To jump to the additional adage.com exclusive section, click here.
If one topic seemed pervasive in industry discussion about the 1998-99 TV season, it was audience measurement. Once again Nielsen Media Research was at the center of the controversy. Subjects ranged from a dramatic drop-off of the 18-to-34 viewership of broadcast networks at the start of the season, to delays in delivering data, to the accuracy of that data.
In an ever-fragmenting media universe, audience-measurement issues become more crucial. To create a forum for these issues, Advertising Age organized a roundtable of four of the most knowledgeable people in the business on the subject: Barry Cook, senior VP-research for the ratings service at Nielsen Media Research; Kate Lynch, VP-media research director at Starcom USA, Chicago; David Marans, senior partner-director of U.S. media research and resources at J. Walter Thompson USA, New York; and David Poltrack, exec VP-planning and research at CBS. Ad Age Media Editor Chuck Ross posed the questions.
Advertising Age:There's been a lot of talk this year that we're basing billions of advertising dollars on data from Nielsen Media Research that people in the ad and network community find questionable. David Marans, could you address your concerns on the kind of data you're getting from Nielsen?
David Marans: There really are a lot of positives and negatives going on. You get what you pay for and we're not exactly spending tens of millions of dollars on the agency side. So maybe our needs are not necessarily being adhered to. We have to applaud Nielsen for having an operation that produces so much national, local and other data. One of the reasons we don't think there can be a competitor so quickly is because it's a gargantuan enterprise. Also, when you're generating extraordinary numbers, there's going to be problems--things that look like anomalies. On the other hand, the idea that diaries are being used is laughable. The idea that People Meters have been labeled the best technology as we get into a new millennium is absurd. It was about a decade ago that we talked about a truly passive meter, but it's disappeared.
I'd like to hear Barry address that.
Barry Cook: The challenges are daunting and we're investing a lot to try to meet them. Thank you for your acknowledgement that we do produce a lot of data. We try to make the quality as good as it can be. The challenge of measuring local markets is huge. The diary is a device--a measurement device--designed a long time ago for a world that is very different from today. It isn't ideal. We have hopes that we'll be able to make improvements at that end of the spectrum. At the other end of the spectrum--the People Meter or the current set meter--these are not the tools for a future with a great deal of interactivity and without channels; a fundamentally different medium.
AA: AP stands for active passive. So this is the passive People Meter?
Mr. Cook:No. The AP is a set meter that uses active codes plus passive signatures to identify what's on the set. The passive People Meter--we, frankly, put that project on hold while we were working on the AP set meter, because we just had to make a tough decision--that the AP set meter couldn't wait. We are considering revving up the effort on the passive People Meter again, because there's a lot of industry interest in getting finer time-segments-of-persons-viewing measured than you can possibly measure with an active measurement system.
AA: This AP set meter, is that satisfactory alone, without passive People Meters?
Mr. Marans: No, it's not.
AA: Kate, do you agree the AP set meter is something that's needed?
Ms. Lynch: It's needed to measure TV in a world where we don't have channels. So it can track what's actually on the TV screen.
Mr. Marans: It's absolutely needed. They will not be able to measure digital TV if they don't convert to an AP meter.
AA: On the other hand, do you agree with David that they need to move forward on the passive People Meter?
Ms. Lynch:Yes. As TV continues to develop over time, we need more accurate information about who's watching at any particular time. That's one of the gripes I have with the current situation. Nielsen collects all this fabulous data, and I believe it's good quality. I'd just like to see some of it on my desk quickly.
Mr. Marans: The truck is on its way. [LAUGHTER]
Ms. Lynch:I know, I know, I'm so desperate for it. And the merging and the linking of the commercial ratings and break information with the program ratings. I had two secretaries watch the [March 21 Academy Awards] from beginning to end to record the minutes of the break, so we could match that up with the minute-by-minute data that we had to buy especially for this project. That's crazy. That should not have to happen.
AA: What do you think on the network side about getting a passive People Meter or some device like it?
David Poltrack: Well, in the first place, TV in the U.S. is measured in three different ways. If you are a network, or a national cable network, you are measured by People Meters; if you are a local market in a major market you are measured by set meters with diary supplements; and if you are in a small market you're measured by diaries. And someone who actually thinks that these three numbers have any relationship to each other is crazy. So, that is one of the major issues going forward.
Mr. Cook:So what should it be, diaries everywhere? [LAUGHTER]
Mr. Poltrack: Oh, right, diaries are the answer. Seriously though, on the AP meter clearly you need a meter that's going to measure in the digital age. You need a metering technology that will be adaptable to the interactive nature and the addressability of TV in the future. That's because--sometime in the near future--we'll need samples of a much greater magnitude than the ones we currently have.
My problem with the current technology is that if you look at the current national TV buying and selling process, it reinforces the acceptance of the limitations of the current measurement system. It's a chicken or egg situation. If we don't change the buying and selling process, the economic momentum is not there to change the measurement system.
If I were in the Nielsen organization, I would say, basically, "I've got a mature product, it's given me a nice margin," and I'd have a maintenance strategy concerning that product. And then I would say, "I have a future, and the future is in measuring the Internet, measuring the interactive world of the future." That's where my future growth is going to come from. I wouldn't make the high-priority be to fix the system. I don't see where the incentive is. So the question is: how do we create the incentive to fix this system?
AA: Barry, does Nielsen have that incentive?
