number of sales are the goal. Then 10 times X are the number who must be effectively reached.


Wednesday, July 14, 1999 #2632
What are GRP's and what do they stand for in a media buy? I am an Account Manager and don't have the Media background but need to explain the GRP levels to my Product Managers. Please help.

The Media Guru Answers(Thursday, July 15, 1999 ):
GRPs are gross rating points, the pounds and ounces of media buying and selling. The target audience of an advertisement divided by the population of the target group is the ad's rating. The sum of the ratings of the ads is the Gross Rating Points. Plans specify how many GRPs of each medium to buy. For print, specifications are more often numbers of insertions in specific titles, but the GRPs can be calculated the same way and one plan compared to another.

Allowance must be made for :15 versus :30 GRP or half page versus full page. A given program or magazine has the same rating (GRP) whatever the ad size/length, but obviously there is more benefit from 100 GRP of :30s or pages than from 100 GRPs of :15s or half pages.


Friday, July 09, 1999 #2620
Hello GURU ! I have 2 questions for you : 1. One of the media analysis we do in our agency,mainly for TV, consists in comparing a competitor's share of spending (calculated as his % of advertising expenditure within the total category) with his share of voice (calculated as the % of his 30 sec equivalent GRPs within the category). Is this correct in your opinion ? 2. How do you define SOV ? Is this the % of the GRPs one achieve within a category or is it the % of money invested by an advertiser within a category in a certain period ? Thanks.

The Media Guru Answers(Sunday, July 11, 1999 ):
1) What do you do with the results of this comparison? How does the ratio of SO$ to SOV help you make decisions? The :30 equivalent step is reasonable, but how do you do that effectively outside of broadcast?

2) Some use SOV to refer to share of spending, others use it to refer to share of weight. The Guru believes share of weight is more descriptive of the marketplace perceived by the consumer, but the person controlling the budget, that is, the client, more often cares about money. They can see the impact of money on the bottom line more easily than they can understand the differences in impact of their :30s versus a competitor's :15s or competitor's radio versus their own magazines.


Thursday, July 08, 1999 #2616
Hello Guru! I would like to ask you what is a 30 sec equivalent GRP and why is that calculated if the spot length (as per some of your previous answers) do not influence yhe level of reach&frequency ? Thanks

The Media Guru Answers(Thursday, July 08, 1999 ):
:30 equivalent is a buyer's convenience. Assuming the standard unit purchased is a :30, instead of dealing with different unit rates in the same program, a :15 is treated as if it had half the rating. It's strictly an efficiency/value issue and has no impact on reach or frequency. Remember that the ratings we have are actually the ratings of programs or time periods and not commercials; commercials are just assigned the rating of the time slot wherin they air, so commercial length is irrelevant to rating.


Friday, June 11, 1999 #2570
how do you figure out net impressions for newspaper. also, how do you show that you are delivering proportionate impressions to the populations of the different markets? would you show the population as a % of impressions? thanks

The Media Guru Answers(Tuesday, June 15, 1999 ):
If you mean "net" literally, this is the reach of your plan expressed in numbers of consumers instead of percent. There are various newspaper reach methodologies. If you mean total impressions, newspapers you are buying should have detailed audience data, from resources such as Scarborough.

If by "proportionate" you mean to deliver impression in the same distibution of age and gender as the population, one wouldn't expect to deliver impressions proportionate to the market: different population segments have greatly different newspaper reading habits.

If you mean the total impressions distribution across a market list should parallel the population distribution between markets, then simply calculate the market-by-market population percent distribution and buy newspaper schedules to that proportion.

For information about newspaper planning and research tools, visit the newspaper media software page of AMIC's sister company, Telmar or for readers-per-copy averages and other research sources, see the The Newspaper Advertising Association site.


Tuesday, June 08, 1999 #2562
I have an outdoor question. If showing size refers to the reach per day, i.e. 25# reaches 25% of a market per day, why aren't the estimated TRPs per month simply 25 x 28 = 700. Most studies I see quote a lower TRP level for a 25 showing. What gives?

The Media Guru Answers(Wednesday, June 09, 1999 ):
The Guru has come across this problem and found the answers.

There are two answers, one sensible, one nonsense, but both real.

Sensible: The "25 showing" is a standard number of panels, based on 25% of adult population. So if your target is Women 18-34, there may be a different number of women 18-34 GRPs in a showing actually bought as 25 Adult 18% GRPs. This is perfectly sensible, and happens ain all media, but the sellers and buyers of other media are fully conversant with these facts.

Now for the nonsense answer, which is most likely the basis of the number you were given. Various research companies, such as MRI have measured outdoor as part of multimedia reports and these generalized reports are being used to estimate target reach for a marketpace showing. Often a completely different source for average frequency is used and these two factors are multiplied to calculate GRPs. It seems invariably to be much lower than the GRPs you would get by the realistic method first described, and so makes outdoor seem less efficient than it should.

The misused sources could, instead be used to provide relative exposure indices between demographics, allowing a simple conversion of GRPs. The Guru hopes the Outdoor industry improves in this area.


Thursday, May 20, 1999 #2516
I'm trying to learn the basics of media research. How does one calculate ROI on media effectiveness and where can I get information on how/ why to conduct market research? Thanks.

The Media Guru Answers(Sunday, May 23, 1999 ):
The Guru has discussed ROI several times. Click here to see past Guru responses about ROI

For information on marketing research, try the Advertising Research Foundation InfoCenter For details about the InfoCenter, call 212-751-5656, extension 230.


Monday, May 10, 1999 #2499
How do you calculate reach "in-market", and are you to combine that with the national numbers? How is this done? Thanks. We are trying to show total "in-market" delivery. Also, back to the average 4 week dilemma, is it only relevant when looking at sustaining levels of a continuity plan? Or would you show average four week even in a launch, retail, or promotional type heavy-up situation? Thanks as always.

The Media Guru Answers(Monday, May 10, 1999 ):
Suppose you had national media with a reach of 40% and a local media plan delivering 50%.

You would combine the national reach of 40% with the local 50%. If you care to go the extra step, you could analyze local variation in delivery of the national plan and adjust the local delivery of the national media before combining with the local. Or if you run only national media you can look at the locally delivered weight to caculate the in-market reach resulting from national media, as if it were local spot media.

Four weeks is a traditional standard measurement period. This standard goes back to the days of the dominance of monthly magazines as an advertising medium. There are numerous ways this rule of thumb is used. Some look at "4-weeks-when-in" and examine four weeks worth of average activity no matter ho many active weeks a plan has. This focuses on the rate of advertising rather than the quantity. Other focus on cume of whatever number of weeks. One has to make a judgement of what tells the story best. The judgement can be made differently when you are comparing possible plans and when you are trying to quantify potential effects on awareness, sales, etc.


Friday, April 30, 1999 #2481
Is there any way to calculate duplication across a media plan using several media (e.g. print and radio and TV), or can I only get a duplication analysis within a media (radio duplicaton and then another duplication factor for print, etc , etc) I use telmar for research with simmons and arbitron access and we also use JDS for buys.

The Media Guru Answers(Friday, April 30, 1999 ):
The standard assumption in media planning is that duplication between different media is purely at random. Therefore, the random probability formula is used:
  • Express the reach of each medium as a decimal (50% reach = 0.5)
  • Multiply the reach of one medium by another to determine the duplication.
  • Subtract the duplication from the sum of the two reaches to get the net reach

So, if you have a 40% reach in TV and a 55% reach in Print, multiply
0.4 x 0.55 to get 0.22
subtract 0.22 from 0.4+.55 and get 0.73 or
73% reach of the combined media.

There are a variety of ways to do the calculation. The Guru actually prefers to use the probablilty of not seeing each medium (reach as a decimal subtracted from 1.0) When these are multiplied they give the net probability of not seeing any of the media. When this result is subtracted from 1, the final result is net reach. This style is particulary useful for combining several media at once.The example would combine this way:

  • 1-0.4 = 0.6
  • 1-0.55 = 0.45
  • 0.6 x 0.45 = 0.27
  • 1-0.27 = 0.73 or

    73% reach.

Telmar's "Media Mix" program uses these assumptions.


Thursday, April 15, 1999 #2450
thanks for your answer about the reach per point from the 14/4/99 but how can i calculate it it it write to devide reach in trp for example if the reach is 50 and the trp is 100 so the reach per point will be 0.5

The Media Guru Answers(Thursday, April 15, 1999 ):
Yes, divide reach by TRP to get reach per point.


Tuesday, March 30, 1999 #2421
The need: I am looking for a way to factor 'page exposure' data into mainstream media metrics such as CPM or GRP. MRI tracks and calculates page exposure using the following: (#_of_days_magazine_is_read * #_issues_read * %_of_pages_read)= avg_page_exposure I believe such data would not only provide a more accurate picture of a readers exposure to the ad pages but could alter CPM & GRP rates. My Reasoning: CPM=(ad_rate/audience); audience=(circ * readers); 'readers' is thenumber of different set of eyes per issue (single exposure). This number does not take into account how long or to what extent the reader looked at the publication -- it could be the mailman delivering the magazine who remebers the cover or it could denote a subscriber who reads the issue cover to cover. Enter the issue of page exposure. Suppose I am considering a magazine with a CPM of .01851 = 40,400/(704,000 * 3.1). However, if this same magazine provided me with a page exposure rate of .99 = (3 * 1 * .33) -- which says that the audience takes 3 days to read the issue and reads about 33% of the issue a day (which I know is unrealistic, but hear me out). Now suppose I take the .99 exposure rate and add it to the 'reader' and recalculate CPM --- 40,400/(704,000 * (3.1 +.99))= .01404 -- I get a much lower CPM. My Question: Why can't I make this type of calculation with page exposure data -- where is the break down in my logic or math? Any insights would be GREATLY appreciated. Thanks in advance!

