Imagine opening up an app, scrolling through an article, being interrupted by a full-size Clash of Clans video, trying to tap a microscopic “×” to close out of the video, and accidentally tapping into the app store. For regular smartphone phone users, this scenario’s all too common.

More user-friendly mobile ads haven’t caught up to the vast potential for monetizing growing audiences on mobile.

“It’s a matter of investment. For some agencies, mobile advertising hasn’t gotten too much of a budget yet in terms of the creative side,” said Staffan Engström, head of mobile at Scandinavian publishing group Schibsted, said. “It’s a big bottleneck to crack the code for successful mobile advertising. As soon as we see bigger budgets coming through, I think the effective results that we so far haven’t seen, what’s been holding the industry back — all will change.”

But what can news readers stomach when it comes to a big disruption on a relatively small screen? Moreover, what do they prefer, and how much should agencies spend on well-made mobile ads?

Schibsted removed some of this guesswork in 2015 by surveying more than 37,000 mobile visitors to its news outlet Aftonbladet about their responses to various types of ad campaigns they saw for products ranging from fast food to luxury cars. Aftonbladet is Sweden’s largest news source, its mobile website reaching more than 5.9 million people per week in a country of 9.8 million. (In Sweden the advertising trade group KIA-Index records web traffic by week.) It has a predictably shrinking print circulation and declining revenues overall despite digital — mobile — gains, trends represented by many other news outlets within the Schibsted group. (Headquartered in Norway, Schibsted is also the parent company to Norway’s leading newspaper Aftenposten and the tabloid Verdens Gang.)

Aftonbladet mobile users saw over the course of a week from ten major advertisers, from Mercedes to Burger King, in three formats: a static image, interactive rich media, and video, for a total of 30 different possible campaigns, each eventually seen approximately 400,000 times. Users then received surveys based on the variety of ad campaigns they ultimately saw. The work here was carried out in partnership with Lund University in Sweden with graduate students Gustav Sundberg and Anton Olivensjö.

The study revealed that the most effective of these formats was the static image. Static banner ads had the greatest effect on a reader’s preference for a brand and intent to purchase the item advertised — and those are among the categories that matter most to advertisers who, at the end of the day, want to sell something.

Readers rarely engaged with rich media or the videos, which are fairly disruptive, and only 0.3 percent ever played the video ad to begin with (“Video is good at building awareness, but this is a very, very, very low number if you want people to actually get your message,” said Engstrom). Rich media ads were a turnoff: The survey revealed a negative correlation to effectiveness of the campaign after just two exposures to the ad. Static image campaigns won out again here, their effectiveness increasing pretty significantly with exposure.

Engström hopes the Schibsted mobile study results will help move his industry away from clickthrough rate — the ratio of people who actually click on an ad, out of all the people who see it — as proof of an ad’s success or failure. Among those surveyed, CTR was negatively correlated with an ad’s effectiveness.

Consider the possibilities for a publisher that decreases reliance on clicks as a measure of success and decreases the number of ad campaigns built only to capture clicks (or taps, whether intentional or stray). Schibsted is not alone here. Facebook, too, for instance, has been building up a new measure for marketers that it’s calling “conversion lift,” finding click-counting insufficient in a world where people access sites from multiple devices. The Financial Times has rolled out a new measure called “cost per hour,” which takes the amount of time that an ad has been seen into account.

Advertisers and the agencies that help create the ads are listening to Schibsted’s findings, said Engström, whose team works across all the company’s brands. For mobile, he recommends specifically short, six to eight-second videos — never 30-second videos, which resemble television spots. Vertical video, though, is a relatively uncharted and potentially dangerous territory, as takeover ads are the most intrusive format for readers.

“We also see a lot of advertisers starting to talk to their ad agencies about this, though it’s still a slow process,” he said. “But we do see far fewer 30-second videos being sent to us after our work.”

“What we’re learning, too, is that the main message of the video should be in the beginning,” he added. “It’s not like, in 30 seconds you build up some kind of feeling throughout, then get to the point at the end. You need to know exactly what you’re going to look at. There is no teaser. You have to have the main message up front. You need to be very clear what you’re going to talk about.”

Engström hopes his team’s work might help Schibsted bypass the mobile adblocking apocalypse altogether — an apocalypse that hasn’t yet arrived, but looms on everyone’s minds.

“Mobile adblocking hasn’t become a problem for us — it’s under 1 percent. Most of our traffic comes on mobile apps. And for our mobile website, we haven’t seen any development at all for adblockers,” Engström said. (For desktop, the overall adblocking rate at Schibsted is closer to 20 percent.) “We’re more worried about creating a really good ad experience so we don’t get into that problem.”

“Good” very often means “relevant.” And as a publisher, Schibsted is “very much in the blind” about useful user demographics like age and income level for its sites without a required login, Engström said, though the company plans to work on a system, sometime in the coming year, to better keep information on its audience by building on data collected from users of the company’s financial services sites, which request more personal information and require a login.

The company also continues to fine-tune its geotargeting capabilities.

“We’ve been using geofencing, creating virtual fences based on where people have been,” said Estelle Douglas, who works on mobile advertising at Schibsted. “For instance, everyone who has been at a soccer stadium watching a game: We can build a profile around them and then, at a later stage, reach out with an ad when they visit one of our sites.”

In one instance, the team helped a large Swedish supermarket chain personalize ads for each city that has one of its stores. It also helped a telecom company shape ads for each of the small local markets it targets. “The results were good,” Douglas said, “mainly because of the relevance of the message.”

Schibsted is also working on a “brand builder” system that makes sure readers aren’t repeatedly shown the same ad.

“Advertisers can’t just create two or three ads and expect them to run for months. You will bore people to death,” Engström said. “The mobile device is so driven by your mindset — what you do at certain times of day — and advertisers need to follow that kind of cycle.”