Is the ad market moving to a CPA model?

RWW publishes an interview with Highland Capital’s Richard De Silva.

Richard essentially makes the point that everytime there is an economic downturn, advertisers want better performance for their their ad spend. The first move was from CPM to CPC, the second, he claims, will be to CPA. Some

Then he goes on to say that to actually track acquisitions you need to put implants into humans, invalidating the first point. There are, of course, other ways. aQuantive, before getting gobbled up by Microsoft, did a lot of work correlating actual offline sales with online ad campaigns, compensating for the last click syndrome, and weighing in the value of various consumer interactions in the end result. At Cossette we’ve done that as well over the years, but many other opportunities exist.

One of the things that fascinates me is that in 2009, pick-up-in-store still hasn’t taken off. To me it seems like a killer app for all things retail, yet (especially in Canada) this is still almost non-existant. How often did it happen to you to stumble upon a brand/product online, check out the price on retailer’s site — and then go to store and buy it? Imagine if there was a coupon thrown in too, making it that much tempting. Coupons by themselves, of course, are another way of tracking offline sales. Granted, the overwhelming majority of conversions from brand advertising are more latent and my examples are borderline DR, but I’d still be interested to analyse pick-up-in-store transactions with respect to their users’ previous interactions with ads and the brand online.

This sort of initiative would help brand advertisers make forays into CPA-based advertising, but without it the economic incentive for publishers is simply not there. The new crop of behavioural and data exchanges (bk et al) will be of great help to facilitate the CPA model.

Let’s not forget that, if brand advertisers could measure their CPA themselves, they could already convert their CPM rates to their CPA metrics. And they could already negotiate down their CPM rates to match an acceptable cost per acquisition.

The problem is thus not with the compensation model, but strictly with the (im)possibility of measurement. And as long as that’s not resolved, publishers wouldn’t be able to readily accept CPA.

What we’ll see in this recession is not a move from CPC to CPA, but a drop in CPM rates and maybe a drop in CPC rates if things get really bad.