The Media Guru Answers(Thursday, April 01, 1999 ):
First, overall, yes it is a reasonable and not unusual concept to adjust CPM according to additional measured factors reported for magazines. However, there are some minor and some major issues with your process in terms of labels and decimal places, etc.

Yes, CPM is ad rate ÷ readers. Readers is circulation times readers per copy (refer to the explanation of MRI information which the Guru did for you in query #2403).

So the basic CPM -- cost per thousand impressions -- in your example is actually $18.51, assuming you mean $40,400 is your ad csot, 704,000 is circulation and 3.1 is readers per copy.

The exposure rate is a factor, not an add on. So the adjustment would be 3.1 times 0.99 or 3.069 or virtually no change. The cpm is now $18.70. If there was a very different page exposure factor it would make a difference. It is a valid way to reaxamine CPMs.


Tuesday, March 23, 1999 #2403
I have been researching these questions for a number of days now and have been unsatisfied with the answers I have been receiving. I am a new member and new to this field, any direction would be most helpful. Thank you in advance... 1) What is the difference between Rate Base (a number guaranteed by publishers and audited by ABC) and Readership (a number provided by, say, MRI) levels for magazine publications? 2) Which number (above) is most often used to calculate CPM (I believe this calculation is ad_page_rate/readership)? 3) Is 'readership' really a composite number (perhaps a result of some other formula)? If so, does Page Exposure Rates factor into 'readership'?

The Media Guru Answers(Tuesday, March 23, 1999 ):
If you went to AMIC's Rates, Dates and Data area and clicked the link
"Audience data from MRI is available for
Fall 1998 for Total Audience, Circulation and Readers Per Copy
" you would see the table from which this image is taken:



The following discussion will use this table as a visual aid.

"Rate base" refers to circulation, the actual number of copies of a publication printed and sold for the average issue over a specified period of time. In the table, "Circulation" is the middle column of data.

"Readership" is the number of readers of the average issue. It includes "passalong" readers, who may not be the buyers / subscribers but read some else's copy. In almost every case, total readership will be greater than circulation. The first three columns of the MRI table we are looking at are readership numbers.

CPM can be calculated based on either circulation or readership. The circulation CPM (Cost Per Thousand) calculation is: divide ad cost by the number of copies in circulation.

The readership cpm calculation is: Divide ad cost by number of readers of an average issue. Often readers within a specified demographic the advertiser is targeting are the divisor in this second calculation. As a planning tool, the readership CPM is more common than the circulation CPM, especially for categories of print that use readership research, such as MRI.

Many people misinterpret the common reporting of "readers per copy." The last three columns of the MRI data are readers per copy figures. What audience research actually measures is readership. A random sample of consumers is interviewed and asked about their magazine reading to determine how many readers there are for an average issue of a magazine. Readers per copy is a calculation done after the fact, dividing the readers measured by the circulation. It is a handy factor used to compare magazine pass-alongs or to calculate other audience elements.


Friday, March 19, 1999 #2400
I need to know the calculation to work out margin of error for TV reach and frequency results. E.g. what is the margin of error of 40% @ 2+ depending on the size of the sample, penetration etc.

The Media Guru Answers(Saturday, March 20, 1999 ):
Assuming you are using a model to calculate reach and frequency, your error is no longer an aspect of sample size but of the reliability of the model.

For instance, suppose your schedule consisted of 20 advertisements with an average rating of 10. And, based on sample size, the 10 rating was +/- 2 rating points (or 20% relative error). But your total schedule of 200 GRP is not going to be +/- 40 points. Because error is plus or minus, there is an equal chance that one 10 rating is really PLUS 2 and the next 10 rating is really MINUS 2. So, in a schedule, most of the error cancels out. This is one reason why ratings minima for buying are often short-sighted.

When it comes to reach analysis, someone might have built a model by compiling several actual schedules measured by the original research and finding a formula for the straight line formed by the average frequency of each. Since the actual schedules came from the orignal research, the sampling error of each (minimized by the plus or minus aspect of the schedule elements, as above) could have been calculated. But now the "curve" coming out of the model is only judged by its ability to match back to actual schedules.


Monday, March 01, 1999 #2363
Dear Guru, We found a formula that is supposed to calculate newspaper impressions. (insertion X circulation*2.28*0.58)/1000) We thought that this was calculated simply by multiplying the circ by the # of insertions. Do you know what this formula could be doing? Thanks

The Media Guru Answers(Monday, March 01, 1999 ):
The Guru believes this formula is meant to estimate persons impressions. "Circulation" represents just copies, not readers.

The 2.28 factor looks like the Sunday average adult readers per copy. The 0.58 then would seem to be the composition for some demographic group, perhaps 18-49. The "/1000" seem to be meant to set the decimal for the particular context in which you found the formula, it will cause the result to be expressed in thousands. Perhaps the number of impressions derived from this formula was going to be used to calculate cpm.

See The Newspaper Advertising Association research databank. for other readers-per-copy and demographic composition data


Tuesday, February 23, 1999 #2353
Is there a standard frequency distribution chart? If so where could i find it or how can it be calculated without a computer program?

The Media Guru Answers(Tuesday, February 23, 1999 ):
Frequency distribution varies depending on the medium (media) used as well as the elements within those media. There are several standard statistical algorithms which can be manually calculated, such as "Beta Bimodal," "Lambda" and others, which can be found in statistics texts. Some are more appropriate to one medium or another.


Wednesday, February 17, 1999 #2346
Could you please provide the basics on how to read a crosstab? Also, the definitions of the terms %col, row, composition, coverage, index - what do all of these mean? This would be very help to folks who are new to media planning and research, so that they could explain crosstab results to others. Thank you!

The Media Guru Answers(Wednesday, February 17, 1999 ):
Crosstabs, those typical computer analyses of data from MRI, Simmons, The Mendelsohn Media Research Affluent Study and other respondent databases, are an essential tool of media planning, used for target selection, media selection, etc.

Here is a section of a typical "crosstab," taken from The Mendelsohn Media Research Affluent Study Reflecting Households with Income of $70,000 plus . It concerns Cosmetics users, persons who visited the Caribbean and Vodka drinkers:

To the left, first the description of each row appears. The top "row," which consists of five lines of data, describes the total population. The next "row" of five lines of data describes readers of Money Magazine, etc.

The next sets of text to the right describe the data content of each of the five lines making up the data rows. "Projection" is the total number of persons the research estimates to be in each category (in thousands, in the total adult universe, which is specified at the top left of the table. This is sometimes labeled "[000]"). Often the term "Audience" appears instead of Projection, especially, though not exclusively, when magazine audience is being analyzed).

The column headings, such as "Total," "Cosmetics," " Drink Vodkas" etc, describe the data in the columns below each heading.

So, at the #1 mark, we learn that 24,855,000 Total affluent adults used Cosmetics in the past year.

At the #2 mark, we see that the number of respondents (persons in the sample) whose educational level is college graduate or better and who use Cosmetics is 3469. In other words, the overall study found 3467 members of its sample who fit both descriptions as to education and cosmetics use. It is important to note this is a whole number and not in thousands. The number 12295 above this indicates that, from this sample, the study projects there are 12,295,000 college (or better) educated cosmetics users.

At the #3 mark we see 12.4 on the %Column line. This means that 12.4% of the column definition (Vodka Drinkers) also fit the row description (Money Magazine readers), that is, 12.4% of Vodka Drinkers read Money. Another way we refer to this is to say that Money's coverage of Vodka Drinkers is 12.4%

At the #4 mark, "%Row" is 16.0, so we learn that 16.0% of the Row definiton (Money readers) drink Vodka. Or, we can say that Money's Vodka Drinker composition is 16.0%

Finally, at the #5 mark, we have an index of 131.3. This is also called "index of selectivity," indicating how much more likely, as compared to the average affluent adult, the persons in the row are to also be in the column. (Traditionally indices are used with no decimal places, so, in application, one would refer to this in future use as a "131 index.")
In this case, the index tells us that a person in a Household which has $100,000 or higher income is 31.3% more likely to have taken a Caribbean trip than the average affluent adult. The index can be calculated either dividing the %Column under Caribbean visit by the %Column under total:
59.6÷45.4
or
in the Caribbean visit column, dividing the %Row in HHI $100,000+ by the %Row in the "Total" row:
21.6÷16.5.


Wednesday, February 17, 1999 #2344
How do you calculate the average radio or TV CPP for a specific market?

The Media Guru Answers(Wednesday, February 17, 1999 ):
It depends on what data you have to work with and how you want to describe your result.

The general rule is to add up all the GRPS delivered and all the costs. Then divide total cost by total GRP.

But this assumes you are working with some real numbers, either a past buy or proposed schedules. Averaging CPPs directly is usually wrong. If you have only CPP's to work with, you will need to get either their associated costs or associated ratings to work back into the numbers you need for accurate averaging.


Monday, February 15, 1999 #2335
Dear guru wanted to find out what is the role of BDI and CDI in market prioritisation. How do you arrive at BDI and CDI and is there a point of saturation on CDI

The Media Guru Answers(Monday, February 15, 1999 ):
BDI is Brand Development Index

CDI is Category Development Index. In either case the index is calculated by dividing the percentage of sales in a local market by the percentage of the population which is in that market. It is done based on a Brand's sales or the whole category's sales respectively. This index then reflects the per capita sales in the market and is used to indicate sales potential.

