In digital advertising, there’s a tipping point on revenue and reader engagement. That tipping point can mean the difference in positive revenue outcomes at the sacrifice of engaged readers.
In a members-only INMA Webinar on Wednesday, participants learned how to stop choosing between revenue and readers, and to pivot to serving both.
Desmond Farrelly, head of audience and insights at DMG Media, and James Laver, strategic partnership executive at Marfeel, discussed how publishers can create a reader-focused monetisation system that will boost revenue while also improving reader engagement metrics.
Teams at Marfeel, a publisher platform that emphasises monetisation, often work with publishers trying to create a profitable monetisation strategy without harming the user experience, Laver said.
In the digital publishing landscape, the traditional TV advertising model is not very effective, Laver said — yet there are still many ways that publishers can monetise while improving the ad experience for users.
“Because users hate advertising so much, if we give them the tools to ad block, to get rid of them, they are.” A study by eMarketer found one-fourth of Internet users in the United States block ads. Another survey by Pagefair.com on audience response to Web site advertising found 60% of users were dissatisfied with the experience.
These are both indicators that publishers are not doing a very good job with Internet advertising, and they need to find out why and adopt a different strategy.
“I don’t think users actually hate advertising,” Laver said. “It’s just the majority of digital ads are annoying, they’re disruptive, they create privacy concerns because users don’t know where their data is going. They slow down the loading page, which we know users hate, and sometimes they’re inappropriate or not relevant to the users.”
A prime example of this is the use of prestitial ads, which Laver said destroys the user experience. In the short term, these can be valuable to publishers in terms of high eCPM and revenue — but in the long term, they drive users away.
“Since we know that users hate how we’re doing it at the moment, we need to find a solution that benefits both sides,” Laver said.
According to Hubspot, 83% of people agree with the statement, “Not all ads are bad, but I want to filter out the really obnoxious ones.” This tells Laver that publishers need to focus on quality, smart delivery, and the user experience rather than short-term revenue goals.
“Of course this leaves us with a big dilemma as publishers,” he said, which was the focus of the Webinar — how to solve the problem of revenue-bringing ads that don’t drive away customers.
Optimise quality, not quantity
Laver shared four key areas where publishers can increase ad revenue while improving user experience.
- Competition: Increase bid competition and reduce latency.
- Speed: Create frictionless user experiences with fast page loading.
- Quality: Premium, relevant, brand-safe content.
- Intelligence: Use data to measure user satisfaction.
Adding bidders is the best way to increase competition for ad space — 53% of publishers only work with between two and five ad networks. Doubling the amount of bidders can increase the CPM by around 60%.
This needs to be done in the right way, Laver added, which is why Marfeel suggests using server-side header bidding.
When it comes to speed, publishers must be fast or they will be ignored. Google now includes site speed in its ranking and reports that as page load time goes from:
- One second to three seconds, the probability of a bounce increases by 32%.
- Three seconds to five seconds, the probability of a bounce increases by 90%.
- Five seconds to six seconds, the probability of a bounce increases by 106%.
- Six seconds to 10 seconds, the probability of a bounce increases by 123%.
However, fast loading speed can result in the dreaded “white space” in which ads can struggle to load as fast. Marfeel found that every 10% increase in ad viewability resulted in a 15% increase in eCPM. Their solution was to prioritise the first ad impression and load it in parallel with the content to avoid blank spaces.
“What we found was that viewability shot up, they were seeing all of the content, they were seeing no blank spaces, and went hand-in-hand with the revenues,” Laver reported. “By just increasing the quality of availability, we were able to increase those revenues.”
In fact, the average viewability for the first position ad increased to over 60%.
The relevance of the ad alongside the content must also give value to the user. “That’s why at Marfeel, we only suggest working with premium networks.” These include Rubicon, OpenX, Google Ad Manager, AppNexus, Criteo, and Index Exchange.
Another benefit of working with the premium networks is they are always brand-safe. Working with better ad standards and being on-message is vital.
Publishers do not want to ever serve the ad formats that are the biggest culprits of user annoyance, interruption, slowed loading time, etc., Laver said. The Coalition for Better Ads includes among these the pop-up ad, prestitial ads, auto-play videos, etc. Though they might generate short-term revenue, the disruption means users will only see the ad once, then leave the page to find content elsewhere.
