AdAge highlighted Google’s Campaign Monitor, a display ad measurement tool which has been in beta for the last year:

With Campaign Insights, Google takes data from the advertiser’s server logs to determine who was shown an ad and when. Then compares that to web searches and site visits culled from data from the millions of Google toolbars on computer desktops. Those results are compared to a comparable group that didn’t see the ad.

Then Google measures the difference between the number of brand searches and site visits between the two groups. To filter out the impact of other media or influences, such as a TV campaign, Google compares the data to the two groups’ behavior before the campaign began. The incremental difference is attributable to the display-ad campaign.

So you have to give Google your server logs, and in return you get Google massaged data. Yet another form of market transparency through opacity.

Off the start I am sure it will be fairly accurate, but 3 years from now when…

• the market is more mature
• Google introduces quality scores and other such levers
• Google is trying hard to hit the quarterly numbers

…I am not sure how much it would make sense to trust Google for these kinds of measurements. Of course opinion won’t matter if Google is the only game in town. It would benefit Microsoft and Yahoo! to push hard on this front because whoever gets the data partnerships locked up first will end up with a more efficient market and higher yields…eating market-share from other ad networks.

If they are honest with the data it might be a bit of an up hill fight. Even within the search channel the idea of attribution is iffy. Andrew Goodman wrote:

Problem: 74% of purchases can only be attributed to a single click anyway. By definition, that’s the last click. Attempt to “attribute” that purchase to prior clicks or other factors? Good luck.

…

So the next revelation: when you bucket keyword queries down into distinct types, only 7% of sales can attribute significant influence to some kind of click other than the last click.

And large advertisers, growing more aware of that lack of search buying funnel, are starting to spend more on longtail search keywords while avoiding bidding wars on top keywords, as reported in the WSJ:

Sprint is buying the top ads tied to phrases consumers tend to search for when they are close to making a purchase, such as “cellphone rate plans” and specific products like “Samsung Reclaim,” rather than more generic phrases they search for at the beginning of the shopping process, like “Sprint,” “AT&T” and “cellphone.” Pricing for the more-generic terms tends to be higher, yet less for important to driving sales, Mr. McPhillips says.

Of course if lots of big advertisers follow the same strategies they may leave holes in some markets. In a recent interview, researcher Jim Jansen claimed just the opposite…that the broad keywords were cheaper and provided more bang for the buck:

In terms of classifying queries in terms of what advertisers’ payoff is, I think the most interesting finding was that the purchase queries, the last stage of the buying funnel, were the most expensive and had no higher payoff than the awareness or the very broad, relatively cheaper queries. From talking to practitioners, that is a phenomena that they have noted also… which is why a lot of people bid still on very broad terms, to snatch these potential customers at an early stage.

My  SEO experience and PPC experience has been more inline with seeing great ROI on the longtail keywords. But with SEO for some really long-tail cases the page can be set up to pull in traffic that is so niched that it becomes harder to find relevant advertisements and offers to match up against the traffic stream. And of course the issue with longtail PPC is getting much volume.