Some marketing philosophies allocate advertising dollars or advertising impressions delivery according to such an index.

Since the usage ia an index, "saturation" would not be a factor unless sales were bizarrely skewed geographically. For instance, a new product in test market might have 90% of sales in a market accounting ofr just 1% of the population. In such as case it would be ridiculous to use BDi to determine allocation.

The Guru has discussed this frequently. Click here to see past Guru responses on BDI and CDI


Tuesday, February 02, 1999 #2303
I am working on a BPA/ABC circ analysis for 2 trade publications. One is BPA audited and the other ABC audited. On an apples to apples basis, how would I go about calculating a renewal analysis of average qualified subs for the BPA audited pub? I.E. The ABC audited pub calculates the renewal rate accordingly: Total Qualified Circ 1000 One Yr Total Subs Deleted* 200 Annual % not Renewed 200/1000 = 20% Annual % Renewed 800/1000 = 80% *Can this formula be applied to calculate renewal % for a BPA audited pub? I question whether we can integrate the "one year total subs deleted" figure as part of the equation. Thank you.

The Media Guru Answers(Thursday, February 04, 1999 ):
If the same data is reported on the same basis, comparison is fair. If comparable data is not reported, perhaps this is not a crucial analysis. The Guru believes this sort of analysis is likley to be a last-chance tie-breaker at best.


Friday, January 22, 1999 #2285
Dear Guru, This is a bit of a theoritical problem.I am currently working on a shaving cream brand which has been on decline for a few years now. Currently it has a market share of 3.9% and is ranked 7th.The markets where it is doing relatively better are actually the smallest markets, but here too, it is not better than 5th on market shares. It has a media budget which is about 1/5th of the biggest spender, which incidentally is not the market leader. My dilemma is - in the given scenario, for a relaunch, where should media focus be - on the overall smaller markets but where the brand is but marginally stronger or on the bigger markets for the category, where a greater potential lies ? The distribution strength is the same in all markets and no directions have been provided by the marketing team on priority markets. Thank you Guru. My name is Abu Huzaifa and i am media planner in Bombay, India.

The Media Guru Answers(Friday, January 22, 1999 ):
Firstly, these are really marketing issues, not media issues, but to try to look at it from a purely media perspective, consider:

Think beyond the "bigger opportunity of the larger markets," because the impact you can deliver in a market is important. In other words, do you get more consumer response to 100 GRPs against 2 million people or 200 GRPs against 1 million people?

For example:

1. Assume that every impression delivered, no matter the market size, has the same potential to generate sales and / or share growth - where will your budget buy the greatest number of impressions?

2. Assume that the ability of the impressions to generate sales growth is influenced by current share of market. Estimate the value of this effect, plus or minus. Apply this weighting to the impressions you can buy and recalculate sales potential, according to paragraph 1.

3. Or assume that every exposure after the third one (or a level of your choosing) is some degree more effective. How many "effective impressions" can you deliver to each market set?


Thursday, January 07, 1999 #2253
How can I use Spot Quotations and Data (SQAD) quarterly TV Cost Per Rating Point Report (HH Ratings)to calculate CPM? Whether SQAD has info. of CPM for each DMA?

The Media Guru Answers(Friday, January 08, 1999 ):
There is a small difficulty in that the total audience usually used to calculate cpm extends beyond the DMA. But for your purposes this can probably be ignored.

"Cost per Rating Point" means the cost of buying audience impressions equal to one percent of the population. Therefore, dividing DMA CPP by 1 percent of the DMA population yields DMA cpm.


Thursday, January 07, 1999 #2251
Hi there I am a media planner from India and would like to clarify the method of calculating BDI & CDI for a country like ours where population dispersion is not uniform across the SocioEconomic Class (the parameter used for setting the target audience. Iit is a cross tab of education and profession of the chief wage earner of the household) in different markets In such a situation is it advisable to use the total population of the country rather than Target Group Population. Sissors and Bumba advise using the Total Population but i guess thats more applicable to developed countries where TG dispersions are uniform Thanks a lot andrew@LoweIndia.com

The Media Guru Answers(Thursday, January 07, 1999 ):
The concept of BDI and CDI is based on different sales rates (units or Dollars of sales ratio to units of population) within specific marketing regions.

Logically, the same demographic should be used locally and nationally. If each person in demographic "X" consumes 10 units of "y" nationally, than the national rate is Y ÷ X.

In each market, the demographic population is also compared to consumption and a similar ratio calculated. Then the market's ratio ÷ the national ratio becomes the BDI or CDI, depending on whether Brand or Category data, respectively, was used. If the total national population is the base but you use the target pop. in markets, then each market's CDI is inflated by the same percentage. That is, if the target was selected because its members, nationally, have a 150 index of consumption of the product, then each market's BDI would be inflated by 50% if the National population was used as the BDI base.

On the other hand, there may be not difference in effect, because in either case, whatever the national base used, the realtionship between markets will be the same.

However, since it is really sales, not people with which you are dealing, it is cleaner to use total, not target population in each case. Otherwise you assume that in every market, target members consume the same, which obviates the BDI excercise. Suppose someone other than the target is a major consumer in some geographic are, why mask that in planning market allocation? After all the whole idea of BDI/CDI is based on the concept that a product's consumers are not evenly distributed demographically, even in countries where some demographics may be.


Tuesday, December 22, 1998 #2233
Dearest Guru, I am in desperate need of your help. I have a hotel client that would like both a Corporate anylysis & indivdual property anylysis based on what was spent last year. I need to calculate total impressions by business unit (there are 4 units) & also what the cost per impressions were. This is where it gets tricky, can I calculate this information without last years media plans? I am new here and the person before me kept, shall we say, no records of last year's activity. If I can get total dollar's spent by business unit for 98' how can I calculate the above impressions? (ie: last year's circulation per publication / by total cost per media buy) Please help!

The Media Guru Answers(Tuesday, December 22, 1998 ):
If the plan was all print, and you know the spending per title and the magazines' actual rates charged as well as the creative units used, you will be able to do your figuring. But. . .

this means you are using circulation for impressions while audience would tell a fuller story.

Someone in your financial area should have the bills for the schedule, which would tell you the number of insertions and eliminate all that calculation with questionable spending and rates.

Better yet, be a real media pro and contact the salesmen at the magazines to give you the impressions anaslysis you need, they'll probably be delighted to introduce themselves to the new media person this way.


Wednesday, December 02, 1998 #2194
Dear Guru, can you name any media analysis tools and media predictive tools that media planners use on a regular basis without being too technical, of course. Many thanks

The Media Guru Answers(Thursday, December 03, 1998 ):
Here are several:

  • Reach: the number of different target households or persons exposed to a campaign (most often expressed as a percentage of the target universe, and most often calculated over a 4-week period).
  • Frequency: The average number of exposures of the campaign to those reached.
  • Gross Rating Points (GRP) / Target Rating Points(TRP): Essentially interchangeable terms for the sum of the audiences of all the ad units in the campaign, expressed as a percentage of the target universe.
  • Gross Impressions: Same audience count as GRP/TRP but expressed in whole numbers rather than percents.
  • CPP / Cost per GRP and CPM / Cost per thousand impressions: should be self evident from the previous. These are referred to as the "efficiency."
  • Effective reach: Those in the "Reach" who experienced a specified minimum number of exposures (effective frequency)

All the above stem from the audience research tools and investment figures. So called "reach and frequency" systems typically generate all these figures.

Other tools, especially in print media are also occasionally used. These may include "time spent with" media vehicles, "page openings", attentiveness, etc.


Wednesday, December 02, 1998 #2193
Dear Guru! We were asked to prepare a presentation for one of our clients about media planning, since he works with several agencies and wants to concentrate the media planning in one of the agencies' hands. I visited the "parts of a Media Plan" which I found very helpful. Do you have some other tips? Specifically, we were asked to present a formula for a benchmark acocrding to we recommend to define what reach is needed for a campaign. Basically, we define it according to various factors such as competitors' share of voice, share of market goals etc. but we don't know any formula. We should be grateful if you supply any guidelined in this matter.

The Media Guru Answers(Wednesday, December 02, 1998 ):
You just need to formularize the thinking you are already doing.

For example, you could say that your formula to set reach for a campaign to equal competitor's share is:

Competitor's Reach times an index calculated by comparing the goal share to your current share. (i.e. to increase share 25%, exceed competitor's reach by 25%).

The Guru is not recommneding this particular formula, just illustrating how to turn philosophy into something apparently quantifiable.

Another approach is to build a matrix of your factors and set a 5 point scale for each; for example competitor's share: 0 point if it's equal to yours 1 point if it's 10% better, 5 points if it's 50% better, etc. Suppose you have 8 factors based on the sort of considerations you mentioned. Suppose further that you set a minumum for all campaigns of 50 reach (reaching the majority of the target). Now add a reach point for every point in the matrix. You have a maximum of 40 added points (90 reach), and an apparently highly logical "formula" for getting there.

The cleverness will be in setting up each 5 point scale. Or perhaps youy will have fewer factore and more possible point on each scale.


Wednesday, December 02, 1998 #2192
Dear Guru. It is not still clear to me how to measure or calculate Reach of the ad campaign using media mix. For example, my ads on TV provided 90% reach, and ads in print reached 25% of the target audience. What is the total reach, frequency of the campaign? What other indexes can we find for such campaign? And my second question is about outdoor advertising. It is essential to measure the effectiveness of the ad campaign comparing awereness and sales before and after the ads placing. But that is somehow the post- campaign analisys and my client would like to see some feagures before the campaign starts (pre-campaign). What indexes (like reach, frequency, GRPs, OTS) can we provide to the discription of the outdoor ad. campaign? Thank You very much.

