Ratings, ratings everywhere... How much do they stink?

If you pay attention to the press covering advertising, you probably know about the latest brouhaha over Nielsen's attempts at switching over to their People Meters into Chicago, Los Angeles, New York and San Francisco. The people meters are meant to be a more accurate means of measuring audience than the diary system that has been in place since 1950.

Seems there are some flaws with the system and those who have the most to lose from these flaws are none too happy. Those who have the most to lose, of course, are media companies. The one that seems to have the most to lose of those is Fox, UPN, and Univision, because, according to reports in both the New York Times and The Los Angeles Times, the people meter system are undercounting - or altogether misrepresenting - minority household viewers.

According to an Ernst & Young audit of the system, the flaws aren't borne of an issue with the technology, though there has been some grousing about viewers being inadequately trained in its usage, but rather the problem is with the makeup of the sample audience that Nielsen has recruited to use the people meters.



Since the dawn of single-source ratings data, there have been skeptics, nay-sayers, and all around legitimate criticism of Nielsen. More often than not, the criticism has been around what the problem is this time, which is a lack of representative sampling of a population relative to the universe under surveillance.

What this demonstrates, in my view, is the continuing importance to all other media of being able to identify and buy advertising against actual audience.

My partner and compadre, Tom Hespos, wrote this week that perhaps planning and buying against media consumption is preferable to planning and buying against audience. With online, we deal in currency dependent and based upon direct measurement.

Though conceptually this should leave little room for doubt and error about what is it we plan, buy, and reconcile in the way of media, funny that there is a healthy amount of quibbling over the verification of media consumption units (impressions) as there is now with the ostensibly more accurate audience measurements Nielsen is trying to make. Of course, with television, the stakes (read: money) is higher, so the debate has been turned towards a more political charge with the introduction of media race politics (long a tender spot in the media industry and long an ignored one).

But when all is said and done, advertisers need to be concerned not by just how much media is consumed, as a whole, or how many (and what kinds) of people are consuming it. It is both of these concepts that need to be married at the altar of media measurement. And, in a way, that is what ratings - a marriage of reach and frequency - try to simulate. Advertisers want to spend the least amount of money in spreading their value proposition to maximize their income and profits. Reach is how many people the advertising speaks to, and frequency is how often they are spoken to; basically, how loudly do you speak and how often do you say what you are saying.

The good thing about online's present use of its version if media currency is that it is ostensibly 100% verifiable. The downside is that it does not currently track to people for the purposes of reconciliation, which isn't all that great, seeing as how the point of advertising is ultimately to sell goods, and, last time I checked, impressions don't buy things, people do.

So, if someone can just figure out how to track people... Oh, wait. We can. The technical strengths of the web allow us to do just that.

Let's avoid the ugly political back-biting television is going through right now and move towards doing the right thing.

Next story loading loading..