Stampen Media increases programmatic revenue by US$4 million in 5 years

By Håkan Hamrin

Head of Programmatic

Stampen Media

Gothenburg, Sweden

Connect

At the beginning of 2020, Stampen Media set a budget for the programmatic display business at US$3 million. Now, at the end of 2024, revenue is at about US$7 million. This has happened in a world where the consumer business in media companies is taking on a greater focus, which, of course, affects the advertising business.

This development in revenue is the result of a lot of effort. How was it possible?

Stampen Media's programmatic monthly revenue in Swedish Kroner from October 2016 to October 2024. January 2020 is marked with a circle.

Implementation of header bidding

Most publishers today have implemented header bidding, but in 2020 it wasn't as obvious as it is today. Stampen had it in place and, therefore, we could add demand partners in a more efficient way than we could while working with a waterfall.

A crucial factor was the server-side header bidding our supplier provided. That made it possible for us to add far more bidders than would have been possible with only a client-side solution.

A simple explanation of the client-side and server-side options is that, with the client-side option, the auction is made in the user's browser. However, with the server-side option, this is made on the servers between the selling and buying part before the bid is sent to the browser. There was also an easy user interface that made it possible to add more bidders in an easy way.

Less friction on for the buyer

Our strategy was to make it easy for the buyer side to use whatever path it found easiest to access our sites. That could mean a certain sales company, a contextual actor, a bidder from an agency, or some other alternative.

Of course, we were aiming to get more buyers to bid on our inventory - this would mean we could both raise the fill rate and CPM (cost per thousand) levels.

Don't think we added anything; we said no to a lot of actors, and we removed several as well. However, we kept the ones showing good CPMs and good quality in business. They didn't necessarily bring the most revenue to the table, but as long as they sent us good CPMs for a certain kind of inventory, we believed they offered a high level of value for us.

Since we had the server-side solution, there was no problem keeping them as bidders even if the total revenue wasn't at the high level of our bidders.

We added several sales companies and bidders from agencies, supply-side platforms (SSPs), demand-side platforms (DSPs), and several other companies, which offered some kind of combination between ad tech and sales.

We reached quite high levels only by adding more partners and raising both the fill rate and sold CPMs. However, after a while we saw the inventory was setting limitations for us. We had more demand, but there was a lack of inventory to meet the demand. Since the consumer business was so important for us, it wasn't possible to easily add more inventory.

Converting readers into paying customers

We wanted the content we created to convert readers into paying customers. We also wanted the sites to be a good experience for our users. Adding more ads placements isn't that easy if you want to make the experience good for both ad buyers and readers. Additionally, viewability is crucial if you want to be a relevant partner for the buying side.

Then we stumbled upon Browsi. In a short period of time, we received mail from the company, saw a recommendation by an SSP in an ad-tech pod, and then met the team at an event in Stockholm.

Browsi was actually able to solve some of our problems. It could add more ad placements on our site without our tech being involved (and the internal tech is always a shortcoming, right?).

It could also make sure the placements would have high viewability (or the desired viewability, to be honest; we could have low viewability if we wanted to but that wouldn't be a very good idea). Most importantly, the company could make sure the user-experience (UX) metrics such as scroll depth, scroll velocity, and time on page was unchanged.

When we also found out the solution could be integrated through our header bidding wrapper without any effort from our own tech, there was no reason not to try the product. The result made us happy. Since then, we have gone from 1.8-2 ads per page to 3.3-3.5 ads per page with unchanged UX metrics.

New inventory

Because we had such a high level of demand, we could just fill that new inventory while it was growing.

However, even with these important improvements, we found the focus on consumer business leads to less inventory for the ad business. Yet, as I've previously noted, we started experimenting with putting fewer articles behind a paywall during the busiest days on the advertising side.

That was quite successful for the advertising revenue while there was no impact on the consumer business. Over the past few months, we have also seen a higher reach on our biggest site, Göteborgs-Posten. It's not only because of this change, of course, but this also shows us there hasn't been a negative side effect in regard to reach.

Other revenue generators

In addition to these fairly significant changes, there are lots of other things that have also helped generate revenue.

These include:

• Using Blockthrough, a new solution for reaching those with an ad blocker installed. This is also implemented through the header bidding wrapper.
• The reloading of ads every 30 seconds.
• Adding several formats to all placements. For example, all mobile placements are approved with formats that are 320px wide and 100px to 480px high.
• Using identity (ID) solutions. Several ID solutions are activated within the wrapper so we can get higher CPMs when the buyer is using those IDs.
• Adding first-party data from our data platform on deals and in some DSPs.
• Adding a pre-bid on native placements, and even letting display ads compete on these placements.

3 most important takeaways

In summary, there is not one single solution that has made it possible for us to increase the programmatic revenue. On the contrary, there are several different things that have made this possible. But, if we had to choose the three most important factors, those would be:

• Server-side header bidding, which makes it possible to add more demand.

• A solution to combine more ads with a retained user experience and high viewability.

• Experimenting with different shares of articles behind paywalls to meet the demand from the advertising side without harming the consumer business.

YOU MUST BE A MEMBER TO ACCESS THIS CONTENT

Become An INMA Member or Log In

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

About Håkan Hamrin

Håkan Hamrin is head of programmatic at Stampen Media in Gothenburg, Sweden. Håkan can be reached at [email protected] or @hhamrin.

Related Articles