Mr. Cook:I think we do. Convergence is the pathway that links the growth opportunities of interactivity with the core business. Do we measure them both in the same sample yet? No. But, it is inevitable. As TV viewing becomes intertwined with interactivity it doesn't make sense to measure TV viewing without interactivity in that sample. It's also questionable whether a stand-alone Internet panel that doesn't measure the TV usage in that panel is really the right tool to express what's going on in the Internet.
AA: But there isn't a competitor your clients can turn to, as of today, anyway.
Mr. Cook:We get a lot of encouragement to improve our system. That encouragement occurs in part through the graces of print vehicles like yours and in direct communication with customers. We have every incentive to grow our existing business any way we can--to provide better data and to measure more complicated and different viewing behaviors that will be occurring.
I have a concept of where we fit in the big picture. And I've been selling this internally at Nielsen. There's a lot of change that will happen in the way people view the medium. That viewing behavior is the piece we have to measure. We have to keep up with it and keep ahead of it in a sense. But then there's another piece--the business models--that straight ad-supported TV isn't the only game out there today. There's a subscriber model; there's a subscriber plus ad-supported model; and there's a pure ad-supported model. And other models that are nothing like those things become possible with the new technologies.
The advertiser can play a role in different parts of the distribution chain. The programming can be sold without ad content. All these things can change the revenue model for everybody in this room. We want to get thinking about it as soon as possible, because there is lag time in building the system to measure it.
Mr. Marans: I am alarmed by something Barry said a little earlier, because it is true. The question was, "What happened to the passive TV meter?" And he said Nielsen put it on hold so it could marshal resources to focus on the AP meter. How could a company play triage that way? It makes me feel uneasy about the changes that might occur unexpectedly in the coming few years if this is a corporation that's unable to walk down a couple of paths simultaneously. Is the revenue not there? Is the interest not there? There's been other problems like delays in N-Power. Short term I'm very optimistic about Nielsen, though I could be wrong. But when we talk long term, I'm concerned.
Ms. Lynch:I'm optimistic long term, but unhappy short term. What I saw in Palm Springs [at a Nielsen client meeting in March] was some great thinking and ideas and questions from Nielsen about the future. Lots of different systems, news of technology and new tools. But nothing that helps me with my business problems today. That's what really frustrates me.
If you think about the way the agency world is changing, we have media independents, media specialist companies merging and people focusing beyond just cheap TV.
Media buyers are focusing more on strategic planning. The tools you have to help us with these areas are not up to the business needs of today. There's very little that allows us to create a competitive advantage for our agency vs. another agency. We're wasting a lot of money not being able to evaluate quickly and inexpensively the value that we get from TV today.
AA: Barry, do you want to respond to the short-term and long-term issues David and Kate raise? Or if you want to leave at this point I understand.
Mr. Cook:I'm just reflecting on why you didn't tell me who was going to be here.
AA: Oh, no, I thought I did mention who was going to be here.
Mr. Cook:No, I'm fine. I love my customers. [LAUGHTER] Couldn't have a business without customers, could we? No.
David, maybe I didn't give you all the background that went into our decision to put the passive People Meter development on hold. We didn't actually stop it, we just cut way back on the staffing for it. There's still people working on it, but not at the level that we had been working before.
Mr. Marans: I don't need a step-by-step, but that answer is given on several projects, whether it's N-Power, or People Meters locally. That's the answer we hear from Nielsen: "We're interested, we think it's important, but, frankly, we can't juggle everything necessarily at once." That's the end result.
Mr. Cook:Like any business, you have to make choices. On the passive People Meter, the choice was easy. The choice was, should we try to lead the world in the development of a difficult engineering and software problem--image recognition technology? Or let the world do it because they're going to do it anyway, and then we come in at the point where the technology is ready to be applied?.
We could beat our head against this wall for a long time, but we don't have the money to invest against it, like banks, like the U.S. Defense Department and whoever else has need for image recognition technology. From what I read, progress is really starting to happen on image recognition, so we're thinking of cranking up our project to take advantage of that.
In general, we don't do everything at the same time--you're absolutely right. Maybe we need to come to you more aggressively and say, "We need to invest in things for your future. Here's the proposal. Let's look for a partnership or a subsidy for this development work. Because we can't do it out of existing resources."
Mr. Marans: It just never happens.
Mr. Poltrack: One of the issues I have with Nielsen [relates] to the sweeps and local measurement. We came up with this idea for expanding the sweeps and checkerboarding and getting rid of the sweeps concept. The American Association of Advertising Agencies and the networks were involved. There's a lot of interest and support.
The year before we did this, there was a problem with local measurement. Nielsen sent a letter to its clients. I believe they said they wanted to increase the sample of every local market 50%, and the first 15% of the increase, as I recall, they were going to do for free, and you could go up to 50%. They gave a pricing of what it would cost to do that.
I would argue if you increase the sample 50% and spread it out over some more weeks, you could accomplish what I wanted to do. I went and computed that for every single market in the country, and figured if all the stations agreed to do it in all the markets in the country, it costs $4 million.
When we proposed checkerboarding to Nielsen, we got back an estimate of $15 million. Those kind of numbers mystify me. Where is the difference between the $4 million that Nielsen had proposed the year before, and the $15 million? I don't see where that comes from. I was told--again, in a frightening way to a certain extent--that it's because the entire staff of the diary-base measurement is part-time, and Nielsen would actually have to hire full-time people.
Well, I don't know how many millions of dollars the local measurement system brings in, but I think a business of that revenue stream would usually have some full-time employees. There's not a major kind of investment.
