Commentary

The Ratings War

This week, the most significant currency in the media business, provided by Nielsen, lost its accreditation with the organization chartered by Congress to keep media ratings honest, the Media Ratings Council.

This post is an attempt to describe the chessboard, players, and possible outcomes regarding a set of moves that one publication called the biggest “measurement legitimacy crisis since 1964.”

First, the problem. Network TV is currently like a cheese shop with a broken scale. If advertisers decide the scale is not too broken, despite having been de-certified, it will make a mockery of the MRC, and the attempts at fairness it represents. The maker of the scale says “it’s only a teeny problem.” Meanwhile, buy a pound of cheese and they’ll cut you a hunk that might be a pound -- or not.

Now go spend a billion dollars on cheese and see if you can keep your job. 

At the center of the conflict is the historically tense relationship between media companies and Nielsen.  What’s driving the conflict is simple: Nielsen makes the measures that control buying, selling, accountability, planning, etc. -- the whole enchilada. Media companies don’t like a third party in their shorts, but buyers, for obvious reasons, don’t trust the supplier to tell them how much they've got.

advertisement

advertisement

Catalyzing the entire milieu, it looks like several media companies are getting ready to use alleged inaccuracies in Nielsen’s measures as a reason to build or buy their own measurement systems. They can probably build good metrics, but who will use them?

Nielsen’s dominance over TV ratings is more than half a century old, and it keeps hanging on. However, bruised by streaming TV, and beaten up by the internet, networks are at the breaking point. They need their numbers to look better, and they apparently believe that can happen if they move to a different currency. Plenty of alternative currencies are available (e.g., Video Amp), but Nielsen’s reaction is to double down.

It’s starting to look as if Nielsen is ready to challenge the entire concept of MRC accreditation, and test whether advertisers will continue to buy media using its ratings even without that accreditation. That would not be the end of the world. A great deal of media is transacted using unaccredited measures already.

And what do advertisers have to say about all this? Not much.

That factoid is fascinating. Either they don’t know what to say or they don’t know what this means to them -- or, possibly, there is nothing they can say that won’t get them in trouble.  Advertisers could force their media suppliers to use Nielsen’s ratings with a few phone calls, but that might not be a good move.

The strategic issue for advertisers is that is if they force a decision on ratings, they risk breaking a fragile system. For example, what if Unilever wants to buy Disney media using Nielsen ratings, and P&G wants Disney to use Video Amp ratings via Dentsu? The entire system gets less efficient in all phases. This might cost more than it would cost Nielsen to fix its problem.

There are, however, some interesting alternatives out there.

In most countries, a joint industry committee (JIC) decides who will provide the universal media currency. The currency provider is appointed.  That solves the problem of multiple currencies, and keeps the provider of currencies honest. The Advertising Research Foundation once suggested JICs for the U.S., and Nielsen’s response was to threaten a lawsuit.

That was then, this is now. The Association of National Advertisers could begin this move on behalf of advertisers. It might be time.

Regardless, this is our big chance to find out whether the marketplace values a common, universal measure over an accurate or a certified measure. 

Or maybe it just wants a cheap measure. If buyers and sellers continue to use the existing Nielsen GRP as currency, inaccuracies notwithstanding, it will confirm that common (or universal) is more essential than accurate.

Exacerbating an already confusing situation, Nielsen’s ratings were built to enable a marketplace, but the parties who inject 100% of the money into that marketplace (advertisers) have taken a hands-off approach, preferring to let the supply chain sort it out.

It’s arguable that letting suppliers spend billions of dollars on other suppliers who sell a diffuse and ephemeral product (consumers’ attention), which is counted by yet another company motivated to spend less on the counting, using a methodology approved by a tiny organization who outsources the analysis of the research … is a process that might benefit from a little hands-on investigation. 

You might think this is the part where industry associations swoop in for an easy win, but don’t bet on it. Progress will require a concerted effort from the entire industry, not just an investigation, or a manifesto. 

Who should own the process? You can bet advertisers want “the industry” to solve their problem, since they usually want that. It is not unreasonable to want that. It’s just that wanting a $100 billion system to change is unlikely to change it.

As always, asking a broken system to repair itself is an exercise in futility.

6 comments about "The Ratings War".
Check to receive email when comments are posted.
  1. Douglas Ferguson from College of Charleston, September 2, 2021 at 1:49 p.m.

    Sometimes at the gasoline pump, I pause to read the certification sticker from some agency in my state that promises me that the gallons I purchase and the price are in close alignment. I wonder what I would do if the sticker was removed?  Or if there were two conflicting stickers? Ratings are only estimates, as Arbitron was fond of saying before its demise. I am optimistic that there is too much money involved not to reach some sort of resolution to the Nielsen problem.  Maybe.

  2. Ed Papazian from Media Dynamics Inc, September 2, 2021 at 4:48 p.m.

    Good one, Ted. As most of us realize, chaos would ensue if each network or cable channel selected whatever rating supplier thay preferred---invariably one which makes them look the best---while time buyers trying to meet a brand's GRP requirements have to reconcile the inevitable discrepncies between the various sources. So a single, unified supplier---as Nielsen has been on a national basis since 1950 is a must.The question is how does Nielsen---as the most likely to win incumbent--- mix and match various sets of data---including set-top-box info which doesn't tell you who is watching---and device usage for digital venues---which  has the same issue---into a comprehensive single source rating service to provide the needed currency? That's not as simple a deal as the smartset and STB folks make it out to be. And panel size isn't the problem---its getting comparable data from what we assume is a representative panel that's at issue.

