In digital, we can measure so much. But so much of what we measure is completely meaningless. What does a click mean, other than that a click has happened? If you have an answer, is it the same answer as the next agency or client would give? Is it even the same for each campaign of yours? As an industry, we’ve tied ourselves to metrics that are very often unrelated to what an advertiser really cares about.  And that’s not how the ad “performed,” but how the audience was moved.

The problem is that the ad server that reports our metrics back to us is only good at measuring actions when what we need to measure is attitude. 

Some examples of popular but misleading metrics are:

• Impressions and eCPM 

The ad server tells us that an ad was delivered, but not whether it was seen. Many sites deliver fistfuls of ad impressions on each page.  We pay more for the larger ones and those above the scroll because we know that large and prominent has a better chance of being seen. But we don’t know that they are seen. This is further complicated by the ease with which sites can burn impressions no one sees – by placing them on the bottom of the page, by instigating many page refreshes during the content consumption experience,  and even by placing ad tags on redirect pages that never render on screens (we’ve really seen that).

• Views

The television model of interrupting content with advertising has jumped full force to the web. Ads are being delivered online in pre- and mid-roll formats that can’t be fast-forwarded… but that doesn’t mean people are watching.  In a piece of research we did in the US, we found that people take the “your content will start in 30 seconds” countdown as an invitation to do something else for 30 seconds.  “I open a new tab and look at something else, then when I hear my clip start I bounce back,” said one of our respondents.  

• Cost

When low cost is the measure of success you’ve failed before you start. Striving for low cost only means you’ve bought something cheaply. Here’s the news: it’s easy to buy cheap. But it’s hard to buy valuable, and it is in the nature of valuable not to be cheap.  Often an agency will buy valuable, pay its cost, and then buy a lot of cheap mass impression inventory to drive down the overall costs. Is it a great use of budget to waste money in order to look thrifty?

What matters is not the delivery of an impression but the making of one. And there are only a few ways to understand whether you’ve captured attention and moved an audience’s perception of a brand. One is to buy on cost-per-click or cost-per-engagement so that the risk of not getting attention rests with the publishers. This way, the only people you pay to reach are the ones who want to be reached. 

In a broad study of the effect of engagement, we learned that people who choose to engage with a display ad are four times more likely to recall the ad than people who were delivered a forced view of a pre-roll-like video product. And engagers were two to three times more likely to be able to repeat back brand attributes and significantly more likely to be moved to intention or likelihood to recommend than people who were merely exposed.

Regardless of chosen approach, it is important that the advertiser thinks critically about the metrics that apply to their core goals.  In an on-demand world, capturing attention is the single biggest challenge, so we must focus on understanding whether we have successfully done so and what the affect of gaining attention was on the viewer rather than on the ad server.