But the investment at Nielsen is [made in] challenging Media Metrix as the measurement service in the new era. Part of the reason I've come to that conclusion is because if I put myself in the position of the management of Nielsen, that's what I would be doing. That's what is really frightening about our system. Not only is it not very good, but every indication is it's going to get worse--relating to what Kate says--short term. At some time we may pull a switch and go to a whole new high-tech measurement system that will change the whole paradigm. But at this point in time, the current system is more than likely going to get worse, not better.
Mr. Cook: I'm still delighted to be in my shoes. [LAUGHTER]
AA: Barry, could you respond to what David has said about local measurement?
Mr. Cook:Sure. If there were a ground swell of enthusiasm coming from the entire local customer base it would have happened. It's not that we could stop it. And we couldn't make it happen if the local customers didn't want it. Frankly, it was a good idea, and the marketplace wasn't interested in that good idea as far as we could tell. Because we brought it to them.
AA: Doesn't Barry have a point? One of the problems is that Nielsen serves a lot of constituencies.
Mr. Poltrack: And I understand that. That is, again, the problem with the system. Why would we expect a TV station with limited resources in market No. 85--that currently has a system that is biased in its favor--to make the integrity-oriented decision to pay significantly more for a different system?
That it is necessary [for that to occur] for our industry to move forward is a real problem. The station shouldn't be expected to do something that's not in its best interest, and Nielsen shouldn't be expected to do something that's not in its best interest. But, collectively, in our best interest, getting rid of the diary-based measurement system is definitely something that we all would like to see happen.
So the question is the economics; how do we put together the situation that allows us to do that? In this particular case--local measurement--cable TV operators offer us some opportunity. At least there you can point to a constituency that has an economic reason to do it.
AA: On the local measurement angle, and sweeps, it seemed that CBS CEO Mel Karmazin was putting the onus on the agencies at the Four A's meeting, saying if you guys aren't happy with this local measurement system, don't use it. Is that realistic?
Mr. Poltrack: You can develop a system that says we're going to take the sweeps ratings and the network ratings for 52 weeks, and we're going to downgrade all your numbers. If you run a contest, we're going to take 10% off all your numbers. Then, all of a sudden, the incentive to maintain that system goes away. Or at least is diminished.
AA: David Marans, your agency runs a big spot operation. Do you like that idea?
Mr. Marans: The research company is supposed to be the universal provider. It's a little alarming that I would have to, therefore, work on all these other adjustments--which we do, and others do. Has it come to that? They're supposed to provide the currency.
Ms. Lynch:It has come to that. We as a group of agencies and media specialists should be doing that as part of the Four A's and part of those kind of organizations. If we do it with a collective voice, that might lead to more change down the line.
Mr. Marans: I want to say that we're having second thoughts about local market People Meters. The incredibly low but viable ratings for more and more vehicles is making us have to think more about the sample size. Do I really want to have 300 homes in Buffalo with a People Meter? With low ratings you must have a larger sample size.
AA: And economically it's not feasible to put huge sample sizes in the local markets?
Mr. Marans: With People Meters? As the pricing currently seems to be? No.
AA: So then if you're moving off that position, what would you like to see?
Mr. Marans: I don't want to say this--we applaud Nielsen for putting more set meters in more markets, it's been accelerating--and I think it's a very valuable addition to the research.
AA: But on the other hand, you opened up by saying the diary system is just a disaster.
Mr. Marans: Yes.
AA: So where do you go? If you're moving away from local People Meters, and you don't like diaries, where are you?
Mr. Marans: Well, if we had the absolute answer to that, you'd be interviewing us about a new research organization, which has not been created, called dot-com.
Ms. Lynch:I agree. People Meters, long term, in a world with fragmenting audiences and more viewing choices, are not going to be an economically viable option. But the set meter with some degree of modeling--of other data about demographic audience, different programs--has got to really be the route forward.
Mr. Marans: Agreed.
AA: David Marans, although you've been critical of Nielsen, I know you've only been lukewarm about a potential competitor, Smart from Statistical Research Inc.
Mr. Marans: There are other organizations out there, other than Smart, to tackle some of these issues. We're talking to people at Adcom. We're talking to people at TCI. There are other organizations that exist. There's a slight flurry of activity. With convergence, my goodness, we may have five or six people at our door.
The reason we've been hostile to Smart is that we believe it's "me too" technology. The People Meter, and the cost of an infrastructure, and the talent involved in replicating that service are extraordinary. Where is the sales staff going to come from, where are the engineers going to come from? I'd rather leapfrog ahead if you're going have a rival to Nielsen, and use very different technology.
AA: So it's not the technology for the new millennium, in your opinion?
Mr. Marans: Nielsen currently isn't, and if that isn't, why should we sign with someone who's doing the same thing?
Ms. Lynch:What I liked about Smart is its software and access to the data. That's one of the reasons I'm heavily supportive of it. I also like the idea of competition. But I also agree with David. As we move forward five years, I don't believe we're going to be looking at traditional TV panel measurement--we're going to be looking at more universal measurement using cable TV set-top boxes.
AA: David Poltrack, CBS has been intimately involved in the funding of Smart thus far. Is this a system that even before it rolls out--if it rolls out--is obsolete?
Mr. Poltrack: There are two roles that Smart plays. The first is the competitive role. The second one is that it can end up being a better system. Now, we don't know at this point in time whether it'll end up being a better system. But it is a competitive system.
