Is that any way to make an impression?

"Click fraud" is something we all know to be on guard against, thanks to investigation and reporting that’s extended even into mainstream news.  Based on recent research, we’d encourage advertisers and agencies to be on the lookout for impressions, too, that aren’t 100% genuine.  And we’d urge publishers to work on eliminating "˜bad’ impressions before they create a new round of headaches for all the people who are trying to make an honest living in the online media space.

One content format that enables publishers to place an especially heavy thumb on the scale and artificially generate impression volume — intentionally or unintentionally — is the slideshow.  Local publisher boston.com uses it in a way that isn’t terribly objectionable, as in this cute photo feature about "best places to kiss" in Boston.  Look at these three pages and notice what changes about the ad.

Nothing changes, right?  Well, not exactly.  Each time a reader clicks to the next picture, the ad is reloaded and the ad-tracking tag is fired again.

In this case, there’s probably no cause for outrage.  Each ad impression happened only because the user stayed with the content, actively requesting the next "˜page.’  It’s also possible that the advertiser paid for exclusive share of voice, in which case they wouldn’t be paying for each new slide, anyway.  Nevertheless, the impressions generated by the slideshow can pose measurement challenges, since it can generate high impression frequencies but still constitute just a single, long exposure from the reader’s perspective.  If these impressions are lumped into aggregate campaign reach and performance data, they could cause confusion and lead to bad judgments.

But if confusion were all we had to worry about, we probably wouldn’t be writing this.  Many other sites have implemented slideshows in a way that is almost guaranteed to generate valueless ad impressions, without the advertiser’s awareness.  Take the following excerpts from a Parade.com photo-feature on Parade All-American high school football players:

As static screenshots it doesn’t look any different from Boston.com’s slideshow.  But it’s significantly different when seen live: the slides cycle automatically, even when another tab or window is opened on top of it.  A consumer would have to scroll down to find the pause button.  And the slides change every ten seconds — many of the ads we saw didn’t even get to finish their animation.  In the last month, at least one of our anonymous panelists was exposed 50 times to the same display ad in less than10 minutes.

Other facts need to be considered before passing final judgment.  How much did the advertiser pay?  What did they expect to get?  But we don’t need any more information to give some advice to advertisers who are paying or measuring by the impression:  Know what you’re getting into if you advertise on sites with a flair for slideshows.  If a network buy means you won’t know where your ads will be running, then do your best to inspect your ad serving data for impressions that look as though they’ve been served on autopilot.