The Media Guru Answers(Wednesday, December 02, 1998 ):
Reach of a medium in a plan is simply a statistical probability. Further, it is generally thought that each medium overlaps each other medium randomly.

So, in your example, if you consider the reach of each medium as a decimal, the probability of not being exposed to TV is 0.10 and of not being exposed to print is 0.75.

The probability of not being exposed to either one, is therefore 0.10 times 0.75 = 0.075.

Therefore, total reach of the mix is 92.5 (if 0.075 or 7.5% don't see it then 92.5% do see it).

Other basic "counts" for a campaign are impressions (OTS), cost per rating point and cost per thousand impressions.

All of these counts; reach, frequency, GRP, OTS, etc are possible for outdoor, if the research has been done, in your country, to count the audience of the locations used.


Tuesday, November 10, 1998 #2144
I need to find out more information on how to figure reach and frequency, especially four week averages as it applies to print, radio and television. What is the best source to use for finding R/F analysis including some work samples. Help me Guru, I want to be like you!

The Media Guru Answers(Tuesday, November 10, 1998 ):
When the Guru started out, Reach and Frequency was calculated manually with the aid of tables and factors. Since then media have become more complex and measurement more detailed. Complicated, multi-step algorithms such as numerous iterations of the Beta-binomial function must be calculated. Now, the computer is virtually the only way Reach and Frequency is analyzed.

Some of the measurers such as Simmons, and MRI have systems for R&F on the media they measure. A few, rare, media such as Telemundo Spanish TV Network, offer sytems (STRETCH2) for their medium.

Most common is the specialized, all-medium software system, such as the one provided by AMIC's sister company, Telmar.


Tuesday, September 22, 1998 #2052
I am working on a national cable buy. First question, please explain VPH. I have been asked to provide the following information: -How many households will my schedule reach and how many times. Of course, I have to have all this information by tomorrow at noon. I have selected my networks and have asked for proposals from each network. The networks inform me that it will take several days to pull a reach and frequency. So my question to you is, can I take the HH's thousands and add them? It this the right way to approach this project. How will I calulate for a frequency. I can give the client the total number of spots, but is there a way to calculate frequency? Please Help? Thanks.

The Media Guru Answers(Tuesday, September 22, 1998 ):
VPH is "viewers per Household" and is used as a simple way to express persons audience in relation to housholds. In other words, if a network has a measured average quarter hour (aqh) audience of 1000 Households and a measured aqh among women 18-49 of 550, then its VPH for women 18-49 would be .55

Estimates of reach are based on modeling from actual past schedules and are typically calculated with computers. These calculations take only minutes, but you are probably facing a backlog in your vendors' research departments or, typically, a turnaround time policy which can be overriden if you apply the right charm or pressure to your sales reps.

Because these models reflect varying audience duplication between one spot and the next and between one network and another, adding household impression would be wrong. Such a calculation would produce "gross impressions" which is much greater than reach.

Frequency is calculated by dividing reach into gross impressions (or percent reach into gross rating points), so you need reach to calculate frequency.

If you have any media planning software at all, such as Telmar's AdPlus or Maestro, you would find that these system usually have a general calculator of cable reach built in.


Wednesday, September 16, 1998 #2045
what is persuasion rating points?

The Media Guru Answers(Wednesday, September 16, 1998 ):
Persuasion rating point is not a standard term. It would seem to imply an adjusted rating point based on testing of the effectiveness of specific media in specific situations. For example, the Roslow Study, for Univision, summarized in the Abbott Wool's Market Segment Resource Locator area of AMIC, measured effectiveness of English language advertising among Hispanics, using persuasion as one of three metrics.

Based on theis study and other data, many advertisers use an effectiveness adjustment when planning Spanish language media.

E.g: If I (the advertiser) want to have 100 GRP per week among my Hispanic target, before buying Spanish media, I wish to account for the fact that Hispanics watch some of the English language media in which I advertise. From Nielsen's NHTI I can see that for every 100 general market media rating points I buy in English language media, I get (for example) 60 Hispanic rating points.

But I know from the aforementioned studies that the GRPs Hispanics receive in English are less persuasive that GRPs of Spanish media. So I apply an effectiveness adjustment to calculate effective Hispanic rating points to which I might refers as "persuasion rating points".

Now the 60 GRPs among Hispanics might become only 33 persuasion rating points. So instead of buying only 40 Spanish language media rating points (100 Goal minus 60 delivered), I should buy 67 (100 Goal minus 33 delivered), to have an effective (persuasive) media plan.


Tuesday, September 08, 1998 #2031
Dear Guru, I'm new in the Advertising field. I would like to know how to calculate the Target Market Reach1+, Reach2+, abd the Average Frequency. TIA. -- SKY

The Media Guru Answers(Wednesday, September 09, 1998 ):
The answer depends upon what data you are starting with. At its most simple, "1+" reach is the same as just saying "reach". If you know the GRPs, and the reach, then the average frequency is calculated by dividing reach into GRPs.

At bottom however, in each medium, TV, radio, print, etc. reach was actually measured at some point, rather than calculated . That is, using respondent level measurement, such as Nielsen or MRI or Simmons, actual schedules advertiser were evaluated for gross audience accumulated and the net reach accumulated, as well as how many people saw exactly one advertisement in the schedule, how many saw 2, how many saw three, and so on. As the Guru stated above, reach is defined as those who saw one or more (1+) advertisements. 2+ or 3+, etc, is determined by adding those exposed to each discreet number of ads.

Taking the results of many of these schedules as a scatter graph, a classic reach curve may be plotted. Or, by arraying GRPs and frequencies in a table, a formula equivalent to the curve can be determined statistically. This formula then becomes a "model" for calculating reaches of other schedules in similar media. Formulae for 2+, 3+ frequencies can also be calculated. There are no simple formulas for doing this. "Beta Bimodal" is one statistical function frquently used. These functions and models are usually built into large computer media planning systems like Telmar's.


Friday, August 21, 1998 #2009
I'm having a difficult time trying to find any material that will teach me how to use CPP when buying network and spot radio. My contacts in the advertsing field tell me that they learned how to use this formula through experience. I know the formula for CPP. However, is there any resource that will show me how to implement CPP when buying network and spot radio? Something that will show me some shortcuts or maybe some examples? Thanks again for your help.

The Media Guru Answers(Friday, August 21, 1998 ):
Once you know how to calculate CPP, its uses are pretty straightforward.
  • CPP is an efficiency indicator. Media proposals you are considering can be ranked from lowest CPP to highest if efficency is a goal
  • CPP can be a goal you give to sales people, e.g. "I'm buying a $100 Women 18-49 CPP, what avails do you have at that pricing?"
  • CPP can be weighted with other factors like rating size or cume
  • CPP can be used to caluculate cpm when the universe is known: CPP divided by 1% of the universe expressed in thousands yields cpm
    (this is only valid when rating and audience in thousands come from the same geography as they do in Network. In spot, where you may use a metro rating but TSA thousands , the formula is imprecise at best)
  • CPP can be used as a bottom line number to compare possible schedules you are putting together

Remember - never average CPPs themselves; total the costs and total the ratings of schedules and calculate the bottom line CPP from the totals.


Sunday, July 26, 1998 #1974
Dear Guru! I have cilent that wants to know the accumulated reach of a 5 months campaign. The campaign was based on a recency strategy; 4 "flights", each flight - 3 weeks, and a break of about 2 weeks between one flight to another. It seems to me not right to sum up the reach of all 4 flights as a total, but to show each flight by its own results Can you please give your professional advice in this issue? Thanks a lot, Irene, Israel.

The Media Guru Answers(Sunday, July 26, 1998 ):
A "recency" strategy generally calls for continuous advertising, not flighting. However this is neither here nor there in responding to your question.

A four-week reach has long been the basic standard of evaluation of a campaign, most likely based on the one time dominance of monthly national magazines in the plans of major consumer goods advertisers -- in the U.S., at least.

"Recency" argues for concentrating on the reach at the point in time closest to the purchase decision, so average reach during the typical purchase cycle is a reasonable way to focus on a recency plan. Of course, in reality, despite an average purchase cycle, in most cases, decisions are made every day. You may end your four-week purchase cycle of laundry detergent tomorrow while your neighbor's four week cycle ends a week from Tuesday. Equally, there may be a day of the week of more opportunity than others, when the product is purchased during a main grocery shopping trip.

A five month cume reach can be calculated. Its usefulness is questionable when recency is the guiding principal, but for other issues, like awareness, it may be relevant.


Friday, July 17, 1998 #1957
Dear Guru, I am curious to find out how user-centric Web research firms maintain "freshness" of their samples. How often and how many samples are being replaced on what intervals by companies like Media Metrix, Nielsen, Net Ratings and Relevant Knowledge? I also would like to know how Media Metrix and RK drawn business samples from what sources. Thanks for your help.

The Media Guru Answers(Friday, July 17, 1998 ):
"Freshness" is not necessarily a desirable element in sampling. Panel data, such as is provided by some services mentioned is explicitly not based on fresh samples.

Results based on a sample of 10,000 are about 3 times more stable than results based on 10 consecutive samples of 1000, and equal to a result calculated from a consolidation of the 10 smaller samples, assuming both samples are equally random within the universes they are supposed to represent. From a practical point of view, the cost of recruiting and installing a sample precludes major "freshening." Virtually all syndicated surveys with large samples (10,000+) and frequent reports (e.g. monthly) are based on panels, rather than new samples for each report. The natural churn of panels is probably 20-30% per year.

The services you inquire about are relatively open about methods. They've been written about in the trade press and the respective web sites are also informative.


