Tracking Non-Click Conversions

We’ve all known for a long time that banner ads have an effect on users who don’t click on them. And one of the ways we theorized that we might measure some of this effect is to count the users who see a banner then later return to a site without clicking on the banner.

The best method for measuring “non-click conversions” seemed simple in concept. Simply drop a cookie on each user that receives your banner ad. That cookie can then be matched up with itself, if and when the user comes to a page on your site that’s been pixel-tagged by your adserver.

Seemingly, this is an easy way to track all conversions back to the ads that referred them, whether those ads were clicked on or not. It would make several new effectiveness metrics accessible to media planners and would give us a better idea of online ad efficacy.

Here’s the problem: This method produces a lot of data. Think about it. Every cookie request has to be logged, whether it’s from the banners or from the pixel tags. On a large site with a large banner campaign, this could amount to terabytes of data.

This data then has to be crunched by machines that can match up all those cookies, which is a tough job if you want to track users who return to the site up to a week after being exposed to a banner ad. And the data processing job becomes exponentially tougher as you extend that window of time from a week to a month or beyond.

Due to the data warehousing and processing demands, many third-party ad management companies were reluctant to track ad effectiveness this way. However, a couple of the ad management companies are now offering this type of tracking, having evidently addressed these two problems.

Whether or not this is due to the “Great Ad Management Consolidation” of a couple months ago, we’ll never know. What’s important is that it’s being offered now, and there’s no need for online advertisers to invest in their own distributed networks, adservers, and data processing machines in order to make this type of tracking possible.

Won’t it be terrific to be able to see conversion rates broken out by frequency of exposure? In my mind, such a measure would be a more accurate dimension of online advertising effectiveness than the metrics we’ve been using, which have only addressed the immediate response to banner advertising.

Personally, I think we’re on the verge of a complete shift in the way we look at banner advertising effectiveness. I would urge all the other media folks out there to take a long, hard look at this method of tracking effectiveness, as well as at the new effectiveness metrics that it makes possible. I think we’re about to see in black and white what the IAB Branding Study has been telling us for a while now – that the Internet is an effective medium for increasing brand awareness and recall.

Now that tracking non-click conversions has become easier and more technologically sound, I think all online advertisers owe it to themselves to take advantage of it. Banner advertising has been selling itself short for a while now, and tracking all the conversions attributable to banner ads is a better way to measure effectiveness.