As a news medium, you constantly want to draw your readers' attention to relevant articles. For example, in newsletters or on the website - two platforms in which Froomle has built up a great deal of expertise. The Belgian scaleup helps media companies on four continents personalize their recommendations. Froomle also uses that knowledge for AI powered push notifications, a form of art that takes years to master.

‍

Push notifications are one of the most powerful ways to suggest an article to people. Sending a notification to hundreds of thousands of people is often the way to go to boost an article. At the same time, it is also risky: because these notifications are intrusive, people quickly drop out when they get one that is not relevant to them. For Breaking News the risk of ‘annoying’ users is small, but for specific articles it is key to get your audience right. With the help of personalization via AI, Froomle manages to counteract that churn!

‍

Look at people’s reading behavior

One of the publishers that Froomle works with is Mediahuis, the leading publisher in the Low Countries. They own newspapers such as Het Nieuwsblad, which more than a million people read daily, and several popular regional titles. Instead of pushing all “non-breaking news” to millions of users, Froomle helps select the key interested users.

‍

“There are various ways of personalizing push notifications," says Jonas Boonen, Head of Product at Froomle. “All too often, these start from explicit preferences. In that case, people have to select which notifications they do and do not want to receive. It’s the first step, but ideally, you combine explicit preferences with implicit behavior.”

That's why Froomle's experts base themselves on the same parameters as they use for personalization in newsletters or websites: by looking at people's reading behavior. In the case of our work at Mediahuis, we look at which articles someone read over the past 90 days. When it comes to a regional article, we combine explicit and implicit preferences. People can subscribe to a specific region, but we also look at whether someone reads a certain number of articles about a particular area.

Moreover, we make sure not to annoy the user by limiting the amount of messages each user can receive to an agreed cap, like 5 push notifications per day or 1 per hour. In this way, Froomle creates segments of users who are potentially interested in a specific article. The algorithm only needs a few days to be fully up and running, and afterwards gets smarter and smarter by taking into account more and more data.

‍

Nothing and everything changes at the same time

Unlike other personalization specialists, Froomle is API based. "This means that customers continue to use the tools they already know. We are an invisible component that plugs into their current flow," says Boonen. “Our platform uses different sources of data. The analytics of your online channels, for example. What are people reading? What interests do they have? Do they use your app or the website? How do they respond to push notifications? There is also an article log with metadata about the articles. What is the subject of this article? At Mediahuis, we work with categories and tags, but Froomle can also analyze the content of an article using NLP. This way of working enables us to enrich the metadata."

Because the newsroom continues to use the same applications, their process doesn’t have to change. Those who work with Froomle and want to send a push notification can always choose how many people should receive it. They can also set other parameters by specifying how many relevant articles those people should have read. For regional articles, for example, publishers can choose whether they should also target neighboring municipalities.

The result? People get push notifications they care about, and the churn decreases. The average churn rate for notifications (= daily users that unsubscribe compared to the total subscribed users) is normally around 1,5%. In this case, it’s 0,25% for users that receive Froomle personalized pushes and practically non-existent (as low as 0,02%) for users that click on Froomle personalized pushes. On average, Froomle achieved a CTR that’s ten times higher with the personalized push notification compared to general news pushes for Het Nieuwsblad.

It is also the ideal way to hype an article. Like the TikTok algorithm, publishers can serve articles to a small audience first, and when it's popular there, scale it to a broader audience.