If you managed to get through middle school math (I’m hopeful that means most of you), you’re familiar with the term “Lowest Common Denominator.” In math it’s a way to combine unlike fractions by finding a common ground. In business, it’s a way to screw yourself up. You see, there’s another nonmathematic use of LCD and it refers to the lowest or least sophisticated level of something, and that’s the subject of today’s screed.
As anyone who has worked in broadcasting will tell you, the ratings system is a sort of shared myth. Nielsen puts out numbers, TV executives believe them and TV buyers believe the TV executives. Of course, it says right on the front of the ratings book that they’re only accurate up to a point, and like any number based on a sample the results are really a range. That range can be pretty wide as the number of folks in the sample who did something declines (so the published rating for American Idol is probably closer to the truth than the rating for a show ranked 125).
Which is why I find this disturbing:
TubeMogul is bringing Online Campaign Ratings to its RTB video ad platform. The agreement between TubeMogul and Nielsen means advertisers and agency trading desks can cross-reference GRPs for audience age and gender demographics with impressions and clicks to get a fuller sense of a campaign’s performance.
Simple announcement which a number of folks covered. Except, of course, when one reads further:
While TubeMogul is able to relay metrics like impressions and clicks in real-time, Nielsen’s GRP numbers are only available daily, as with their broadcast GRP metrics. Also TubeMogul’s advertisers will have to log in to the Nielsen dashboard separately to view GRP numbers alongside metrics on TubeMogul’s platform.
In other words, we’re bringing down digital’s great system of non-sampled measurement to the LCD of TV. That’s bad business in my book. I realize that the advertising ecosystem isn’t quite able yet to deal with a completely different set of metrics, especially metrics presented in real-time, but the further we dumb down the standards the more likely it is that those lower standards become the norm instead of temporary fixes.
Digital measurement isn’t perfect. Faulty implementations, disreputable folks cheating via bots and other ways, and an overwhleming amount of data we don’t often present well are issues. But even with these and other faults the reporting and accuracy is better than what we used in TV, which any TV or agency person will tell you is pretty much a fantasy if you get them talking over a drink.
We can do better!