Bearing The Bull Or Baring the Bull: A Need For Set Top Box Straight Talk

The set top box realm is clicking. I don't mean the zillions of streams daily. Rather every couple of weeks for the last few months another company - whether researcher, dataminer, technologist, manufacturer, and/or vested third party such as advertisers, advertising agencies and content providers - is jumping into the fray, or what one might have perceived months ago, frolicking with consensual strange bedfellows.

Some recent examples (chronologic order):

  • Set top box and video on demand measurement company Rentrak has announced that it will combine its STB data with data provider approval segmentation systems - both syndicated and those from advertisers that they use in their day to day business processes.

  • Middlewarer OpenTV unveiled plans to open up its TV measurement platform by publishing the format of key data points that its software is able to measure on the set top box to allow the developer community, data analysis partners and other service providers to integrate with its measurement solutions more easily and help foster the creation of next generation set top box services. Partners include AdsVantage, Nielsen, Rentrak and TNS.



  • Technologist NDS and TNS Media Research launching an opt-in audience measurement initiative that will track viewing patterns of individuals.

  • The formation of The Council For Innovative Media Measurement (CIMM), a new coalition whose members include advertisers, advertising agencies and content providers (AT&T, NBCU, Viacom, CBS, Disney, ESPN, WPP, Interpublic, Omnicom, Starcom, MediaVest, P&G and Unilever) whose mission is to develop set top measurement and cross platform measurement as well as provide guidance to start-up technologists and proffer introductions to industry players.

  • Digital Video Recorder TiVo's launch into a commercial ratings service (second by second) for three local markets.

  • The unveiling of Microsoft's Navic Reveal, a census based near real time measurement and reporting of TV program viewership.

  • Interactive TV program guide maker Rovi, formerly known as Macrovision, has joined forces with media measurement company TNS Media Research, to analyze TV viewer actions, including use of IPGs and reactions to IPG advertising.

    The list of players invested or investing in this burgeoning arena is becoming legion: set top box clickstream data-ists (Navic Reveal, Nielsen Media Research, Rentrak, TiVo, TNS Media Research), cross industry collaboratives (Collaborative Alliance Set Top Box Think Tank, Council for Innovative Media Measurement, the Nielsen funded Council for Research Excellence), data driven TV buying platforms (Google TV Ads, Invidi, Microsoft's Navic Admira), dataminers (Acxiom, Allant, Experian), exotic flavored data analyzers (TiVo/Quantcast, TRA), predictive modelers (AdsVantage, Google TV Ads, Microsoft's Navic Admira), and technologists/manufacturers (NDS, OpenTV, Rovi and rumored BigBand and Motorola).

    A renaissance of sorts for audience measurement.

    My concern: as the industry mandarins squabble for hegemony within this expansive, fertile space - one that is necessary to prove the efficacy of TV advertising as well as evolve cross media televisual metrics (linear TV, broadband, mobile and all varieties of on demand) - they will promulgate both the promise and perils of set top box data in ways that obfuscate and mislead the community. Case in point: in late summer, a major research entity published some analytics comparing program viewership in digital cable homes versus satellite only households versus NPM sampling and the effect it would have on revenue generation. "Some examples from the '08-'09 season:

  • Cable networks would do much better in digital cable homes - with some networks getting a lift of 20+% in audiences
  • The Fox broadcast network would do 4% better with digital cable only homes while CBS broadcast network would lose 6% of its audience
  • Ratings for "Desperate Housewives" would be 12% higher with digital cable only homes but 6.5% lower with Satellite only homes - that is a swing of 18.5% for a single show
  • Cable networks would benefit from using digital cable homes only - to the tune of $2.5 billion in additional ad revenue
  • If C3 ratings estimates were based upon viewing only from digital cable homes, it would cost the broadcast networks approximately $340 million in ad revenue so far this season."

    Is this sampled reporting suggesting that when pay TV operators approach potential customers and are able to sell through a subscription, the platform insists on a loyalty oath that prohibits viewing of certain channels by their subscriber - whether they be broadcast, cable, satellite or telco originated. Or more simply, that certain subscribers have a propensity to view certain types of programming given their channel availability, packages (analog, digital, double or triple play or quadruple bypass), premium services, broadband connectivity (speed or not), number and age of occupants in the households, household income, location (rural, suburban, urban), adoption of new technology (DVRs, high definition), interactive features... And these factors, coupled with contential desire, determine what they will watch, not the platform itself.

    Applied mathematics whose applications are self serving will only hinder our understanding of the value of click stream data, the marriage of panel based metrics with STB analytics and myriad of other possible measurement combinations to bring more targeted value to the advertising community and greater satisfaction to the consumer.

