The favoured method of measuring ad effectiveness online amounts to fraud. That’s the contention of the Financial Times’ global sales director Dominic Good, who, speaking at the IAA What’s Coming Next conference, said that the current system of selling ads on a cost-per-mille (CPM) basis, the performance of which is then measured by click-through-rates (CTR) is, at best, a proxy for an actual measurement of engagement with an ad:
“Intuitively we know that these ideas, borrowed from the direct response industry, and now nearly 20 years old, are fraud. These metrics are, and can be, gamed.
“Impressions themselves are really only a proxy for reach; unique users are surely the real measure of reach, yet rarely is that specified in media insertion orders.”
Those limitations of the old model are what spurred the FT to develop their cost-per-hour (CPH) ad model. Under the CPH system, only banner ads and middle-page-units (MPUs) which are in full view on an audience’s screen for five seconds are counted as ‘viewed’. Otherwise, the advertiser will not be charged. Taking as standard Google’s estimate that 50 percent of ads served worldwide aren’t actually viewed by humans, Good argues that the CPH model is a more honest approach to ad serving, one that guarantees “100 percent viewability”.
Beating banner blindness
In an interview with TheMediaBriefing following the press conference, Good clarified what he meant by 100 percent viewability:
“We use Chartbeat’s methodology, so as long as Chartbeat is counting it as viewed, i.e. in view, and we get the five seconds in view of active time, so scrolling, mouse-movements. That is what we’re counting.”
So while it doesn’t solve the overall issue of ad blocking or, presumably, any sufficiently determined ad-farm that wants to game the system, the CPH model is an attempt to alleviate the overall issue of viewability. Good suggests that, by focusing on brand metrics which have engagement over reach as their primary focus, the CPH method will better serve advertisers looking to engage with the FT’s executive-level audience:
“It’s all about quality. The whole premise of what we’re saying is that not all impressions are equal. Some work hard, some don’t. If you are interested in branding you need those impressions that are in view for a long time.
“You need that active engaged time between the user and what is on the screen. What we do is we produce quality content that has deep engagement with readers. That then creates a smaller quantity but much higher quality inventory.”
The initial pilot scheme – which saw partnerships with BP, Microsoft, IBM and Repsol and eight others – delivered positive results from the FT’s point of view, achieving Good’s goal of increasing the efficiency of the FT’s ad inventory by delivering the same amount of ad impressions with 20 percent less inventory.
Dominic Good speaking at the IAA’s What’s Coming Next conference, the launch of the FT’s CPH scheme.
Mobile traffic wasn’t included in the pilot, for technical reasons, and Good says that there’s no fixed time period for when that can be resolved. As mobile traffic accounts for over 50 percent of web traffic to the FT, that’s an issue that needs to be resolved. As Good says, they can measure around 50 percent of mobile traffic using the CPH metric, but that’s not a high enough proportion to deliver upon the promise of 100 percent viewability that’s at the heart of the new model.
Complementary rather than replacement
But Good is quick to note that CPH is best used for highly valuable audiences such as those enjoyed by the FT, and that there are plenty of reasons why publishers might want to stick with CPM as their default measure of digital ad supply:
“At the end of the day you still have to deliver your ad unit by an impression. As long as you’re delivering impressions you can buy on impressions.
“An impression still has value, particularly if you’re into broad-reach campaigns or campaigns which have simple creatives, simple executions, simple messages. You don’t necessarily need to have that length of time we’re talking about. We are disproportionately skewed towards brand advertisers.”
So the new CPH system is perhaps better used for publishers with a focus on brand metrics, targeting a habitual audience who make the site a regular destination. That’s supported by the FT’s implementation of ‘sequential advertising’, by which audience members who regularly visit the site are taken on a narrative journey through a series of ads for a single, hopefully further alleviating the issue of banner blindness. Good explains:
“We use brand metric studies to prove performance, and moving away from the outdated click. Our aim is to help the industry to grow by focusing on what matters to advertisers…that commodity is time.”
It’s early days for the FT’s CPH scheme, and despite the retention of some of its initial partners in the project there is a long way to go before the validity of its premise is proven. But as we see other publishers – notably the Guardian – moving more explicitly towards brand metrics in an attempt to reestablish the quality of their audiences and halt the race to mass audiences that caused CTR to become a default measure in the first place, the FT’s highly valuable audience is an excellent test-bed for a time-based ad system. If successful, the scheme could become the standard for paywalled sites like the FT, whose habitual audience makes a time-based system viable.