Intelligence: Better experiences keep readers returning
Once publishers have developed the right ad formats to engage audiences without disruption, they need to develop metrics to measure the results. Laver shared the metric that Marfeel uses and recommends.
Average Revenue Per User (ARPU) = Total revenue/Total number of users (in the thousands)
“The value of this metric is it allows you to see how every individual change affects every individual user,” Laver said.
He shared a case study of a Marfeel client that showed implementing the right ad strategy resulted in fewer ads but higher revenue.
Case study: Mobile-first strategy of Extra.ie
Farrelly lead INMA members through exactly how this process worked for Extra.ie. Started in 2017, Extra.ie is part of DMG Media, one of Ireland’s largest media organisations.
Extra is focused on creating smart, short-form, engaging content that holds the reader’s attention and creates potential for additional engagement. In the first year, the company had acquired one million users.
“But we came to the sobering realisation, one year later, that those users aren’t guaranteed,” Farrelly said. By the end of year two, Extra’s audience had decreased by 48%. “We essentially didn’t change our tactics. Everything we had done so well to acquire that audience, we continued to do.”
Those tactics of distributing and promoting Extra.ie content were primarily through Facebook — but as that platform changed its algorithm, the correlating users and page view numbers were declined. The tactics that were so successful in year one were no longer reaping the same rewards from the same output.
Difficult periods such as that can turn out to be the best thing that happens — a “wonderful misfortune,” Farrelly said. This experience forced Extra to do a complete audit of the content it produced on a monthly basis, cut waste content that wasn’t proving of interest to the target audience, and produce more content that was of interest.
“It led us to produce far less of what we thought was relevant to the user, and far more of what was actually relevant,” he said.
This led to Extra developing a content creation strategy it calls FLIP:
- First or fast.
After implementing the FLIP content creation strategy that consistently did well on Google AMP, Extra.ie achieved unprecedented growth of more than 252-fold increase. This led to a 164% increase in Extra’s audience by the end of year three.
Collaboration with Marfeel
Farrelly explained why the Extra team decided to collaborate with Marfeel. While they were very good at attracting users to the Web site through the unique mix of content, they recognised engagement levels could be improved.
They went into the partnership with very clear and strict goals:
- Improve the UX of Extra.ie (navigation and site speed).
- Improve the users’ ability to access all the great content their journalists were producing every day.
- Monetise more of Extra’s content.
The company immediately saw an increase in its Web site speed and in programmatic revenue — 78% in the first month, increasing to 190% by the end of the fourth month.
The collaboration also enabled Extra to have:
- Swipeable page and section headers.
- Article time indicators.
- Inline related articles.
- Sticky social bar.
Extra increased the year three audience increase by another 47%.
“We managed to increase our pages per user pre-Marfeel, but we’ve increased it again a further 38% in year four,” Farrelly added.
Before, Extra was serving about six commercial elements in a 300-word article (or one every 50 words); 50% of ads were being served in the first 33% of a page. Now, they have brought that down to three commercial elements in a 300-word story.
“We took a longer-form view on the whole thing,” Farrelly said. While they reduced the number of commercial elements within short articles, they actually increased the vertical distance between ads and added three new ad positions on longer articles.
“It meant the ads were far less intrusive. When you’re able to have the ads complement the content, there’s a far better relationship between reader and revenue.”
Ultimately, Extra.ie was able to decrease the amount of ads shown on shorter articles — but the revenue associated with those ads actually increased.
“This led to the highest revenue we’ve ever had,” Farrelly said. “We’ve just come through a period of the highest acquisitions we’ve ever had, around 2.2 million users [in Ireland], and worldwide 3.1 million users.”
This unprecedented level of growth has happened in a short amount of time, leading to Extra.ie becoming the new face of digital Irish news, sport, business, and entertainment.
“We’ve done that by doing what we said at the very start — putting the audience at the centre of our decision making. Through our collaboration with Marfeel, we’re able to monetise the content in a way we’ve never done before and upgrade our engagement levels,” Farrelly concluded.
Banner image courtesy of Pixabay.