The fact is that the Nielsen People Meter agenda was basically accelerated when AGB came on board. And the Nielsen N-Power initiative started and was accelerated when SRI introduced its Smart software. That's the value of competition.
I agree with David that it would be nice if we could invest this money in a system that's more designed to carry us for the next 20 years, and does deal with things like addressability, which are going to be a major part of our future. And it doesn't at this point in time. But, if you get two organizations up and running, trying to build a system for the future, you're more likely to end up with a system for the future than if you have one system, one company in a monopoly position trying to build that system.
Mr. Cook:Let me bring it back to something that Kate said before, which was that she wanted us to be more helpful in providing information that really helps her do her business. She mentioned commercial audience measurement is more than a chore right now. With diaries it probably isn't very meaningful measurement.
Ms. Lynch:For sure.
Mr. Cook:People do not attend to their measurement task that much, when they're paying attention to TV or the rest of their life. You do need different kinds of tools to get data you should trust at that level of time-resolution. One of the reasons we don't report very much about commercial audiences is because our data collection tools aren't designed to measure commercial audiences. But if that's important to you, ring in and we'll work on it.
Mr. Marans: I'm sorry, but I don't buy that.
Ms. Lynch:The big issue on commercial audiences is getting the people to turn over during the breaks. The amount of variation between programs, and between times of day and between networks of the tune out during the commercial breaks is enormous. It's incredible that billions of dollars are traded in this country that don't reflect that.
Mr. Marans: I concur. One of the reasons we pushed for the passive meter was so we'd be able to see what the eyes on the set were during the commercial breaks. The buyers have a slightly different agenda than the sellers in terms of what's important to us. We're not in the programming business, the networks are; we're not in the local TV business, the networks are.
I don't really care what the rating is for "ER," nor if it's on cable, or syndication, or a broadcast network. I want to know, for a target audience that we've selected, how many of them are viewing the Taurus or the Listerine commercial during the break. And, ideally, what influence the commercial has on them. When we're talking about age and sex, the data we get now is very rudimentary. It's almost laughable.
AA: So how far away from getting this information are we?
Ms. Lynch:A lot of people are starting to model based on zip codes, Spectra, Prism kind of data, and then apply that data.
Mr. Poltrack: Along those lines, I think the Simmons Market Research Bureau/-MasterCard thing is really very interesting. Simmons has got an arrangement with MasterCard, where they are going to get a tremendous amount of product usage information from . . .
Ms. Lynch:. . . a census of all MasterCard owners and their transaction database, and then merging that with . . .
Mr. Poltrack: . . . the Simmons database. That kind of information can tell us a lot more about which TV programs are better for a particular advertiser than the age and sex of the people watching.
AA: Once you have that with the Simmons data, how does it translate to the way you advertise on a particular TV show and so forth?
Ms. Lynch:You could use all TCI subscribers with some modeling, some ability to add demographic information or people viewing. An ID code for each person. So you've got a census of TCI subscribers viewing data, you've got all the credit cards that they own, and then you've got some kind of another mixed media survey as well.
AA: Would you know that I live in a certain household and I have a MasterCard? How does that work?
Ms. Lynch:I wouldn't know your name, but I'd have an ability to reach you through TV via TCI.
AA: So you would know that someone in that household, that TCI household, has a MasterCard?
Ms. Lynch:Yes. And I'd know your viewing habits. In my dream world you would be sticking your credit card into your TV, or there'd be some kind of a key card or an ID that showed you were watching, and you were buying things.
AA: What are you going to know from this Simmons/MasterCard project?
Mr. Poltrack: Simmons has very rudimentary TV data. So, what you would know from Simmons is this person is a "Touched By an Angel" viewer, and this viewer has this purchase profile vs. automobile sales.
But the weakness in this particular case--is the weakness of the viewing data you're getting out of the Simmons data bank. But there are a lot these things that are coming forward like DirecTV. DirecTV is addressable. DirecTV can tell me a lot of things about the people who are watching my programming on DirecTV that I don't know about the people who are watching over the air, or the people who are even watching on cable. There are beginning to be a series of players that have large enough bases of viewers with addressability that can allow the kind of things that Kate's talking about to take place in the near term future.
But all of these things are going to involve some degree of modeling, because you're going to have a lot of information about certain people, and virtually no information about other people. So how do you model the haves and the have nots?
Mr. Cook:I'm not against modeling. I think modeling to relate purchase transactions to media behavior is important and necessary. It's probably adding a lot of value to the media information. It isn't changing the currency of the media information. Modeling to get the currency makes the currency model dependent. That's where I think we have to be careful going that way--I'm not saying it's impossible--but we have to go very cautiously.
Mr. Poltrack: But you're also talking about a universal currency. One of the things about adults 18-49, adults 25-54, was basically, it created a universal currency for the advertising marketplace. There is really no universal currency.
If I am an advertiser and I have a hot product in a hot category--if I'm right now trying to establish an e-commerce Internet site--I'm a lot different than a brand of household cleanser in a rather mature marketplace. The whole valuation of the medium is totally different for me. One currency will never effectively cover both situations.
Mr. Cook:There are lots of ways to come to a determination of what the CPM for "Touched By an Angel" should be. You want to bring in a lot of information about the demographics and maybe beyond that you want to bring in information about purchase behavior.
Mr. Marans: To add a philosophical note here, my question is can we do business right now the way we've been doing it? Can we purchase time nationally, and in cable, and syndication? Yes, we can and we do. The question is do we have the resolve to move to the next level where we want to be and need to be? There seems to be scant resolve--or negligent resolve--in local TV stations to do anything. And one of the things we suggested down at the Four A's in New Orleans is it's time the agency buying community get together--and the letters are going out--to spend about the next year figuring out through this plethora of options where we agree we can take a stand to do something about it.
