Why Econometrics Won't Save the 30-Second Spot

If you follow the predictions of WPP’s Sir Martin Sorrell, and many who follow advertising do, you might believe that econometrics will find the data to save the 30 second spot from oblivion.

But in this case, Sir Martin is jaw-boning a likely acquisition and looking to spin things in his favor.

The reality is that econometrics has been around a long time and has produced some very valuable models. Why hasn’t it already been applied to TV advertising?

There are several good reasons:

You need lots of data to build reliable models. All we have for TV is the buying data. Data about timing, impact, attitudes, behaviors, purchases, wholesale and retail prices, promotions, coupons, store locations and data to support correlations between spots running and products moving off the shelves is virtually non-existent.
-Econometric modeling costs a lot and requires sophisticated researchers. So far the attacks on TV haven’t been sufficiently threatening to warrant the level of cash outlay required.
There are too many related yet uncharted variables. Even the best models would be sophisticated guesses that would have to be validated over time by even more data (that we don’t yet have). That’s an awful long way at an awful high cost to be so easily dismissed.

Sooner or later advertisers will want to understand how and why TV ads impact sales. They will want measurements that assist them in making budget allocation decisions based on media effectiveness in generating immediate and short-term demand.

Just don’t bet on econometrics to provide the answers anytime soon.