Saturday, July 11, 1998 #1945
Dear Guru, I have seen you use "advertising weight" in other response. Please clarify the meaning of this percentage. Thank you.

The Media Guru Answers(Saturday, July 11, 1998 ):
Advertising weight refers to the gross audience of a campaign. It may be GRP/TRPs or impressions. It may be considered in total or by individual demographic segment. While some look primarily at expenditure, "weight" is a better guide to communications impact.

In competitive analysis, each advertiser's weight is compared to all others as a percent of the total weight in the category to calculate "Share of Voice."


Thursday, July 09, 1998 #1941
Dear Guru How do you define Selectivity of media vehicle ? How do you measure it ? Therefore how do you calculate a Press Selectivity Index ? Is it similar in concept to a brand development Index or a category development Index (BDI & CDI) ? andrewwilliam25@hotmail.com

The Media Guru Answers(Thursday, July 09, 1998 ):
The Guru defines "selectivity" as narrowness of targeting audience. For example, if your target is women 18-49 and there is a magazine, all of whose audience is women 18-49, than that magazine is highly selective. A magazine with 80% w18-49 and 20% W50+ is less selective.

Standard print audience measures such as the U.S.' Simmons, MRI or U.K.'s TGI provide these data.

Logically, a selectivity index would compare the incidence of a given demographic group within the population to its incidence in the audience of the medium, with the population incidence set as equal to 100. Thus, if women 18-49 are 50% of the population but 80% of the media audience, the selectivity index would be 80 divided by 50 or 160 (the decimal is moved 2 places to the right for an index). In this sense it is similar to BDI which indexes product purchase in a market to product purchase nationwide, in terms of percent used in the market compared to percent of national population in the market.


Friday, July 03, 1998 #1933
DEAR GURU, THANKS FOR YOUR SERVICE.HOW IS COST PER THOUSAND PRICING IS calculateD IN ONLINE MEDIUM.WHAT ARE THE CRITERIA GO TO DETERMINE CPT PRICING METHODS.WHAT ARE THE POPULAR SITES WHICH CHARGE ADVERTISERS ON CPT METHOD.

The Media Guru Answers(Friday, July 03, 1998 ):
The Guru has discussed this issue many times. If you go back to the Guru Page and select Guru Archives and do an advanced search for this coding:

(web | internet | online)&(price | "cost per thousand" | pricing | cpm),

you will find 25 Guru replies on your topic.


Tuesday, June 09, 1998 #1886
how do i calculate reach of TV+PRESS, Is there a formula

The Media Guru Answers(Tuesday, June 09, 1998 ):
As a rule, TV and press are thought to duplicate in a random pattern. That is, the random duplication formula is appropriate. The reach of each medium is treated as a decimal. To calculate net reach, we combine the probabilty of each medium's NOT reaching the target, to get the combined probability of neither reaching the target. The remaining people are the ones reached.

The formula works as follows when TV reach is 45 and press reach is 37.

People not reached by TV would be 0.55 of the target

People not reached by press would be 0.63 of target

Total people NOT reached are 0.55 x 0.63 or

0.35 of target.

The remainder of target is reached (1.0 - 0.35 = 0.65)

so reach is 65


Friday, May 29, 1998 #1613
1.what is osto's model? 2.In case of an absence of duplication data for publications, how do l calculate the effective reach using 2 or more media vehicles? in such a scenario, is it safe to use the random theory even if multiple readership is negligible?

The Media Guru Answers(Tuesday, June 02, 1998 ):
1) The Guru is not familiar with Osto's model. It may be specific to India, from where you are writing.

2) The Random method is a starting point. If you can find two other similar publications with measured duplication, you can use the duplication ratio from those publications. If you literally mean "effective reach," that is, reach at or above a minimum exposure level, then you need a more complex formula or a computer program like Telmar's ADplus.


Monday, May 11, 1998 #1588
Is this any accepted measurement standard in place to effectively price web-page advertising? Also, has any service been able to obtain more direct response information, re:web-site hyperlinks, other than just counting "hits"?

The Media Guru Answers(Monday, May 11, 1998 ):
Web (banner) advertising is typically priced in one of 4 ways:
  • CPM- (cost per thousand ad exposures)
  • Flat fee ( usually calculated, by the seller, and compared, by the buyer, on an exposure basis)
  • Cost per click ( based on the number of times the banner gets clicked by a visitor to the site where it is displayed)
    This is not a very reasonable method, since the responsibility for making the copy attractive enough to stimulate clicks and fresh enough to keep stimulating clicks properly belongs with the advertiser, not the medium
  • Transactional ( where the payments to the site are based on the visitor's action when arriving at the advertiser's site; including making a purchase, requesting information, or navigating within the advertiser's site)

"Nobody" really counts "hits" anymore. Page exposures of the site and exposures of ads are the coin of the realm.

The latter two models above are direct response measures.


Saturday, April 11, 1998 #1564
Dear Guru, could you please tell me how can I calculate the frequency(exposure) distribution of an advertising media schedule using the Sequential Aggregation, Cannonical Expansion and Conditional Beta Distribution models? Thank you in advance for your answer.

The Media Guru Answers(Friday, April 17, 1998 ):
A good university or public business library should have statistical or advertising methods text books with this information.

For example, in NY City, the Public Library's Science, Industry and Business Library has such books and might even give you the specific references to locate copies near you, if you inquired by phone or e-mail.


Wednesday, March 25, 1998 #1553
How can be measurement error calculated? I would like to know is there any correlation between sample size and data validity? Thank you

The Media Guru Answers(Monday, April 06, 1998 ):
Sample size and data reliability are in a "rule of squares" relationship: A sample four times as large is twice as reliable. Note that "reliable" is the statistical term referring to the chances that a duplicate study with the same size random sample will get the same results, give or take a specified range of error.

"Validity" refers to correctness. It might have to do with whether a question asked can corectly produce a result that is desired. For example, a question like "What will you have for breakfast a week from Tuesday?" may not be a valid predictor of what people will actually eat on that day. But, with a proper sample, it will be reliable in predicting with a set degree of variability what people will eat.

The formula for calculation of error for a given sample is:

The Square root of (P x Q over N)

Where:

p = the percentage result to be tested (e.g. 10% of the people will have bacon)

q = the complement, or difference vs 100% (if p = 10% then q = 90%

n = the sample size

So, if a sample of 500 produces the result that 10% will have bacon, the sampling error for this result is

the square root of (10 x 90)÷500 or

+/- 1.342

so the answer of 10% should be read as "between 8.658 and 11.342" and really means that 68% of the time the same study repeated would produce a percent of bacon eaters between 8.658 and 11.342.

If the sample is quadrupled, to 2000, then the error is halved, to 0.671.


Monday, March 23, 1998 #1548
what is the correct television weight for a campaign

The Media Guru Answers(Monday, March 30, 1998 ):
The "Correct" weight depends on many factors, there is no one correct weight.

One way, but certainly not the only way, to calculate an appropriate level is to follow this checklist:

  • (A) How many new sales / product units, etc are your monthly sales goal?
  • (B) What percentage of the prospects who are successfully exposed to your campaign are likley to buy what you are selling?
  • Divide (A) by (B) to determine with how many prospects per month your advertising must effectively communicate.
  • Using the reach and frequency calculating system of your choice and your judgement of "effective levels of communications, calculate what level of weight delivers the desired effecively reached audience.


Sunday, November 30, 1997 #1466
Hi Guru: I am a marketing student and now doing a promotional campaign. My team needs to develop a media flowchart with reach, frequancy, GRPs, and yearly schedule by by month. I know there are softwares that does that, but we are only students and have no money to buy professional softwares like that. I would like to know if there is freeware that we could use to develop this chart or if there is anything we could refer to. Thanks a million. Please let me know ASAP cos this is due pretty soon. Sarah

The Media Guru Answers(Monday, December 01, 1997 ):
As with most specialized software that makes a job easier and a result prettier, there's also a tedious, less attractive choice.

Telmar's ADplus / Flowmaster is one of the best programs for media flowcharting. However, with no budget, you can probably do an adequate job for your school project with whatever spreadsheet program your computer has installed: Lotus 1-2-3, MS Excel, or MS Works, etc.

Just set your columns to 2 characters wide (enough for dates) for each week and you can create a flow chart. Activity bars can be filled in with special characters or shading.

Any text can be accommodated and total can be calculated; heavier lines dividing months and quarters are also easily done.


Wednesday, November 19, 1997 #1460
I am trying to calculate the value of various components of a partnership (co-op) program. I am particularly interested in knowing how to go about calculating the value of in-store POP/signage. Could you please help me. Thanks.

The Media Guru Answers(Saturday, November 22, 1997 ):
There are two key issues:

How many exposures you get, and

The value of those exposures

It should be simple enought to determine how many people will see your POP; the vendor should have store traffic data.

The value is judgmental. Using as a base something standard like a TV :30 is a starting point.

For how many seconds will a POP sign engage the viewer?

5?

10?

That numbers of seconds' ratio to the :30 is a first step in calculating relative value.

Then, what is the selling power of a sign versus a full sound/motion TV commercial?

50%?

60?

30?

All these factors should tell you thata single POP exposure is worth "X"% of a TV :30 exposure. Combined with the exposure count, you have your valuation.


Wednesday, November 19, 1997 #1459
Does it make any sense to calculate GRPs not having reach and frequency stated? My campaign brings me 530 GRps - whatdoes it mean for me? Could I calculate OTS if I have only GRPs? Thank you

The Media Guru Answers(Saturday, November 22, 1997 ):
GRPs are simply a summation of all the audiences of all the ads in a plan. They give you the "boxcar" size of a plan without any detail. This can be used to compare to other campaigns or other times, in crude terms.