AA: So, is this leading to a joint industry committee? The U.K. model type of thing?
Ms. Lynch:Illegal in this country.
To read the first page of the interview, click here
AA: Well, I don't know if it is illegal.
Mr. Marans: I didn't say joint industry committee. I just say that the Four A's has committees--research, spot buying, national buying and policy committees--that can discuss issues so that if . . .
Ms. Lynch:So you can brief Nielsen . . .
Mr. Marans: Exactly. I'm just saying I would like to get our houses in order by having, for example, spot people talk and say, 'Hey, you know, if you change the business this way, we'd better prepare to do this.' And to have the media director say, 'Hey, if the research people on the agency side want to do this, then we're going to have to spend this type of money.' I would like to do a situation analysis over the next year.
AA: So what comes out of this a year later, David? Just another position paper that everybody ignores?
Mr. Marans: First of all, I think it's important that we get out of our own bedrooms, where we talk to ourselves--at Thompson, at Y&R, at Ogilvy, at Starcom--and speak together and see if we can find out what's in the air.
Ms. Lynch:About what's important.
AA: Does this not go on in the Four A's various committees now? Isn't it supposed to?
Mr. Marans: I'm suggesting a little more joint effort so that the spot people and the national broadcast people are in the room together. We can do it. I would like Nielsen to hear a voice--a unified voice from us--instead of 17 different positions on the 12 different issues we've discussed thus far.
AA: And then what happens at the end of the year? OK, so there's a unified voice, and so someone writes it up as a paper, and then? Nielsen says, 'OK, that's fine,' and they ignore it and everybody keeps doing their own business.
Mr. Marans: The past is not necessarily a prologue for the future. I want to be an optimist.
Ms. Lynch:I think that the world we're living in is changing. In five years there are going to be five media buying shops, and we're going to need to either work together and have better tools and better data, or we're going to be competing ferociously against each other. And it's whoever is able to deliver better product back to the client who's going to win. And that's going to involve working with the AdCom people, the TCI people; working with Nielsen to build a new tool or database for the Simmons people. But I think maybe in the short term joining in together as an industry will achieve some of the things we need.
Mr. Marans: I want for us to focus more clearly on things together. And, as Kate said earlier on, we'd like to have tools from the audience research providers so we can do that.
AA: But we have had lots of delays by Nielsen in getting tools to you. For example, Dart, now called N-Power, has been subject to extensive delays, and, at the time of this roundtable, it doesn't look like it will be out in April either. Barry, what's going on?
Mr. Cook:We have a lot of people working on it--they're bright people and it's a tough project. We have an enormous amount of data and we want to provide access to the data in a way that the customers have asked us for. And we're very disappointed that the delays have come because we didn't want them.
AA: Is that explanation satisfactory to you guys?
Mr. Marans: No, and I'll tell you why I was disappointed. I found it very odd that as soon as Smart became a more viable option, the more you heard about Dart.
Mr. Poltrack: The frustration is that Smart came up with a system similar to the European system that basically cost you nothing on top of the base subscription rate. Nielsen introduced a system [originally called Dart] that costs you a huge amount of money on top of the base subscription rate--and they haven't delivered it. So it's not like they haven't delivered something that they were pressured to provide. They've built a model that basically is an economic model where they're going to make money on this system. They're charging a lot of money for this system, and they still haven't provided it.
Mr. Cook:If we gave it away--but didn't provide it--you'd say 'they're not serious because they know they're not going to make any money from it.'
Ms. Lynch:I can't wait to get it.
Mr. Poltrack: We were the first ones to sign up, Barry. We signed up, we're paying. The level of frustration is just magnified by the fact that we're looking and we're saying, 'well, if all these people are signing up there's got to be a lot of money there.' Yet it doesn't seem to be getting done.
Also, there's another aspect of this, which is when an organization misses deadline, after deadline, after deadline in terms of providing something, you begin to wonder about how reliable the product is going to be once it's there. It's a question of management control. If it's so difficult to mount, what is the maintenance going to be like? And how much faith can you have in the product once you actually get it? That's the other thing that's a little frightening.
Ms. Lynch:I'm totally optimistic, and hoping that, because of the delays, it's going to be a very stable, solid product.
Mr. Cook: Our plan is to get it right and then release it, really. Not to delay it indefinitely.
AA: There's been a lot of talk, especially at the beginning of the season, about the decline in 18-to-34-year-old viewing, especially on broadcast networks. David Marans, where are we with that now?
Mr. Marans: As long as the Nielsen data are the currency for making decisions, for buying and selling, and as long as there is nothing from the company saying anything's wrong, we are going to accept it as a fact of life.
On one hand, there are certain aspects of the last few months that make sense that there might well be a decline. In much of the U.S. the weather was warmer than usual. I understand Blockbuster Video had a fairly good quarter, as other video stores did.
The Internet's allegedly growing, the measurements indicate. So, why wouldn't there be a relatively small, but unusual decline? It's very plausible. Particularly, if you [asked me] which age group might have a decline, I would say probably people in their 20's would be the ones you might expect.
The only thing that concerns me is seeing if it's a sustained trend. If it's not, then I might be a little bit worried. Hmm, why now? Why didn't it last?
AA: Are you also worried if it is a sustained trend?
