The company then turned to Rouven Leuener who, as group head of digital product, has pushed the newspaper forward creating a series of innovative new products.
At DIS2019 Rouven will examine those innovations in detail. Here he explains why the company chose to focus on its audio offering and why its paywalls are among the most technically advanced in the world.
How did you end up in your current role?
During my studies I founded a Studentlab, a web technology startup that is still in existence. Then I went to Swiss media house Ringier, where I was responsible for all types of digital strategy. Among other things I created the biggest sports mobile community in Switzerland and was behind the development of the company’s video offering.
After that I went to Neue Zürcher Zeitung (NZZ) with the task of reordering everything – in five years. In that time I have created six new digital products. At the same time I was also studying in London for a Business Master in Digital Media Management and Innovation with Hyper Island.
NZZ has pioneered an innovative text to speech system which went fully live a few weeks ago. How does it work and why did you choose to create it?
At NZZ we are gradually becoming a user centric company, and that, in my understanding, means you cooperate with the user and take the lead from them. We listened to our audience and some of them said “I really like your content, but it isn’t enough. And at the end of day when I go jogging or I am in the car I would love the opportunity to listen to it.”
And we took what they said very seriously and tried to find out how we could offer voice products. We published every article we created as an audio file and soon discovered that it was too expensive to do it manually. So we developed the text to speech engine that takes the content and converts it. We started developing the programme last year and we fully launched it a month ago. It has been successful. It is something people love and now we’re rolling it out to all our customers.
If you listen to the content on a smart speaker, for example, you have to connect via the app, so we have a good idea of people’s listening habits but it is too early yet to have any meaningful insights. We do however see that people are very engaged with the audio. We initially did a silent launch and now we are seeing more and more people use the product as we encourage our audience to try it out.
What is the business model for the product?
The business model is based around our dynamic paywall – within 4-20 articles the paywall is breached and you are asked to pay for the content. As we are currently collecting data from the product we are also thinking about other business models too. Maybe, for example, a listener could have 10 minutes free and then pay £1 to listen for an extra ten minutes, or something similar.
At the end of December we will make a decision about which is the best business model for the product. Fortunately we have had lot of feedback, including lists of things users would like to change. For example, they asked for a female voice and they wanted more variation of the speed. These are a couple of the things we’re going to change. Also, at the moment we don’t offer streaming – the user has to download the whole MP3. This is going to change next month when we will have new features ready to roll out to everyone.
NZZ has also developed a personalised news app – how successful has that been?
This was an innovation, as we had need for personalisation both on the user side and the company side. We developed our own system because we were not very impressed with how others did personalisation. You know the ‘you get more of X because you have Z’ – etc which to us didn’t make sense. We wanted our personalisation products to be different and developed in heavy corporation with the users.
We have a personalised newsletter. This is really important for us and it generates more engagement and helps us tailor the content for the user. The feedback has been very good. Initially in the beta phase we got a lot of love from users who are passionate about personalisation. We are however, careful about how much we personalise. For example, we would never personalise the front page. Users also need to know which content is served to them editorially and which is personalised. Other companies offer different types of personalisation and in some instances boast very good click through rates. But we don’t care about click through rates. We want people to visit us frequently and highly value us when they engage with us.
What other innovations have you been working on?
At the moment we have just finished our homework, we were catching up! Five years ago there was nothing and the business had no future profile. Where we are really good is the innovation around our dynamic paywall. We calculate the propensity of how likely to pay for content every user is every night, and then combine this data with machine learning to promote our paid for content. This is unique, though the FT has a similar model. In the new year we will open up a second horizon type of project. However there’s no point in us using AI or ML just for the sake of it. I believe that the technology should support the solution. It should not be the solution and then you try to find the problem it solves. That’s the wrong way round.
When I look at Civil and other blockchain experiments I get excited. We will use the technology to do some tests at some point in the future.
More like this