If by OTS, you mean "Opportunities to See," which is equivalent to Impressions, then the calculation is simple. GRPs are a percentage of the population. Whatever your GRP's target group, you need to know the total "universe" of that population for which the GRPs are stated. Then, if you have 500 GRPs, you have impressions equal to the population, times 5.


Saturday, October 18, 1997 #1438
Dear Guru Could you please give me your views/suggestions on the following: 1. How can you set media objectives for a banking client in a market with only two major competitors; both of whom do not have a clear-cut advertising campaign? Would a % above last years GRP levels be appropriate; in proportion to the market share desired? What other parameters should I consider? 2. Qualitatively or quantitatively, how can front page solus positions in newspapers be compared with inside pages and ear panels? 3. And lastly, how do you add TV and press GRPs; for a specific audience? Sorry about the long query. Thanks in advance

The Media Guru Answers(Saturday, October 18, 1997 ):
As a rule, the Guru sets media objectives based on marketing goals, not competitors' activity. Some marketing goals do indeed lead one to comparsions with competition, and awareness of competitors' plans is always a consideration.

If the key marketing goal is share growth, then a proportional increase in weight is one approach. But consider that share, like reach, exhibits an asymptotic curve. In other words, it can't pass 100%, so the higher it goes, the more effort is required to "move the needle."

Consider: You first assume that "X" amount of GRP's are required just to maintain share, on the assumption that competitive activity doesn't vary (and that advertising is the only variable influencing share).

Have you considered whether current share is proportional to share of GRP weight among competitiors?

Would 50% more GRPs grow share by 50%? No, if only because it increases the size of the total advertising arena. Your 50% increase in GRP does not increase your share of GRP by 50%, so calculate the right number to increase share of GRP, if you follow that philosophy.

But since there are competitors, perhaps it takes 50% more weight to gain 25% more share?

Newspaper positions can be compared on a basis of noting, reading, recall, etc. In each country or culture (you are writing from India), the relative power of media and the way consumers relate to them are different.

In the U.S., for example, a front page ad in a newspaper would be quite unusual if not unheard of.

Contacting the U.S. Advertising Research Foundation or ESOMAR, the European Survey, Opinion and Marketing Research organization, or your own country's newspaper advertising association may turn useful up research on positioning.

The Guru treats GRPs of different media as simply additive. When there are established effectiveness factors, as some advertisers have developed, GRPs may be accordingly adjusted before adding, in comparing plans.


Thursday, October 09, 1997 #1427
Does the length of the commercial determine the amount of grps reached? If I schedule a 30ss and achieve X amount of grps, and schedule a 15ss the same amount of times I achieve the same amount of grp's as with the 30ss?

The Media Guru Answers(Friday, October 10, 1997 ):
Very simply, yes. Whether they are watching for 15 or 30 seconds, the audience of the commercial is the same, so the GRPs are the same(never mind theories of channel switching, or we'd be adjusting commercial audiences based on partial viewing).

What can be confusing is that TV buyers often use formulas requiring :15's to be treated as if they had half the rating of a :30 in the same time slot, so that they can most readily calculate a ":30 equivalent" c.p.m. or Cost Per Point.


Monday, August 18, 1997 #1392
Is there a company, or a source, which is capable of measuring Reach/Frequency of any/all media combined?

The Media Guru Answers(Monday, August 18, 1997 ):
The ADPlus system, from our sister company Telmar, can combine reach and frequency from all media. Some media must first be calculated by other systems and then be brought into ADplus for combining


Sunday, August 03, 1997 #1382
I’m looking for any information on how to calculate ratings for future periods under condition of an unstable schedule. Thank you.

The Media Guru Answers(Sunday, August 03, 1997 ):
The potential implication of "unstable schedule" is unclear. Beginning from the simplest application of total usage, seasonality trend and most recent share, up to refinements in trending vpvh or programming and counter-programming changes, all depends on being able to predict the future. "Unstable"" implies that no trends or reliable information are available.

The Guru would be forced to use available data and hope for the best.


Friday, August 01, 1997 #1381
I would like to have an explanation concerning the variables a,b,c,d used in S. Balasubramanian's model of 1990 to calculate the A&P/S ratio.

The Media Guru Answers(Friday, August 01, 1997 ):
If the report you are apparently reading does not sufficiently explain the variables, you can contact Dr Balasubramanian directly. One of his studies is reported here on AMIC within Abbott Wool's Market Segment Resource Locator among the research samples. It's "ETHNICITY AND SHOPPING BEHAVIOR" Ethnic CHoice Orientation (ECHO)" , complete with a Mailto: link to the professor.


Thursday, June 19, 1997 #1366
Dear Guru, I have a set of urgent questions to ask of you. I have a meeting tomorrow, and need your help! 1. How is effective reach calculated? 2. Reach v/s Frequency -- when should one be given priority / importance over the other? 3. Is there any way of taking creative into account while analysing competition? If yes, can a system of weights be worked out? 4. How do you reconcile to the vast difference between reach/frequency deliveries from a Peoplemeter system as opposed to the Diary system? My client refuses to accept a 4+ reach of 30% being accustomed to levels of 70% for the same plan! Would greatly appreciate your immediate reply.

The Media Guru Answers(Thursday, June 19, 1997 ):
1) In any schedule of several commercials, some of the target group will see only one, some will see two, some will see three, some will see four, some five, etc, etc.

The actual measurement is based on tracking the cume of several different advertisers schedules in a single measurement period such as one month of the PeopleMeter.

A mathematical model that will match the measured GRP/Frequency is calculated so that plan deliveries can be predicted. Going more deeply into the actual measurement, it can be determined how many people of each demographic group were exposed to each commercial in the schedule and a model calculated which will predict that performance for a plan.

For example, below is the typical output of a computer models' frequency distribution, showing what percent of the target saw exactly n commercials and what percent saw n+. (this example is from Telmar's ADplus):

                    Frequency (f) Distributions 
                           ------------------------------------- 
                                  % who saw
                                 ---------------
                          #seen exactly  at least     
                          ----- -------  -------
               Target:      f     rch    rch    
               P18-49      ---   -----  -----   
                            0     69.1  100.0   
                            1     11.5   30.9    
                            2      6.0   19.3    
                            3      3.7   13.4   
                            4      2.6    9.6   
                            5      1.8    7.1    
                            6      1.3    5.2   
                            7      1.0    3.9   
                            8      0.7    2.9   
                            9      0.6    2.2   
                           10+     1.6    1.6   
                           20+     0.0    0.0    

2) Reach vs Frequency: The determination of emphasis here can be a complicated analysis making up the greater part of a plan's documentation, under the heading of "communications strategy." A commercial so powerful that it's sell is overwhelming in one exposure might take the "Let's buy one spot in the Superbowl" route as did the Macintosh computer with the classic "1984" execution.

In more competitive situations, competitors' levels are taken into account, clutter in the media of choice, copy quality, etc. Obviously a balance must eventually be struck between reach and frequency based on judging all these factors.

3) There are several ways to take creative into account while setting up reach vs frequency goals;

The complexity or simplicity of the message

The number of commercial in the pool

how close your commercial is to the established "wear-out" level

The balance of :30 to :15

etc, etc. can all be assigned factors and totalled or averaged to give a reach vs frequency emphasis factor

a similar exercise can also set effective frequency thresholds

4) There should not be "vast" differences between effective reaches based on people meter and diary systems if schedule GRP and other aspects are the same. 5 or 10% would be the range the Guru would expect.

A plan with a 70 reach at the 4+ level would be delivering in the range of 98% total reach. It sounds as if your client may be confusing a plan with 70 reach and an average frequency of 4 with 70 at an effective frequency of 4. Or perhaps confusing 4-week reach with a long term cume?


Monday, June 09, 1997 #1364
Many television stations are assigning dollar values to sponsor audio/video tags on television promo spots which are essentially promoting a THEIR television special. Do you see any inherent "VALUE" to these audio / video mentions on promos? If so, how would you value it? Thanks.

The Media Guru Answers(Tuesday, June 10, 1997 ):
It is reasonable and normal for sponsor mentions in promotions of on-air programs or off air events to be presented as having some sort of value. Most marketers would accept that any Brand mention has some value. Certainly, the value is not comparable to a commercial announcement; it lacks any sell and may be mixed in with other Brand's mentions.

The Guru has always calculated the value of these mentions in two steps:

1 - calculate the pure time ratio of the mention, i.e. if the metion runs 5 seconds, that's one-sixth of a :30.

2- Reduce that value by half, to account for the lack of selling power. Of course this is an arbitrary factor. In any case, the Guru would not pay for these mentions, but only consider them as added value in making a purchase decision.


Tuesday, May 13, 1997 #1345
Since "PRICING WEB SITE ADVERTISING" was first published (it's not dated but I'm guessing '96?) have there been any 'advances' in the methodology for pricing web advertising beyond either the ModemMedia model or the alternatives suggested? I am not an advertising professional (and they said us geeks use obscure achronyms?), and I am also looking for a concise FAQ type document that might explain the formulae and jargon (CPM, Frequency, Impressions in your excellent on-line dictionary and Depth which isn't) within the context of web advertising. Are there and other specific media terms (new or old) that are pertinent in a web advertising context (I got page view and hits)? Thank you.

The Media Guru Answers(Wednesday, May 14, 1997 ):
The AMIC article was wriiten in the latter part of 1995, not long after the appearance of the Internet World May 1995 article which it discusses.

