Behind the Numbers: Making Last-Click Assumptions

Behind the Numbers: Making Last-Click AssumptionsThe purchase funnel may be clogged

You know the saying about assumptions -- they make an ... This is especially true in online advertising, where marketers and publishers have relied on the idea that the final ad a consumer sees before clicking the buy button is the most important marketing message of all.

But giving credit for a sale to the last ad, at the expense of the life cycle of the campaign, is a marketing mistake Microsoft's Atlas Institute wants to eradicate. Because the interest, intent to buy and final purchase occurs as part of a digital purchase funnel, says Esco Strong, senior group manager for the Atlas Institute, part of Microsoft's Atlas Solutions division, which provides digital media technologies and tools for agencies, publishers and advertisers.

The purchase funnel is a familiar concept to most marketers - those who know they need to build awareness and expose consumers to their messages over time. But most advertising metrics ignore the issue, or where and when the exposure occurred, and instead measure the conversion based solely on the last ad clicked on, Atlas said in its research report "The Long Road to Conversion: The Digital Purchase Funnel." The report measured more than 17 million conversions in campaigns run by 250 advertisers in the summer of 2008.

"Most information has been about the moment right before a purchase," Strong says. "This has been the gold standard for measuring ads, but it's wrong. You have much less accountability with offline media, newspaper and print, but the difference with online is you have accountability and measurability and we need to evolve this. So if I see 10 ads and half come from Yahoo, maybe Yahoo should get half the credit for the sale. The point is we can't learn a lot about our media if we only learn about the last ad. We know a lot of people spend a lot of time online and see a lot of ads. And you have to move someone through the purchase funnel and build awareness, consideration and purchase intent to close it."

Similarly, marketers should assign value for a sale to the Web sites and the ads that occur early on, midway through, and at the end of the purchase funnel.

Here's why: In its research, Atlas has found that in the final two days prior to a purchase a consumer is usually exposed to 5.5 ads. But, those final 48 hours before a buy only account for less than one-third of total media impressions, Atlas determined.

If advertisers back up the purchase funnel a bit, they'll see consumers were exposed to 8.5 ads in the week before a purchase, 11 ads in the two weeks before a purchase and 14 ads in the month prior to a purchase.

Strong suggest advertisers evaluate how a conversion was earned based on impressions, not just clicks. Measuring in this way also reflects the reality of media buying because buyers purchase ads on portals and news sites to get started on a campaign, then drill down to targeted sites for the next layer, and then close the deal with a search engine ad or ad network buy.

Measurement should reflect the reality, Strong says.

3 comments about "Behind the Numbers: Making Last-Click Assumptions".
Check to receive email when comments are posted.
  1. Mark Hughes from C3 Metrics, June 26, 2009 at 7:26 p.m.

    Compare Atlas to C3 Metrics...White Paper detailing cost savings using C3's Attribution Gateway. White Paper available for direct download from home page here:

  2. Steve Latham from Encore Media Metrics, June 30, 2009 at 1:52 a.m.

    Glad to see more articles on the subject of lead attribution. Now that we understand advertising matters, the million dollar question is this: what is the relative value of impressions vs. clicks? If a customer sees 8 display ads and clicks on 2 search ads before converting, is each "touch" worth the same amount? Or will there emerge a standard rule of thumb such as "1 click = 4 impressions"? Let the debate begin!
    Once this is settled we can look at that pesky issue of adjusting for cookie deletion and multiple machines :-) If you want to discuss visit

  3. Toby Korner from [x+1], July 7, 2009 at 5:46 p.m.

    Assumptions are exactly that, and we need to move away from them as an industry in order to push online ad budgets into the same sandbox as the traditional media. The best way to achieve this is through common metrics. Online ad measurement techniques and metrics need to be similar to more traditional media metrics, and not assumptive. There are piles and piles of information around online ads and purchase behavior (even more than for offline media) but it is not being used to its full potential. Looking at both online and offline through the same measurement lens is the key to decomposing the true purchase funnel.

Next story loading loading..