Visual data to give Yle better understanding of users
With the increased impact and popularity of Yle’s online content, there is also a greater need for data about the users. The data is needed to be able to provide better public service to all Finnish people, and especially to those who do not use Yle’s services often.
Television viewers and radio listeners have been described in detail for quite some time already, whereas online users have been perceived as faceless browsers. In order to solve the problem of the online audience, Yle turned to agile modelling together with its partners Futurice and Reaktor.
The result is a new and unique way of predicting the age and gender distribution of online users. The model provides reliable data about popular content as well as programmes with fewer viewers. The model was also expanded to include audio listeners and article readers.
The results and the data visualizations are intended to provide answers to the following two questions:
- How can Yle provide better service to people who use only a little of Yle’s online content?
- How can Yle obtain user data that all Yle employees can utilise better in their work?
Making use of all browsing on Yle’s website, the model does not, however, affect the users’ views or user experience of the online services. It is more like a tool that helps Yle understand what kinds of content people use.
A result of agile modelling
“In our everyday work, the model gives editors a better idea of how well a programme or article works for a specific audience, and we can monitor this almost in real time”, says Eija Moisala, head of the Smart Data and Audience Insight at Yle.
Yle’s model has generated interest in other European media companies. The BBC and the Nordic public service broadcasters have also been keen on learning more about the data architecture behind the model.
“In addition to age and gender, interesting subjects for modelling are also the various motives and contexts of media use. We are already working on solutions for these areas”, says Eija Moisala.
"In addition to age and gender, interesting subjects for modelling are also the various motives and contexts of media use.
Better data is the foundation for creating better content
The data modellings have made Yle a pioneer in understanding audience data. It has also brought about changes in Yle’s internal routines. All Yle employees, from the editors to the CEO, can use the data in their work. The data can also easily be rendered in visual form.
“The general audience has quite clearly discovered online viewing, which is growing rapidly right now. Some episodes of Sohvaperunat (Gogglebox) have reached new record highs, up to more than 120,000 starts", says Arttu Nurmi, commissioner of culture and entertainment programmes at Yle.
“The new way of online audience modelling has provided real-time information on viewer profiles. At the same time, we have been able to provide both better content and offering.”
"Some episodes of Sohvaperunat have reached new record highs, up to more than 120,000 starts.
The available data is getting better and more accurate constantly, improving Yle’s insights in user preferences. The better Yle understands what individuals need, the better it can serve all Finns.