  • 10 comments about "Bearing The Bull Or Baring the Bull: A Need For Set Top Box Straight Talk".
    Check to receive email when comments are posted.
    1. Mike Einstein from the Brothers Einstein, October 6, 2009 at 12:12 p.m.

      With all due respect to "the list of players invested or investing in this burgeoning arena", brands are made or broken on the strength of our feelings about them, not on some intermediary's specious interpretation of data about us.

      Pity major brands like GM for instance, who, when reading articles like this, can't help but conclude that their agencies know more about who isn't buying cars than they do about their good clients who make them. Worse yet, these same poor guys are picking up the tab.

      Closer to home, the vicarious thrill we get from Mad Men should tell us something about how far we've strayed.

      Can you imagine a TV show about the Coalition for Innovative Media Measurement? Neither can I.

    2. John Grono from GAP Research, October 6, 2009 at 4:14 p.m.

      Excellent post Mitch. I liken this situation to the online world over the past decade where every man and his dog came up with the new "wonder metric", truth be known each serving to be little more than a sales tool rather than an audience metric.

      The need for clear measurement guidelines is imperative and that need is NOW. The key is not in the statistical gymnastics that one system can perform better than another but in actually understanding the 'humanistic' side of what the raw data streams actually mean so that accurate pictures of the audience can be drawn. As we've discussed before Mitch, the real illumination will occur when panel-data is successfully fused with STB data in a 'hybrid system' - may the best man win!

      Now ... Mr. Einstein. With all due respect. Mitch was not talking about brands being made or broken. Mitch was talking about audience measurement. These two topics are world's apart. Whether you like it or not (and clearly you don't), robust quantification of audiences at both the macro and micro level is imperative in our industry. Just as creating brilliant advertising is in order to convert mere products into powerful brands.

      The interpretation of such data as Mitch talks about is no less specious than the creation of ads that "really taps into the heart and mind of the consumer" - a specious (and unquantifiable) phrase I have heard in many a pitch, made directly to the client who are paying the truckloads for such "trust me ... I know how the consumer thinks and can tap into their emotions and sell you loads of product". I suppose there is more money to be made though when you keep the shroud of mystery draped over the creative process though.

      But you are right - I can't imagine a series about CIMM (though maybe it would make a good documentary). I just didn't simply realise that whether you could spin-off a TV show was a criteria for success in an advertising campaign.

    3. Brian Rock from Network Ten, October 6, 2009 at 7:05 p.m.

      It's interesting seeing progress on fusing STB data with panel data, although some of the early results raise as more questions than they answer.

      For example the data suggests that "the Fox broadcast network would do 4% better with digital cable only homes while CBS broadcast network would lose 6% of its audience".

      OK - why?

      What is it about digital cable homes that makes them more likely to watch Fox and less likely to watch CBS? Or is this begging the question: perhaps this variation is a due data processing and reporting (this is pure conjecture: I have no reason to suspect this is true, but it is possible).

      Same again with "ratings for "Desperate Housewives" would be 12% higher with digital cable only homes but 6.5% lower with Satellite only homes". What's the explanation for such a dramatic difference?

      I'm sure these sorts of issues will be explored in more detail, and maybe they are being looked out but not reported. But the sooner the analysis is done and publicised the sooner we can move on to more accurate audience measures.

    4. John Grono from GAP Research, October 6, 2009 at 7:48 p.m.

      Good questions Brian. One of the reasons could be a lack of consistency in definitions, guidelines, data cleaning, processing, attibution etc. That is at the heart of what Mitch is after and what I am after.

      There is another reason of course. STB data is normally from a 'slice' of the market - that subscription TV operator. I think it is a vanity to expect one subscriber base to mirror the population - unless of course they have massive penetration. Therefore I think that what the data SHOULD be saying is that "within subscriber base X, programme Y performs better that the general population".

      For a proper hybrid fusion you would collect ALL STB data sources (and what do we do if not everyone wants to play) and then fuse that with a nationally (and market and subscriber platform) representative panel. This makes it imperative that all STB data passes through the same 'engine'.

    5. Mike Einstein from the Brothers Einstein, October 7, 2009 at 10:56 a.m.


      You not only miss my point entirely, you prove it in your reply. Listen to yourself. You're not in the advertising business at all, you're a data broker consumed with chasing the stock price down, to wit my brother Jeff's take on how far we haven't come and where we'll never go by making the medium the message:

      "As a society we seem all too willing to swap (mostly meaningless) information for (meaningful) communication. We have chosen to deploy our digital communications tools to shut down communications in the same way we choose to devote all of our time to our time-saving devices, and in the same way we now choose -- at every opportunity -- to eliminate friction by automating the sales pipeline. In the process, we choose to exclude meaningful deliberation and dialog, we choose to waste our time and attention (our only true inventory) on trivial pursuits, and we choose to destroy our own profit margins.

