• by Tom Hespos , November 3, 2003

Sometimes I feel as if some of the work our industry does to isolate the effects of online advertising in integrated campaigns is like the never-ending pursuit of the Grand Unification Theory.

Advertisers and agency folks alike constantly ask themselves the question, "What does online do for my bottom line?" when the question they should be asking is, "What combinations of media are best for my bottom line?" They want to understand, in a quantifiable way, which media are responsible for exactly how many sales in a given period of time. All media should be held accountable by precise measurement, but let's not pretend that sales should be attributed to any one medium in an integrated campaign.

On iMedia Connection this week, Peter Greb of Avenue A showcased a method for differentiating between causal and incidental non-click conversions in an online campaign. The methodology seems sound for an advertiser that fulfills purchases through a single channel (the web). But most clients don't fulfill solely through the web channel. They might have retail stores, or resellers, or an offline catalog.

But questions remain: How do we credit the online channel when customers who see web ads buy offline? Why does online get credit for an e-commerce sale when TV, radio, print, PR and other offline channels impressions contributed to the end sale?

The interesting idea presented in Greb's article notwithstanding, this notion of hyperaccountability is exactly the type of thinking that lands ad agencies in hot water. We try our best to analyze individual media in a vacuum. The good agencies warn their clients that learning from this type of analysis is "directional only" and caveat the living hell out of their findings. The bad agencies attribute sales to individual media when it turns out that all they're measuring is the last communication that a prospect was exposed to before becoming a customer.

But the truly sad thing is that many clients eat this stuff up. They look at click and conversion reports from adservers and say to themselves, "Online is responsible for X sales." By and large, they have no idea how online ads contributed to offline sales, nor are they paying attention to how their offline media contribute to online sales. Drilling ever deeper into the Excel spreadsheets to find pockets of media activity that "work" or "pay out" often results in "analysis paralysis" where both client and agency are spending so much time buried in minutiae that they lose track of the big picture.

I'm not suggesting that we revert back to the advertising Stone Age and fail to hold communications media accountable. But I think the marketing industry needs a healthy dose of realism if they want to avoid analysis paralysis.