Mr. Marans: No, it doesn't matter for J. Walter Thompson, and I assume all buyers of media. We're supposed to just follow the audience, whatever they do or wherever they go--it's not that important to us.
AA: But do you know where they're going?
Ms. Lynch:I believe it's a real change in people's behavior. Some of the programming that's been on over the past few months is not attractive to the audience. And people are doing things differently now. A lot of people are using the Internet for news gathering, for entertainment, information and for sports. If you look at the increase in usage of those sites and visits to those sites, by either heavy print users or heavy TV viewers, it's a real change.
AA: But do you actually have any correlation, any studies that you've done that verify the decline?
Ms. Lynch:No, we haven't done it with TV viewing. We've done it with print readership and MRI data. It's not panel data, so it's not completely checkable. But it makes sense. I only really have the need for lots and lots of detailed research if things don't seem to make sense. If the research backs up what you feel to be happening in the marketplace, then it's fine. It just makes sense.
AA: If this decline is real, should buyers look to other types of media for that target audience?
Ms. Lynch:The planners are doing that already. We've seen a significant increase in the use of the Web by a lot of accounts we work on, and you just do different things on TV. There's lots of places to reach 18 to 34s and 18 to 49s on TV.
AA: David Poltrack?
Mr. Poltrack: If you look at the lifestyle situations, I think Kate's right, There are logical reasons to expect that 18- to 34-year-old viewing is down, and it probably is. The issue is to the extent that it is down that the total loss we're seeing is a function only of lifestyle change. And the problem is it happened too quickly, too dramatically to be totally the result of this phenomenon.
This is a long term phenomenon. It's gone on for several years, I expect it to continue to go on. The issue that we have is all the work we have seen focuses on the fact that in-tab rates do exacerbate the situation. The people who are out-of-tab in low 'in-tap' periods are heavier viewers, particularly in the young categories. Therefore, there is an artificial inflation of this rate of decline when in-tab rates fall below normal levels.
AA: David Marans, do you agree?
Mr. Marans: The data has shown that there's a problem and it's exacerbated. The thing is how much would that affect the overall number--1% or 2% out of the 6% or 7%? We don't really care whether the numbers are high or low, we expect them to be accurate.
Mr. Cook:Let me point out 18-34 adult viewing in late night is increasing. So, it isn't as though they're abandoning TV. There are changes going on in both directions, more so in the direction of reduced viewing in most dayparts. But it isn't as though it's some sort of universal phenomenon you can pin on a methodological thing. There's probably some small relationship between in-tab and reported levels of the in-tab sample. But it doesn't gel as the explanation for what's going on.
The other thing, though, is that it was a wake-up call that there's been a big drop versus last year. That wake-up call was really set by cocking the trigger last year, and having an unusual increase last year. If you look at the trend over seven years, the one year that stands out is last year. Viewing was down every year except last year.
AA: So what do you think was going on last year?
Mr. Cook: Don't know yet. One of the things we're doing is to zero in on that, because it's a fact of life that we don't get as many questions about numbers being unusually up as unusually down.
AA: What about the viability of commercial ratings, which is the currency in the U.K.?
Mr. Marans: Well, first of all, how are we going to get reliable information on commercial ratings if we don't have a device, a mechanism that accurately reflects it? Have you ever kept a People Meter in your house? Have you ever seen how daunting the task is month after month? I don't know if they're going to be as keen on pushing the button in and out. I do remember somebody--I can't remember who--talking about fatigue many years ago in the Nielsen panel.
So, unless the mechanism for collecting the data is changed, we may be able to get, "commercial ratings," and we may be able to get, "marketing ratings," but I don't know how good they're going to be.
AA: So when do buyers and sellers start trading on commercial pod position and the like?
Mr. Poltrack: I don't think there's a buyer out there who basically doesn't try to get an "in program" first- or last-position, as opposed to a middle-of-the-pod position in a break.
Mr. Cook: Or even which break.
Mr. Poltrack: Right. These things are negotiated on a regular basis and the better buyers are getting favored treatment in that regard. So I think it's better if we are able to "dimensionalize" this to a certain extent.
Mr. Marans: What do you mean by the better buyers? Those who pay more or those who are somehow appreciative of the differences?
Mr. Poltrack: I don't think it necessarily has to be the people who pay more, I think it's people who have the information and have the best dialogue with the seller side. If you don't ask for it, you're not going to get it. The fact is, we rotate people through all of those spots evenly. But that doesn't mean if suddenly someone comes in and it's a deal-breaker that they have to be in a certain position; that they may not end up in that position. The good buyer is always going to do a little bit better than the bad buyer.
AA: Over the last few years there's been a lot of talk about optimizers, and how that's maybe changed the buying pattern. We've seen some money shift. But is there enough data that you're getting to justify a big shift to cable? Are you getting the respondent level data you need?
Mr. Marans: How much do we want to reveal during this discussion? [LAUGHTER] First, I'm delighted we have respondent level data. Second, I'm infuriated at the fact marketing data, like long-distance phone usage, car data, and such, are not on the tapes. Third, I'm very alarmed that we have to pay a lot of money for respondent level data. Data that's given without extra cost as part of the regular contract in most other countries. And because Nielsen was so intransigent about the extravagant cost for a while, it kept many of us from wanting to pay for it, and took a longer time for us to get it. So--respondent level data--wonderful.
AA: What about the optimizers?