      These are idiotic and self-defeating choices, and unnecessary concessions to the great narcotic of our age: digital scale -- the exact same addiction that produced three trillion-dollar market crashes in the same decade, and the exact same addiction that is now destroying the media industry and reducing it to a minimum-wage beggar's market. Our fealty to digital scale -- like all addictions -- warps our thinking, champions the trivial, replaces sober perspective and judgement with narcissistic narcosis and illusions of grandeur, consumes all of our time and resources, then leaves us with little or nothing at the end of the day.

      Rather than looking for additional ways to reduce friction in our lives, we should be looking for new ways to induce and provoke it, because the jury is in, and there is no value to be found in a friction-free pipeline. The moment meaningful dialog is deemed too inconvenient to sustain is the moment we sacrifice our humanity on the alter of sheer arrogance. Babel is the only logical conclusion."

    6. John Grono from GAP Research, October 7, 2009 at 5:55 p.m.

      Thanks for that career assessment Michael.

      However, you're extremely wide of the mark saying I'm "not in the advertising business at all". I can only surmise that you reached that erroneous conclusion because my moniker includes the word "Research". Surely, just because I no longer work in either a creative agency or a media agency (and I have worked in both) but am now an independent research consultant to the industry does not mean that I am not in the advertising business.

      I reiterate that Mitch's piece wasn't about the creation of communication ideas, which is your obvious and laudable passion (and by the way, I agree with your concerns around poor advertising). It was about measuring the size of the audience that are exposed (and ideally engaged, persuaded, act upon) to those communication vehicles. Just because clients are demanding that the medium be measured does not mean that the medium becomes the message as you seem to be purporting was the drive behind Mitch's post and my comments.

      As for your brother Jeff's rodomontade piece, I'm so very happy for you that you quoted it as it is obviously very dear to your heart. I just can't see the relevance to Mitch's piece about the need for standards in the use of STB data for audience measurement purposes.

    7. Mike Einstein from the Brothers Einstein, October 8, 2009 at 11:15 a.m.


      Let me give you an analogy of what I mean when I suggest that our industry's obsession with audience measurement is misplaced reactionary foolishness:

      In the movie, Play it again Sam, there's a scene where Woody Allen, preparing for a date, expresses to his Humphrey Bogartesque alter ego his concern that he perhaps has applied too much after shave. To which Bogart replies: "Somewhere you got turned around, kid. She's suppossed to smell good for you!"

      In a similar vein, we somehow got turned around to where we now place the onus to perform on the media. The media doesn't perform. It can't. It just sits there until we figure out what to do with it.

      And I would further suggest that your claim that "clients are demanding that the medium be measured" speaks more to what you and yours can defend and sell rather than to any clear mandate from the advertiser.

      You're a very gifted writer, but your eloquence notwithstanding, at end of the day it still comes down to the fact that Starkist doesn't want tunas with good taste, they want tunas that taste good.

    8. John Grono from GAP Research, October 8, 2009 at 9:12 p.m.

      An interesting analogy Michael. Miind you, my favourite line was when Woody turns to Diane and says, "you've got the most eyes I've ever seen" (omitting the word "beautiful" obviously). But that could be the quantitative researcher coming out in me!

      Let me explain my point of view and where it derives from. When I started back n research in 1977 (owww!) I was hearing calls from clients for greater depth and accuracy in research across all facets of marketing. Better measurement of sales, of promotions, of distribution, of advertising ... it was quite an overwhelming list to a young buck. When I moved into media and audience measurement research in the late '80s the call was even louder and even more precise.

      My point is, that this is not a recent phenomenon. It is also not something that I am trying to "defend". It is a long-term need that as a research industry we're trying to deliver on. And may I say ... it's extremely complex multivariate work. Maybe the mandate you talk about has not existed in the US for as long as it has down here in Australia. Being a small population we often have to do things differently on small budgets in order to deliver.

      With all due respect, I am surprised that you think that any calls for greater accountability are seen as "misplaced reactionary foolishness". I'm not sure who originally coined the phrase "that which can't be measured can't be managed", but that seems to be the mantra of all marketers (rightly or wrongly - and I tend to agree that is right). At the end of the day marketers either directly or indirectly pay our wages. I for one have listened to their call, and am working to deliver on their need.