By the way, please be aware that AMIC has added a new area, called I-Trac, which discusses web terms and measurement and which includes a Web Glossary

In terms of newer thinking, consider the critqued article's central concepts:

1.Determine the ratio of hits between the web site's log and the number of file "hits" that make up the page carrying the ad. Divide logged hits by number of hits making up the page to calculate what we can call "page views." Then call page views "reach."

Since then, the software which interprets log files has developed so that it can distinguish pure "hits" from the more relevant page requests or "page views" . Hits today is taken to refer to any line in a log file, even errors. (Ad) Page requests is the analog to traditional media's "impressions".

2.Determine repeat viewing of that page and call that frequency.

We more commonly use "frequency" in terms of whole campaigns

3.Determine the success of viewings of that billboard ad in moving readers to the actual web site and call that "depth."

This measurement concept has come to be called "click-through" or ad click rate. Depth was a term only used as defined in this Internet World article.

Today pricing is generally based on cost per thousand (CPM) impressions. Rates seem to range from $15 cpm for the broadest, general audience sites' rotating banners, through $50 or so for search engines' keyword banners up to $100+ for "premium audience" on highly targeted business to business web sites. Another pricing model growing in popularity is "price per click," which charges for each vistor who clicks on a banner. The problem here is that the site hosting the banner must rely on the creative to generate viewr response -- it isn't all the effect of the web site itself. Therre is considerable literature today about how to influence clicks, as well as a growing body of research which argues for the awareness building effects of the banners, regardless of clicking response. Finally, simple revenue based models are the rising concept. In this, sites hosting banners are compensated with a portion of the transaction revenue generated by web surfers they send to retail type sites. An offshoot of this is a model for ad placement agency compensation based on the revenue generated by their placement of ads at recommended sites.


Saturday, February 22, 1997 #1039
I am trying figure out the best way to calculate reach & frequency for the following:

Television Flight:
4 consecutive weeks (250 TRP's per week)
Then scaling back and running 175 TRP's per week - Every other week for the following 8 weeks.

How do you calculate R&F when your schedule runs on an every other week basis?

The Media Guru Answers(Monday, February 24, 1997 ):
There is no basis for believing that an alternate week schedule of 700 total points (175 per week for 4 of 8 weeks) cumes to a different total than 87.5 grp per week for 8 weeks, as long as the scedules are otherwise identical in numbers of different announcements, and numbers of different episodes of the same programs.

It is true that if the schedules per week of activity were solarge as to exhaust reach potentials, the answer might bedifferent, but this is far below such levels

So the total schedule of the first four weeks at 250, plus the 4alternating weeks can be calculated as if there were lower levelconsecutive weeks.


Wednesday, January 29, 1997 #1065
I'm finding very hard to calculate the price of an advertise on the web.DO have any tip or formula that could help me ? Thank you very much !

The Media Guru Answers(Thursday, January 30, 1997 ):
What information or goal are you starting with? As a guide line, search engine keyword banners may cost between $30-$70 per thousand ad impressions.


Monday, January 27, 1997 #1067
My client is requiring me to use adjustment percentages whencalculating grp's in print. I was always taught that reach x frequency= GRP's. Now if I calculate the adjustment to my grp's, the formula no longer works. Is this correct, or do I have to do something else to my reach/frequency? Help!!!

The Media Guru Answers(Tuesday, January 28, 1997 ):
There are various approaches. If the GRP adjustment is just an index reflecting characteristics of the vehicles and their audiences, it may be sufficient to show R/F/GRP/AdjGRP

If the adjustments are meant to change actual value of the GRP, it is usual to recalculate reach from the new, adjusted GRP. Since print r&f is usually calculated from actual schedules, via a "black box" algorithym, rather than from a grp "curve," this may be impractical. If your system allows you to enter factors for each publication before calculating reach, that may solve your problem.

Lastly, even with adjusted GRP to represent some abstraction, the people reached would not be reached at a different average frequency, so one quick and dirty answer, if you must use adjusted grp, is just to divide them by the original frequency, to get reach.

It's similar to the concept of changing a spot coverage area, broadcast r/f to its national equivalent: The GRPs are weighted by the coverage area % and the frequencyremains constant, to calculate the reach.


Friday, December 20, 1996 #1090
Media Guru:
What can you tell me about standard error? Specifically, I have three questions:
1. What goes into standard error? If not the actual calculations, can you tell me what affects standard error: it's not just sample size, is it?
2. What is the maximum standard error that is considered acceptable to the media -- specifically, the advertising -- industry?
3. Related to the previous question, do you happen to know recent standard error levels for suppliers such as Nielsen (National), Simmons or MRI?
Thank you for your attention to this humble query.

The Media Guru Answers(Saturday, December 21, 1996 ):
As the adjacent answer to your previous Guru inquiry details, standard error considers sample size and the size of the specific response. Standard error is smallest for a 50% response in a specific way. 10% or 90% answers have the same standard error. When you hear that a study, like a presidential poll is "+/- 3% that is usually the standard error for a 50% response.

It is interesting to note that the size of the sample is the key and not the relationship of the size of the sample to the universe. In other words, when a broadcast rating service uses a larger sample for New York than for Klamath Falls it does so because of the cost of larger samples being more affordable by larger market's media who sponsor the research, not because a bigger market "needs' a bigger sample. Also note that because of the square root aspect of the calculation, a sample must grow by a factor of 4 to reduce error by half.

"Minimum acceptable error" is quite situational. While an error of +/- 2 on a rating of 10 seems small, it becomes important when a buyer need to decide between programs rated 8,9,10,11 and 12 which might all have identical audiences yet seem to vary by 50%.

As stated above, sample size and response control standard error. Neilsen Simmons and MRI each give the information with their reports to calculate error appropriate to the individual report and findings. Most software used to generate reports has the option to display the error witheach cell of data reported. (You may have noticed single and double asterisks on tabulations of Simmons or MRI data, these are indicators of standard error ranges)


Tuesday, December 17, 1996 #1092
Dear Media Guru Guy:Where can I find out more info on standard error? First, how is it calculated, or what affects standard error? Second, I'm thinking about magazine readership research, and I'm wondering what standard errors some of the popular studies have (MRI or Simmons, for example). And then, I'm also curious as to what, among media gurus, is considered an acceptable or unacceptable error. Do the same standards apply for other media research studies, for example, Nielsen ratings? Thanks Mr. Guru.

The Media Guru Answers(Friday, December 20, 1996 ):
Standard error reflects the range of "tolerance," due to sample size, around the reported answer where the "truth" lies. Inother words, statistical data like "10% of women 18-49 read MagazineX" in reality means that within the range of error expected, if thesame study were repeated 100 times, with a sample of 225, the resultwould between 8 and 12 percent, 68 times out of the 100.

The formula isthe square root of (P times Q divided by N)

Where
P= the percentage to be tested
( e.g. the 10% in the Guru's example, above)

Q=100 minus P ( 90% in our example)

N= total sample size (225 in our example)

or

10 X 90 = 900

900 divided by 225 = 4

square root of 4 = 2

One standard error is the amount of variance sampling causes 68% of the time.

So, at one standard error, 10% is between 8% and 12%with 68% "confidence" At 2 standard errors or within+/- 4 points, or6 to 14% we have an answer we are confident our research will repeat95% of the time. This is why the concept is also referred to as"reliability". It is really way to express confidence that the samesampling procedure will produce the same result.

Most statistical texts can give you considerably more on the topic.


Tuesday, November 05, 1996 #1113
I developed and launched a relatively successful websitethat reaches a very specific audience of users. Demographicshave been developed through an online survey, as well asdetailed "hit" statistics provided by the ISP. Althoughthe audience is relatively small compared to other typesof media, it comprises a highly desirable consumer group.How can one calculate the tangible ($$$) value of banneradvertising on such a site (to companies who wish to reachthis specific audience?

The Media Guru Answers(Wednesday, November 06, 1996 ):
There is considerable controversy regarding the value of Banners. If they are merely logo's which, when clicked take the vistor to the spoonsor's own site or an actual ad. But your question goes to the next stage, the relative value of a highly selective but small audience. The Guru recommends checking the pricing of space in magazines which are selective for the same audience as your site, and comparing that pricing to prices of general audience magazines. The comparison should be on a cost-per-thousand-audience basis, which will allow you to price your site's visitors or banner-clicks according to the same index versus reported general audience sites such as Netscape or Yahoo.


Tuesday, October 29, 1996 #1116
How can you figure duplication numbers for trade publications?

The Media Guru Answers(Wednesday, October 30, 1996 ):
Some publications will have done research on their audiences' reading habits from which you can calculate duplication. Given a few such studies, you can model duplication between othe titles. Some judgement will be called for based on how closely realated the chosen publications are. There is software available, such as Telmar's "MakeDemo," which can help with these estimates.


Wednesday, October 23, 1996 #1119
1. What would be the best way to value ($$$) a credit mention at the end of a prime time television program?
2. ... Value to a :10 opening and :10 closing billingard at the end of a prime time television special?

The Media Guru Answers(Thursday, October 24, 1996 ):
The simplest way to calculate these is by a simple time relationship. A :10 is one-third of a :30, though it might not sell for that price. A credit may have viewer attention for only 2-3 seconds and may be text only. So perhaps half of the time value, at best.


Tuesday, October 08, 1996 #1133
I wonder if you could enlighten me on your thoughts onthe following: currently, within our market, we lookat cost per point (CPP) for TV. This has been the casefor a number of years now. Recently, however, certainTV stations have been trying to encourage the use ofcost per thousand (CPT). Is there any right or wrongway of looking at a cost measurement for TV?My thinking is that CPT does not prove to be stablewhen measured across a time period, simply becauseuniverse sizes change over time (thousands do increasebut not necessariily the penetration into a market).Therefore CPT could be used when measuring off the sameuniverse size, but is not feasible in showing trendsover different years using different universe sizes.CPT works to the advantage of the media owners as it isseen as much less of an amount than a CPP is (at themoment)???Please help.

