How The Financial Times tests new ad products to prove they work

“At The Financial Times, we’re moving the focus away from clicks because they are not the metrics we think are important,” said media director Lucy Marchington. “We are seeing an increase in volume of clients asking for more brand uplift studies, but also this is part of our focus on metrics that are more meaningful.”

Brand uplift metrics are having a resurgence as publishers and advertisers stop relying solely on clicks for performance. While publishers like Vice and Bloomberg have invested in brand surveys for proving that large-scale integrated branded content campaigns work, others like The Telegraph are offering them for display campaigns too. Google’s announcement it’s phasing out third-party cookies in Chrome by 2022, a key mechanism for digital attribution, has accelerated the trend.

The FT began looking at how it could understand the correlation of what causes brand uplift by carrying out brand surveys and cross-referencing them with metrics like ad-time-in-view and amount of screen real estate. After six months of testing with five different clients, The FT found that its vertical video ad format increased consideration by 21%, brand preference by 25% and purchase intent by 22%, compared to a control. One of the clients has re-booked so far and it won back one client that hadn’t spent with the FT for 18 months. The publisher wouldn’t share exactly how much revenue this has increased to or how the format performed in specific clients’ campaigns. It’s also winning more briefs especially from luxury clients who are drawn to the full-screen creative.

The publisher is halfway through testing a sticky ad unit in the right-hand rail and brand surveys embedded in content pages. It’s also re-auditing its existing ad products, like its time-based sales trading method. This week it finalized running brand surveys across mobile and desktop to develop cost-per-hour on mobile.

To move this forward, the FT is now testing how to overlay its first-party data when a user responds to a brand survey as part of a clients’ campaign, with brand lift measurement company Brandmetrics. Subscribers share data like job title, industry, area of responsibility and income and the publisher infers gender through the first name with 95% accuracy.

“We took a deeper look at our brand survey offering and how to develop it,” said Marchington. “It’s an important metric but we were reporting on it in isolation: This format offers X percent brand uplift versus the control group. Now we’re thinking about how to move that forward.”

The goal is to get more robust data this year to understand what formats and campaigns the FT can offer clients to hit specific brand metrics. Rather than the format delivering a higher percentage of brand awareness, it can cut the data by audience, for instance, the purchase intent is higher among the FT’s c-suite readers. This signals what behavior is a prerequisite to brand uplift, for instance, whether it was the fourth ad impression that led to higher purchase intent, or the time-in-view or an ad with higher screen real estate. From this, it will develop proxy metrics for the sales team who can say to get higher brand consideration these are the common behaviors that inform the campaign planning.

To make sure it didn’t frustrate paying readers last year it made the call to turn off full-screen ads in its app, while the ads delivered on campaign metrics they had a negative impact on subscriber engagement, according to Marchington.

“There’s this interesting balance with the advertising world and making sure that [the ad product] works for subscribers,” she said. “We don’t play in the scale or rates game. When you buy from the FT, we want to prove you’re getting high impact with the right audience.”