ellipsis flag icon-blogicon-check icon-comments icon-email icon-error icon-facebook icon-follow-comment icon-googleicon-hamburger icon-imedia-blog icon-imediaicon-instagramicon-left-arrow icon-linked-in icon-linked icon-linkedin icon-multi-page-view icon-person icon-print icon-right-arrow icon-save icon-searchicon-share-arrow icon-single-page-view icon-tag icon-twitter icon-unfollow icon-upload icon-valid icon-video-play icon-views icon-website icon-youtubelogo-imedia-white logo-imedia logo-mediaWhite review-star thumbs_down thumbs_up

3 outdated metrics you're still using

3 outdated metrics you're still using Tony Zito

Recent conversations regarding the "Making Measurement Make Sense" (3MS) initiative indicate that advertisers recognize the shortcomings of campaigns focused on impressions or post-impression attribution, and with campaigns optimized for clicks. In explaining the genesis of the 3MS initiative, the Interactive Advertising Bureau's (IAB) website states that "while consumers have embraced digital media and continue to adopt new media behaviors, the marketing and media businesses have yet to create the tools necessary to keep up with consumers' behavioral changes."

Traditional measurement tools, such as impression, post-impression attribution, and click-through rate (CTR), handicap marketing managers and force upon them campaign measurment tactics that should have long been abandoned.

Impression targets

Optimizing a performance campaign to serve a specific number of impressions is the worst way to get substantial results from display media. While impressions are an important component of campaign strategy, companies that base campaign success on the number of ads served are solely focused on increasing ad volume, not conversions. Theoretically, more ads served mean more exposure, which should return more results. This is simply not reality.

RealVu, an advertising analytics company, recently published a report evaluating "ad impressions that appear within the viewable computer screen area" versus ads that appear in non-viewable areas. Their research exposed that "viewable impressions were never more than 56 percent and could be as low as a shocking 6 percent on an individual campaign basis." Another report from AdXpose shows that "more than 50 percent of the ad impressions delivered, and 95 percent of clicks, came from suspected fraudulent sources," and that "nearly all the fraudulent traffic was hidden behind numerous layers of nested I-Frames (ad units that pull ad content from other sources) that can purposely hide URLs and in-view data."

As is evidenced by these reports, networks or other vendors focused strictly on the volume of impression delivery may design campaigns to purchase inexpensive and often non-viewable ad inventory to maximize the number of impressions they can serve within their budget.

Serving ads to users is not the same as creating an "impression" or generating a "view," making both of these terms industry misnomers. Most ads do not show up in a viewable area. Even in cases when advertisers are fortunate enough to place an ad where users can be exposed to it, there is another barrier commonly recognized as "banner blindness." This phenomenon is a testament to the value of dynamic ad content and the need for evidence that ads are influencing consumer behavior. Even brand marketers can benefit from the added value of knowing which impressions were loaded and viewed.

If as much as 94 percent of ad inventory is invisible to consumers, then we can easily explain why the remaining 6 percent of viewable ad space is so costly -- it's scarce. So, marketing managers are faced with a dilemma: sacrifice reach for inventory quality, or access a larger portion of the target audience and guess about campaign efficacy.

Post-impression (view-through) attribution

If advertisers have ever run display-ads with a performance goal, then they've likely paid for post-impression conversions, or at least attributed a portion of post-impression conversions to a campaign. In this case, advertisers were likely charged for consumer behavior that was not influenced by an ad, given that ads may not have been seen at all. All the same, too many advertisers allow partners to get away with attributing too many post-impression conversions to their banner ad campaigns.

Attributing post-impression activity is especially troublesome for retargeted banner buys. Retargeted audiences have already expressed interest in the advertiser's product/service/brand by visiting the website, and, as a result, have an existing propensity to return and convert. Serving ads to site visitors after they've already visited and abandoned (left without purchase or conversion) is a highly effective methodology, but care should be taken in attributing perceived subsequent success.

Imagine yourself as a yield manager for a retargeting vendor that was getting credit for post-impression revenue. Your greatest margin would come from a strategy that allowed you to deliver low-cost inventory to all abandoners (cookie users having been served an ad), and then claim responsibility for a large number of return visitors/conversions that would have happened even without the campaign. Most early retargeting companies are only successful because advertisers are still willing to give credit where it's not due.


Knowing that impression volume and post-impression attribution give little insight into banner ad success, clicks and CTR have become a crutch for those hoping to understand campaign performance. Historically, marketers haven't had anything else to reference.

If asked if they ever click on banner ads, almost everyone will say "no." In fact, according to comScore/Starcom research, just 16 percent of internet users would answer "yes," and half of those (only 8 percent of all users) generate 85 percent of all clicks. Very few people click on ads, and those that do, do so often. So, campaigns optimized for CTR only succeed by narrowing the targetable audience and serving ads to users that have a higher propensity to click. This is highly effective at increasing CTR, but will return deceptively impressive results for those marketing managers who think clicks are the ultimate indicators of campaign performance. The question that marketing managers must ask themselves is, "Is my now narrowed audience of likely clickers the group I'd really like to reach, and does this strategy yield the best possible outcome?"

The answer is no.

Because typical users don't click on banner ads, advertisers forfeit a large pool of opportunity by optimizing for CTR. For ecommerce advertisers, sacrificing profitable incremental revenue for marginal improvement of CTR is unnecessary. Not all buyers are clickers and not all clickers are buyers -- in fact, most of them are not.

In mid-2011, Collective Media released findings from their study of more than 100 million users and more than 1 billion impressions, which found:

  • 99 percent of users never clicked ads

  • Users that did click were 210 percent more likely to click again

  • 18 percent of clicks came from ad impressions that had already been clicked

  • Clickers tend to be older, lower-income users who are less likely to spend money online

  • Users who are economizing click 65 percent more often than those that purchase frequently online

In ecommerce, and certainly in the case of retargeting, clicks have become a standard measurement of direct user responsiveness to an ad. While it's true that clicks represent an immediate movement from the ad to the website, they do not demonstrate consumer purchase intent. Furthermore, by optimizing display media exclusively for improved CTR, advertisers forfeit profitable incremental revenue and brand lift.


The organizations behind 3MS (IAB, ANA, and 4AS) are making a clear call for higher quality performance metrics. What's surprising is that engagement has, so far, been overlooked. As advertisers consider new standards for understanding display ad influence, it's important that they recognize ad engagement, and subsequent post-engagement conversions, as a viable metric that will help improve optimization and attribution techniques.

Tony Zito is CEO of mediaFORGE Inc.

On Twitter? Follow iMedia Connection at @iMediaTweet.

Tony oversees Rakuten Marketing in the Americas, Europe and Australia, leading the development and execution of the company’s vision. As CEO, Tony drives the global organization in delivering consumer-centric omni experience marketing...

View full biography


to leave comments.

Commenter: Lisa Tadje

2012, January 27

Thanks for your comment, Josh! I work with Tony at mediaFORGE. The primary reason we raise this issue is to clarify that marketers don't have to accept this as reality for their display media campaigns. There is a better option. Check out our full whitepaper at http://www.mediaforge.com/resources to read more about engagement and why we believe it's the best optimization metric for delivering strong and verifiable campaign performance. You can also read our follow up article that iMedia posted this evening.

Commenter: Josh Sherwood

2012, January 26

I understand that retargeting view through can be somewhat misleading, but that same argument can be made for other top converting mediums. Paid Search top converters are typically branded keywords, meaning the consumer already has the brand in mind. Email sends are to someone that has already visited the website and signed up. Social is similar as well in that typically someone already uses the brand then likes them (not saying a lot of conversions come from Facebook).