      One area I am very concerned about regarding measurement in marketing is with the 'effectiveness' of the creative idea or its execution. Past research efforts at delving into this sphere have, in my opinion, been misguided because the industry has relied upon written questionnaires asking respondents to enunciate feelings that are occuring in the emotional or non-verbal parts of the brain. There are some interesting developments in neuro-science, but then all they are reporting is that there is some recation to the idea or execution, there may even be an attribution to the neurological reaction, but we still don't know how or why that reaction happened.


    9. Mike Einstein from the Brothers Einstein, October 9, 2009 at 10:57 a.m.


      I must say I am enjoying this exchange.

      There are any number of viewpoints - some more famous than others - that conclude that the closer we observe something, the more it changes. Does the object or event being observed change? Or is it the observer who changes?

      In Mitch's article above, he states the effect of our confusion without citing the cause, which I contend is a basic refusal to acknowledge this uncertainty principle at work. Put another way, we feel compelled to further refine our metrics because what we're measuring keeps changing. And it will continue to change the more we measure it. Better tools don't and can't change the principle. They just help us chase more string.

      But I think our over emphasis on media metrics is revealing of another, more disturbing malaise, namely our reluctance to risk our ideas in the open marketplace. We can hide behind numbers and research, but we're loath, indeed afraid to raise our hands and say "Here's what I think."

      Advertising should be about influencing future thought and action, not about interpreting past behavior. That's why I tossed Charlie Tuna into my last comment. My guess is you haven't heard that line in years, but it resonates today as strongly as ever.

      Go ahead and study the audience all you want in the rearview mirror, but don't be surprised when you discover that not only is the road behind you different than you thought, you know practically nothing about the road ahead.

      Better yet, take a creative risk by challenging the status quo instead of analysing it. You and your colleagues - and your clients - may just find yourselves concluding that you don't really want an audience with good taste. You just want an audience that tastes good.

    10. John Grono from GAP Research, October 9, 2009 at 8:26 p.m.

      Me too Michael!

      And not surprisingly we have different takes on the same thing. Let me try to explain what I mean by audience measurement. It measures what has recently happened. It has no pretence to say that is what will happen in the future.

      As an example, let's say CSI has drawn an audience of 15m-16m people each week for the past four weeks, then I am pretty confident that next week it will do the same sort of numbers (ceteris paribus) this week, and on this basis I make a decision whether to run a ad in it or not based on the "fit" of the ad to the programme and how much the networks are charging for the spot. Clearly, it would be sheer madness to assume that it would do the same numbers in a year's time.

      Importantly, this enumeration is purely a quantification of the mass or people who watched the programme. It does not pretend to understand their motivations, engagement, liking, disliking etc for the programme. However, if we see a big bunch of people returning to watch the programme week after week, do we really need to know why? Frankly, there are probably 15m reasons why. All I need to know is that they do, and I can pretty safely bet the client's money on it. Think of it as "wisdom of the crowds".

      I concur with your opinion that advertising should be about influencing future thought and action. However, I believe that interpreting past behaviour is integral (client-side) as well. A good marketer should be looking at their sales and saying ... well look at that, everytime I run humorous ads on Letterman twice a week I get a spike in sales that isn't there when I don't. The marketer first of all needs to know that the spike occurs. Then we can dig around and find out why.

      In no way does any of this impinge (or should impinge) on the creative process. Believe it or not, there IS some science behind marketing. In a past life I have built micro-econometric non-linear multivariate sales models for clients ... with scary predictive ability. Where these models fall down is with what I call "the power" of individual ads. For example, I may know that a burst of TV spikes sales - but by different amounts. After taking out pricing, distribution, competitive, seasonal effects etc (all pretty easily quantifiable) there is still this variability ... the "power" of the ad. Face it some ads are good and resonate, some don't. No, I have not cracked how to "measure the power" because frankly I don't think it can be done. The closest I have come was to get the creatives, strategic planners and client to rate the ... gold, silver and bronze (though we originally used ... shit hot, pretty good, and could-have-done-better).

      Now to throw another malaise into the ring. That is "extreme futurism". Every day I read another article about "the next big thing". Futurists are perfectly safe with such opinions as you can't measure the future - neither should you try (remember Henry Ford's "faster horse"). The hype and excitement these things generate (remember Second Life anyone) I believe divert way too many people in the industry in to looking for the "pot of gold" instead of single-mindedly doing what they should be excelling at ... making blindingly good creative advertising campaigns. (Then I'll come along and tell them how they went ... hehehe).


    Next story loading loading..