The Media Guru Answers(Wednesday, October 09, 1996 ):
The issues you raise with the use of CPT as against CPP are real, particularly in that when you have people meter data and the universes are changing they will affect the 'thousands' calculated conceivably more than the change in the audience itself.

In people meter panels the universe can vary if it does not form part of the weighting cells.

In addition as a rating using people meter data is a time weighted average it is not strictly speaking possible to convert them into whole 'people'.

Selling or buying TV based on CPT derived from respondent level people meter data would be fraught with hidden difficulties for both the users and the TV stations.

Note that this question was posed about the South African market.


Friday, March 08, 1996 #1266
Guru:Is there a formula for calculating reach & frequency for trade vehicles.

The Media Guru Answers(Sunday, March 10, 1996 ):
There is no truly simple formula for calculating reach and frequency of any medium. The key datain print R&F are pair-wise duplication between different vehicles and between two or more insertions in the same vehicle.

As the number of insertions in a plan increase, the number of data elements to include in a formula increase. The number of possible pairings for just a 10 insertion plan is 45 ((n x n-1) / 2).

Telmar among others, offers software designed to quickly perform these calculations on defined schedules of media measured by SMRB, MRI, MMR, J.D. Power or others. Using measured media as prototypes, reach of various schedules you might want to consider could then be calculated. From these numerous calculations, you could, by regression analysis, develop a "simple" formula of the form y=ax+b to calculate frequency based on GRP of typical plans of the sort you run in these media (y is frequency; x is grp; a and b are factors from the regression).

A formula of this kind is very specific to the audience dynamics of the media vehicles involved. Please understand, this is not a recommended technique, merely a response to your question.


Wednesday, February 21, 1996 #1755
Dear Media Guru- I have a two part question , both dealing with the same subject, tv sampling error. Suppose ER gets a 20% rating and Seinfeld gets an 18%, both off a sample of 1000 resondents. What are the odds of there being absolutely no difference between these two ratings? This is not as simple as looking up the standard error of each rating. I remember that it has something to do with the standard error of the difference, but I just can't recall the calculating process.Could you please explain? Then to complicate matters, I'm looking at the same phenonena on a grander scale. Suppose the estimated delivery in rating points for a tv schedule is 1000 grps and it underdelivers by 10%- ie. 900 grps. What is the likelihood that the difference had to do with pure chance ( sampling error) and how do I calculate that? I know this is more difficult since you have to account for buying many programs in the estimate and the actual. Naturally, we are assuming that the error differences are all due to sampling, and not the idiosyncrasies of the marketplace or the impurity of the sample. In this case I know the answer is going to be technical, but that is what I need. Thanks

The Media Guru Answers(Friday, February 23, 1996 ):
The Guru loves this kind of stuff. The answer is technical but hopefully, in simple terms.

First, if ER has a 20 rating and Seinfeld has 18, with a sample of 1000 (for that demographic), then the ER 20 rating's standard error is +/- 1.265 while Seinfeld's 18's is 1.215 (See formulas in the Jan 25 18:23 Guru Q&A below).

Note that the absolute size of the error on the 20 is larger but it is relatively smaller. Also note that the range of these errors is such that they can make the two programs' ratings equal: 20 - 1.265 = 18.735 which overlaps 18 + 1.215 or 19.215.

There is a 68% probablity that these two ratings fall within this range. But the swing could go either way on either number. And could fall anywhere within the +/- range specified

There is a 90% probabilty that these two ratings fall within +/- 1.999 on the 18 and +/- 2.081 on the 20. The odds are 95% that they fall within +/- 2.381 for 18 and 2.479 for the 20.

These odds actually relate to reliabilty. That is, if you repeat the same rating study 100 times with the same actual facts existing, 68% of those studies will give ER a rating between 18.735 and 21.265.

Now the 1000 GRP underdelivered by 10% is different As the beginning of the explanation showed, while there is a swing around any rating (a 5 would be +/- 0.689 in the same study), the odds equally favor underestimates and overestimates. This is the same as the reason why small samples don't necessarily underestimate ratings. So in 1000 GRPs made up of 500 spots with an average 2 rating, the sampling error on the individual ratings somewhat cancels out.

To calculate this in an Arbitron measured radio buy using a single survey and one station, for example, the formula is

GRP x ((100 x #spots) - GRP) / sample x Factor)))

"Factor" is from a table provided, specific to demographic and #quarter hours in the daypart of the buy.

So, if your 1000 GRP were based on Adults 18-49 ( with a 1000 A18-49 sample), and a Mon-Fri, 6a-7p schedule, the calculation would be:

(1000 x ((100 x 500) -1000) / (1000 x 2.42)))

or +/- 143 GRP at the 68% confidence interval. Obviously, if the average rating were higher, hence fewer spots or if the sample was larger the variance would be smaller. With an average 20 rating, the swing is about +/- 40 GRP.

So, depending on average ratings and sample sizes, the 10% underdelivery could be within the range of standard error.


Friday, February 16, 1996 #1760
Dear Mr. Guru, Thank you for your last reponse on how to calculate GRP's. You had mentioned that you had explained it fully except for Neilson's calculation methodology. I would be interested in hearing more about this method of calculation as well. Also, is there a "better" way to measure the actual "Impact" an ad campaign has had if you know the actual length of each ad, the frequency the ads ran and the channels(and shows) that they ran during. ie. frequency X length X Audience(rate for each time slot)?? This is obviously a simplified formula, but your feedback on this would be greatly appreciated. Lastly, for television advertising, what are some of the other accepted methods of measurement. Thanks (Again) darrylw@conceptus.on.ca

The Media Guru Answers(Friday, February 16, 1996 ):
It is Neilsen's survey methodology that wasn't covered. They would use the same calculation formulae. The full description of Neilsens methodologies for People Meter, household meter and diary would cover several pages. Contact Neilsen who will be happy to send you methodology booklets.

Regarding "impact" there are as many ways to evaluate this as there are advertisers.

Some advertisers use a factor for copy length based on norms from recall tests. For example, 75% of a :30 is a typical value for a :15.

Some use attentiveness by daypart.

Some use a combination of the two factors.

Some apply the factors to GRP as an indicator; some apply to GRPs and then estimate reach from those adjusted GRPs as an impact indicator.

The frequency of a schedule, as discussed so far, refers to the average frequency of exposure for all pesons reached.

There are those who use "effective reach," counting only persons reached at least 3 times (or any designated minimum) when evaluating the impact of a schedule.


Thursday, February 15, 1996 #1761
I would like to know a DETAILED calculation for GRP's for Television advertising. I assume the frequency is the number of times the ad ran across all channels. But how do I calulate the Reach for a T.V. AD. Is it based on the rate cards of the networks?(if so, how) or is it based on direct audience measurement. As much detail as possible would be greatly appreciated. darrylw@conceptus.on.ca

The Media Guru Answers(Thursday, February 15, 1996 ):
Calculation of GRP does not depend on knowing reach, though reach x frequency is ONE formula. Reach is the more complex calculation, GRP is relatively simple (see other formulae in the adjoining question).

Reach has no relationship at all to rates, nor to commercial length, for that matter. Audience research surveys such as Nielsen can tell us the audience of individual programs. The net unduplicated audience, or reach, of actual advertisers' schedules, examined over time covered by a given survey period, typically four weeks, can be determined.

When many such scehdules, usually thousands, have been examined in that way, "curves" on a graph can be drawn representing the intersections of reach values with schedules' GRPs. The graph curves because each added announcement adds fewer new, unduplicated people toi the reach of the schedule.

The curve can be expressed as a formula y = ax+b, which then can be built into the computer model which media planners use to quickly calculate reach from given GRPs and sometimes other descriptive details of scheduling, such as average ratings, numbers of different networks or programs, etc.

The Guru is now nearly out of details unless Nielsen survey methodology is of interest.


Sunday, December 17, 1995 #1806
To calculate the household rating you consider as a base the Total ousehold universe or the total TV set universe, I just wondering because if we take the total household in some moments the HUT coul be more than 100%. Thank You in advance.

The Media Guru Answers(Sunday, December 17, 1995 ):
The base is HH with TV sets. Yes it is theoretically possible for HUT to exceed 100%. HH share, which uses a base of HH *currently* viewing often does sum to more than 100%. In the olden days, when TV's were powered by dinosaurs on treadmills (The 1950"s) we used SIU ("Sets In Use") as a base. Multi set homes were almost non-existent and TV homes were less than 70% of all HH.


Saturday, January 28, 1995 #1876
Do on line services have a reported user count, minute by minute, to calculate interactive GRPs?

The Media Guru Answers(Saturday, January 28, 1995 ):
Most on-line services have the ability to capture minute by minute usage of their services since some of them bill customers by the amount of time they use. Some of them also have a profile of the user based on questions provied at the time of sign-up. So theoretically, they could produce average qtr.hr. ratings or even minute by minute ratings on the demographics they capure and therby calculate Target GRPs. Some may even be able to capture what you viewed while on-line (program ratings) provided such viewing is on their server. Whether any of them would care to part with that information for a fee or otherwise is speculative at best.


Wednesday, January 18, 1995 #1878
How do I calculate GRPs?

The Media Guru Answers(Wednesday, January 18, 1995 ):
Reach x Frequency = GRPs



Back