    As for the advertisers, they are not all of the same mind. A large number buy time on certain program types---sports,news, specials---despite their high CPMs and without much regard for rating erosion, demographcs, etc. simply because that's what they want to do. So they are very silent right now. The others hold back because they have relegated time buying to the "numbers people" and given the matter of rating accuracy little or no thought. As a result, the sellers ---who do most of the funding---will decide everything. And once they batter Nielsen into doing what they want, everything will be peaceful--again.

  3. John Grono from GAP Research, September 2, 2021 at 7:58 p.m.

    A very sage article Ted.

    I agree that the advertiser is key at this crossroads.

    Never before have 'publishers' (networks, streamers, Pay TV, internet sites) had the ability to see their own data in great depth.   They can see how many devices are currently downloading/streaming content at any point in time of the day.   There is however the temptation to conflate all that traffic data as audience data.   Traffic is device - audience is people.   Naturally, the publisher assumes their internal data to be the 'source of truth'.

    Around 20 years ago the internet was measured by "hits", and later "Unique Browsers"?   Surely the Unique Browser (UB) was the most accurate data!   In the dial-up days it may have been an acceptable proxy.   When broadband became prevalent, and devices more common, Monthly Unique Browsers sky-rocketed because of cookie deletion and accessing on multiple devices.   At one stage here in Australia Monthly UBs were 5x our population.   Unique Audience became the standard because the panel could de-duplicate the data.

    I think there is a parallel happening now.   As it is a seller’s market the focus has shifted to first-party 'vertical' data (i.e. a deep dive into usage patterns).   Should that 'vertical' data become some form of accepted currency, two things can happen.

    First, the seller can control their very deep 'vertical' data.   They can choose an IT vendor.  They can re-specify thresholds etc. to pump up their own tyres.   In short, the data would not be comparable between sellers.

    Second, the 'verticals' can't be melded together, they can only be added together.

    On the buyer’s side of the market, the media agency/advertiser need to see the duplication between the sellers, the programmes they sell, and the audience that watches the programmes they sell.

    Media agencies/advertisers primary goal is to provide the most cost-effective highest reaching campaign, with the least duplication of the paid ads.   No-one wants to see the same ad 100 times in a week.

    If the TV Ratings were based on 'Vendor Verticals' (of varying veracity) without third-party measured duplication (which needs to be done at a micro level and not a macro level) it would sadly be a retrograde step in my opinion.

  4. Ed Papazian from Media Dynamics Inc, September 2, 2021 at 8:28 p.m.

    John, one of the media planning issues that gets worked over a great deal concerns the issue of supposedly excessive frequency and the assumed need to reduce it. Too often, these discussions do not specify the time frame or consider that viewers don't pay any attention to a high proportion of their ad "exposures". Certainly it makes little sense to "expose" a consumer to exactly the identicalad message 40 times in a single week---even though in practical terms this really represents only 12-16 actual comercial viewings---not 40. But that's still too much. However 40 "exposures" spread out over three months---with the average viewer watching only 14 times---or just over once per week---and probably seeing several different executions from the brand's pool of commercials-----is quite a different matter as it goes to share of voice---relative to ad exposures by rival brands.

    Also, it is unlikely that TV time buyers will be de-duplicating audiences across platforms as they make buys---corporate upfront buys in particular where many of the shows that will carry the ads don't even exist as yet. And the planners, who might advise the buyers about the mix of platforms  for each brand are shut out of upfront, corporate futures buying deals. Until that changes---and I don't see it happening very soon----the main priority for a new and improved TV rating service will have to be getting a better handle on the average minute audience of each platform---in effect, the GRPs that will be guaranteed by the sellers.

  5. Joshua Chasin from VideoAmp replied, September 3, 2021 at 1:59 p.m.

    Ed:

    Assuming dollars are fixed, the question becomes, at what point is the value of an impression against someone exposed multiple times (frequency), worth less than a first impression to an incremental person (reach)? That's the point where impressions should be retargeted away from frequency to reach. And frankly, it's tough to avoid quickly building frequency against a core audience, because-- an inconvenient truth about media-- the top quintile, for the medium as a whole, or for a network-- may consume 60% of the quarter-hours (and thus impressions.) 

    But you're 100% right that you can't think of frequency absent a time frame. Is 10 exposures too many? In an hour, probably. In a month? Probably not. 

  6. Ed Papazian from Media Dynamics Inc, September 3, 2021 at 3:04 p.m.

    Interesting point, Josh. I happen to believe that the top quintile for a TV buy ---which gets about 55% of the "exposures" is probably more ad receptive to TV commercials as well as program content. If I am correct then it generates, maybe, 60-65% of the actual ad exposures---meaning looking at the brand's commercials.

    However I'm not so sure about the concept of incremental reach necessarily getting you better results in terms of ad impact as here too, the time frame is important. On a given day, you may be able to add incremental reach very easily but in a week it becomes  more difficult ---and costly--- and still worse monthly----and, in each case you don't really know if this is the first time in the ad campaign that the added consumer saw your message. I do know that many scanner panel studies for packaged goods brands have suggested that maintaining share of voice---or GRP tonnage---is very important---and this means that you need frequency not just reach to maintain market share.

    The key is how to maximize the time between your "exposures" while at the same time getting enough of them to fight off inroads made by rival brand ad campaigns. The longer the time span between exposures the less the consumer avoids the ads due to redundancy. This can be got at by clever scheduling---using lots of different dayparts, channels and shows, for example---but how many brands do this?

Next story loading loading..