Mr. Marans: There's lots of variations of optimizers and what we can do with them. Anything that can harness data a multitude of times more quickly than humans can is wonderful. But you don't turn on an optimizer with respondent level data and it turns into the tribute to Mr. Kubrick--"sorry, Dave, you can't do this, I control all." It's not that way at all.
AA: I was asking about their use with regard to money shifting to cable.
Ms. Lynch: It's just the same as the network stuff. What you need to be careful of is that you don't get a cable program spit out of an optimizer and go and expect to be able to buy that just because it's cheaper. That's not the way to use an optimizer.
Mr. Marans: Yes, many people thought that would be true.
Ms. Lynch: Maybe that's what people thought you could do, but that's not what you can do. What people are doing now is looking to supplement the basic optimizer data with other tools and research, such as the Quad analyses, and sales effectiveness studies, and recall-based studies--all those different kinds of things--based on their individual brand needs. We're trying to learn more about how advertising works, not how to necessarily buy and sell programs, or cable, or syndication. That's not the business we're in. With the broadcast and cable networks it's different. They're using optimizers as a different tool, for a different reason.
Mr. Poltrack: Yes, the No. 1 contribution of optimizers is they tell us everything we don't know. For example, is there a difference in the communications in a certain spot vs. another spot? When an optimizer spits out an intuitively illogical answer, then you start to ask questions, and then we realize how little we know about certain things that we should know a lot about. I'm not surprised that optimizers direct money towards cable TV. To a certain extent they're efficiency-driven. But most national advertisers are using cable already. If you took network TV, national syndication, and cable, about 20% of all the GRPs currently bought by national advertisers are in cable. That is probably the kind of mix you are going to see from an optimizer, if it had all the qualitative values put in effectively. Optimizers don't scare me from the point of view that all of a sudden "WWF" is going to get all my prime-time money. On the other hand, they tell me I'm going to need qualitative information, such as the Quad analysis, to support the value of "60 Minutes" and "Touched by an Angel" vs. "WWF."
AA: I want to bring up something else, looking to the future. Forrester Research has put out a report recently about personal video recorders. It says, "Personal video recorders are consumer devices that use an intelligent interface, an internal hard drive to visually record program participation on viewer demand. With features like instantaneous fast forward and reverse, the ability to pause, live broadcast, easier to create viewer profiles, PVR's will broaden viewer options by offering at any given moment a menu of recording programs based on viewer preferences. Today every major consumer electronics vendor has a PVR device in the lab . . . "And it goes on, "To accommodate the PVR audience . . . in the next 10 years these things are going to be practically ubiquitous. And to accommodate the PVR audience, the broadcast networks have three basic options: strike promotional deals with PVR vendors; dump their affiliates and go straight to cable; or increase the value of broadcast content." And now listen to this. "The shift to add free cable and satellite combined with PVR's ad-skipping capability will have serious impact on advertisers. Ad viewership will decline slowly at first, only about 8% over the next five years, but it will accelerate with PVR adoption. By the end of the next decade Forrester expects TV ad viewership to drop nearly 50%, forcing advertisers to accommodate this new environment. To reach the illusive PVR audience advertisers [will have] to develop more alluring campaigns, interactive ads that draw audiences into commercials, and highly targeted ad campaigns for niche audiences."
Mr. Marans: What's the date on it, April 1.
AA: Forrester is fairly well-regarded.
Mr. Poltrack: I've read that report.
AA: Well, Mr. Poltrack is ahead of all of us here.
Mr. Poltrack: I've also seen the devices. They were at NATPE. There's a business there, no question about it, and it will grow. I think Phil Guarascio [VP-general manager, marketing and advertising at General Motors Corp.'s North American Operations] was talking recently about the fact that, from an advertiser perspective, we're going to have to stop talking about counting eyeballs and talk about a broader concept of how advertising is communicated through mass media. Basically from our perspective as a mass medium, the attraction of that is that the advertising budget is a certain percentage of the marketing budget. To participate in that full marketing budget is something that we should be after. I think it's going to change advertising. It will change advertising, and it will change the way advertising is presented in mass media.
AA: Is it overly pessimistic when Forrester predicts that by the end of the next decade TV ad viewership will drop nearly 50%, forcing advertisers to accommodate new strategies?
Mr. Poltrack: As usual in the case of Forrester, they may have the curve a little bit too short. I think it'll take a lot longer than that.
It's a slower process, and the level of penetration may not be that great. However, advertising is going to be transformed. Advertising is not going to go away, but it's going to be transformed. It's going to be in people's homes, and we will collectively figure out how to present it in a reasonable way. Programming content has to be funded one way or the other. If the distribution system is there, and the collection system is there for programming to be funded entirely by subscriptions with no advertising support, then there may be a separation of entertainment from advertising. But simultaneously through the Internet the advertisers will be creating advertising entertainment.
Mr. Cook: I agree with what you're saying. The technologies that facilitate commercial avoidance are the same technologies that empower commercial seeking, and this should not be laughed at. But I think people want information about products and services, the Internet is proof of that. The TV medium will be able, with this kind of technology, to give advertising on demand, to give people the ability to learn more about products, and to see ads when they want to see them. There is still a need to reach people with information when they're not asking for it. And the TV medium is going to have to find ways to do that.
Mr. Marans: Twenty years ago HBO was going to be carrying every single best program. The next thing that was going to happen was everyone would do their own TV programming--VCRs would be ubiquitous--and usage of VCRs would skyrocket. And next, the Internet. If I believe what I saw in NetRatings, the average Internet user spends about two hours a week in cyberspace. So anyone who wants to be a seer and look out a little too far, I think makes a good deal of money as a consultant, and selling newsletters, and will seem foolish in retrospect most of the time.
