Four years ago, the discussions were all about either/or. Should Aftenposten, Schibsted and Norway’s largest subscription newspaper, choose a metered model or a freemium model? There were clear advantages and disadvantages with both options.

Step one: Deciding whether to use a metered or freemium model

The metered model maintains traffic levels and revenue opportunities for digital advertising sales, and it does not require ongoing discussions in the editorial department about which articles should be placed behind the paywall or kept open. The freemium model illustrates the value of being a paying subscriber. It provides unlimited traffic to open content and leaves no question as to what content you need to pay to read.

At the time, in 2013, that was how we saw the digital model.

For the launch of Aftenposten’s digital subscription, the metered model was chosen. At that time, advertising revenues were the main part of the total revenue at 60%. All print subscribers were automatically given full access to Aftenposten Digital.

The next step was to motivate users to log in. The data from logged-in users provided opportunities to keep a dialogue going with users during the relationship and learn more about their reading habits.

However, digital conversions did not occur at the expected speed. With 20 free articles per device per month, only 9% of the users actually hit the paywall and a very small percentage of these actually purchased subscriptions.

The metered model worked well to maintain pageviews and ad impressions, but it did not show the value of paying for content or how to buy subscriptions. For 91% of the readers, the experience of Aftenposten’s Web edition was unchanged because they never saw the sales poster.

Step two: Introducing the hybrid model

During autumn 2014 and winter 2015, it became obvious our current approach would not get us anywhere close to the number of digital subscribers we needed to substantially grow our user revenues. Should we give in or try a different approach? We chose the latter.

Over the summer of 2015, Aftenposten launched what we call the hybrid model: a metered model combined with some freemium content.

We launched our hybrid with our most premium content as the starting point. Every Friday, Aftenposten’s weekend magazine (A-magasinet) is published both in print and on digital.

A-magasinet is Norway’s largest newspaper magazine with engaging stories about essential topics and exciting people, as well as good guides for active leisure, food, and drink. The content is attractive and well-read. Therefore, with the intention of clearly showing the value of a paid subscription and driving more users to hit the paywall, the content of A-magasinet was published behind the paywall for subscribers only.

All other content was still on the metered model. Aftenposten now had a hybrid model with the content of A-magasinet on freemium and the rest of the content behind the meter. The change had an immediate effect.

The increase in sales intake gave Aftenposten a substantial boost in the conversion of digital subscribers, but parallel to launching the hybrid it was natural to explore and experiment with new sales surfaces and sales promotion slots. Data and systematic work processes were key for all of our work and initiatives.

Step three: A tighter hybrid

From 2015 onward, our focus has been on identifying new ways to boost conversions without reducing overall traffic to our site and apps. Among other things, we have increased the share of freemium content on our front page from 10-15% to 20-25% and introduced a tighter meter where we reduced the number of free articles per user/device from eight to six per week. These adjustments helped boost conversions further and is a key factor behind our growth.

Over the last four years, we have worked systematically to improve all parts of the user experienced and our conversion flow. We have identified new and optimised existing conversion points, introduced new ad formats, learned how to convert from Facebook, improved our purchase flow, and started to use propensity scoring and data triggers to promote the right offer to the right user.

Initially we started with the basics and tested three different sales entries in addition to sales posters and conversion from the A-magasinet articles that were considered freemium content.

The sales button in the header is a clear entry that is always displayed in the headline of Aftenposten’s digital products. The appearance and communication is systematically tested with change of text, colour, shape of button, and so on. Data still gives us the answers as to what converts best for that space.

Additionally, sales posters in the editorial environment can be optimised to communicate more of the content the reader already consumes when he or she hits the sales poster. By using data, we test the position, appearance, communication, and content of the poster.

Also, if you visit Aftenposten as a non-subscriber today you will get several teasers before you actually hit the paywall. These can be ads promoting stories or “reminders” after reading two or four articles saying you are running out of free stories to read.

With this teaser of the sales poster, the hypothesis was that the reader might convert earlier. The reader may have aspirations to read more and therefore choose to buy a subscription. A negative consequence we discovered was that some readers choose to screen the reading to save the quota of free articles before they hit the sales poster on the meter.

With the hybrid model and the aspiration to increase sales surfaces and continuously optimise, test, and personalise, Aftenposten is growing digital sales fast and reached a milestone this November. This rapid experimentation has taken Aftenposten from 30,000 to 100,000 paying digital subscribers in two years.

The whole organisation is working continually toward the same goal. In upcoming blog posts, my colleagues and I hope to share more details about how we work to secure digital sales and engagement. The goal is to inspire you and to start dialogues across borders.

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