Ms. Lynch: But by then they'll be rich and on some Caribbean island.
Mr. Marans: Right. So true.
Ms. Lynch: We've been able to avoid commercials since the remote control was invented. And people don't for the most part. Some of them do, some of them don't. A lot of people like watching commercials. So, it's not going to affect commercials on TV at all.
Mr. Poltrack: All these people underestimate, or perhaps don't understand, the actual way that TV's used. You walk into your home, you're relaxed, you turn on the TV set. You essentially aren't looking for program involvement, an interactive involvement.
I tape programs. My VCR is set to record certain programs on a regular basis. But I do not play those programs back with any degree of regularity. I like to know they're there. I like to know that if prime time is particularly bad on a particular night and there's nothing I can really watch, that I can turn on an episode of David Letterman that I couldn't stay up for the night before. But most times I don't do it.
Mr. Marans: I do the same thing. I'll trade you my warehouse of movies I've taped for your warehouse of programs
Mr. Cook: Personal video recorders, or intelligent agents, are potentially something that are going to change the way people use TV. Maybe the 10-year time frame is not overly optimistic if it provides a real benefit to people. It can change the way they use TV.
AA: So it will know, for example, that I watch "ER" every Thursday and then record it, for example, if I don't have it on, right? It would do that kind of a thing?
Mr. Cook: It'll record it and you'll have an "ER" channel available to you in the future, that has all the old "ERs" on it, or enough of them that you've got an "ER" to watch. But it isn't that you have to limit it to a program. It can be a certain type of program, or any program that discusses a certain subject, or any program that has a certain person in it, depending upon how much cross-indexing there is in the program listings. Most importantly, [it can be a potent tool] if it comes up so it is equal to any TV channel, and it doesn't require a separate playback of tape.
Mr. Marans: I can't wait to see the heavy viewers, who are already watching 50-60 hours a week, do this. I don't see it.
Mr. Poltrack: There's also the personal-interpersonal aspect of it that is relevant. It's going to be an interesting environment. Because there will be a significant amount of time-shift viewing. When VCRs came out one of the things that we argued was that they were going to bring movies into the home and compete with broadcast. The positive thing about them was through time-shifting they were going to allow people to watch more TV.
So if you are a fan of "NYPD Blue," and on a given week you find yourself watching the CBS movie at 9 p.m. on a Tuesday, and all of a sudden you realize it's 10 p.m. your "NYPD Blue" is on and you're into this movie, you've got a dilemma. But if "NYPD Blue," your favorite show, is in a stored capacity somewhere, then you can watch your movie and you can watch "NYPD Blue" later.
Mr. Marans: Your Choice TV.
Ms. Lynch: They'd go bankrupt.
Mr. Poltrack: So there will be a certain amount of that. Also there is the whole world of digital, multiplex TV, which represents a competitive threat to this particular product. If, in fact, you can do the same thing because the top programs are being multiplexed across the schedule, then the importance of this type of device becomes less. But one way or the other, increasingly when and where you watch TV programs, technology's going to make it a lot easier for you to set your own viewing agenda. And that will be good news for the major appointment TV viewing product, and it will be bad news for the least objectionable programming product.
AA: I'm going to change the subject. With the continuing fragmentation of viewing choices, and the corresponding ratings decline of the major broadcasters, there have been pressures to include more commercials. Then we have "make goods" as well as the other non-programming things that are already there. Where are we going with all that? As we've seen through the studies The Alliance is doing, this clutter problem's getting worse, not better.
Mr. Marans: We don't begrudge the networks making a profit. One of the real anomalies here is that most cable networks have dual revenue streams, and the broadcast networks don't. That's a political issue--it's also reality. If Lifetime, and A&E and ESPN are getting two revenue streams, they should have fewer commercials one can argue. That will make me popular in the cable community.
I've heard Marvin Goldsmith say, "hey, got to take care of the inventory and keep pricing down." All I know on this subject is the more clutter, the more it's hurting the advertiser.
But Marvin usually comes back and says, "okay, I will run less commercials if you pay me more money for those spots." But no one's willing to ante up for that.
Mr. Marans: That's true. Of course, one can also take this to its logical conclusion--why don't we just put on more commercials, and to hell with the programming?
Mr. Poltrack: If you look at the FM radio model, when FM started the model was to run less commercials. They basically said, "we're going to be better than AM because we're going to give you a clutter-free environment." I find it fascinating cable has done the reverse. Cable started with more commercials than network TV. They've created an overabundant supply. At a certain point someone's going to say it's worth money to be in a more clutter-free environment. And maybe there will be premiums attached. I mean "Hallmark Hall of Fame" really believes in that concept. Because "Hallmark Hall of Fame," buys the program, owns the two hours and chooses to run less commercials. Despite the fact that its commercials are very entertaining.
AA: And those shows are consistently highly rated.
Mr. Poltrack: Someone is going to start doing it, and I'm surprised that some entrepreneurial cable network hasn't come up with that idea.
Mr. Cook: I think the viewer doesn't mind the commercial load very much. And I infer that because you're looking if the increase in commercialization was lowering program ratings it wouldn't be worth it.
Mr. Poltrack: I agree with you. I don't think they mind . . .
Mr. Marans: But I'm worried about them remembering our commercial message.
AA: On that note we've run out of time, and I want to thank all of you for your participation.
Copyright May 1999, Crain Communications Inc.