ALERT: INMA Global Media Awards recipients unveiled

Somedia learns the value of frequently testing its push notification strategy

By Corinne Raguth Tscharner


Chur, Graubünden, Switzerland

Four months ago, we — a small team from the Swiss media house Somedia — embarked on a journey along with 18 other local media companies from Germany, Austria, and Switzerland. The journey, called Facebook Accelerator, turned out to be an immense help on our way to figuring out how to build reader revenue, increase the number of paid subscriptions on our news portal,, and create more reader engagement.

The journey started out with high hopes from our end, and the accelerator programme lived up to them easily. The vast knowledge of the involved experts and coaches, the joint and individual sessions, and the dialogue with other publishing companies all played a part in contributing to meeting our expectations.

The idea

The programme taught us one very important lesson: test, test, test. To figure out the best way to accomplish a goal or tackle a problem, we can’t be afraid to conduct tests and to fail. A presentation by another accelerator participant inspired us to take a closer look at how we handle push notifications in our day-to-day news business. We wanted to find out what happens when we send out more push notifications than we had done so far. If we increase the frequency of the daily push notifications, would we increase the engagement of the push subscribers, and would they build a habit of visiting our news portal?

Before we started the three-week test period, we had approximately 17,000 active push subscribers and 90,000 subscribers in total. They received about two or three push notifications a week, exclusively about the most breaking news in the region. We feared the readers would perceive more push notifications on mobile and desktop as intrusive. Spoiler alert: The test results show clearly our fear was not justified.

The parameters

At the beginning of June, we stepped up our push frequency to at least two notifications a day. Usually, we sent one around lunchtime and one in the evening. The increase of notifications meant that we could not be as picky as before with the topics of the pushed articles and could not solely push breaking news anymore. So additionally, we began to push interesting regional and local news stories, quizzes, and texts that took a lot of effort to research and write. We also pushed a promotional article about our editorial morning newsletter once a week.

An example of a push notification sent to a mobile device.
An example of a push notification sent to a mobile device.


We did not differentiate between paid and free content, nor did we notify our subscribers in the push text if there would be a paywall once they clicked on it.

The results

The results are clearly showing that our test was successful and that we could indeed grow user engagement.

Compared to a similar time period before the test, 30% more subscribers clicked on push notifications and accessed our site to read the article. We pushed about four times more than before and generated 50% more sessions during the three-week test period.

Furthermore, it stands out that nearly 5,000 dormant push subscribers became active again. Readers who had subscribed to our push notifications but never paid attention to them nor clicked them suddenly got active again and were interested in reading the articles.

Results showed the push notifications were working for Somedia.
Results showed the push notifications were working for Somedia.

During the test period, we also managed to increase our number of newsletter subscribers. On days we sent a push notification with a promotional article of our newsletter, we saw up to five times more new subscriptions than on any other day.

To summarise, the key results were the following:

  • 30% more users clicking.
  • 50% more sessions.
  • 4,932 users rengaged (about 25% of former active users).
  • Four- to five-times newsletter uplift in newsletter subscriptions on the day of the push.

Last but not least, our readers did not get fed up by the increased number of push notifications, which was the fear we had going in. If anything, we are now encouraged to keep the higher frequency of pushes and to find out more about their best timing and topics.  And the results encourage us to keep on living by one of the credos of the Facebook Accelerator programme: Keep on testing, testing, testing. 

The Audience Analytics Accelerator programme is a joint initiative between Facebook and INMA. The Facebook Journalism Project’s Accelerator programme helps news publishers build sustainable businesses. Funded and organised by the Facebook Journalism Project, each Accelerator includes a three-month period of hands-on workshops led by news industry veterans, grants administered by non-profit journalism organisations, and regular reports on best business practices. The Accelerator’s executive director is Tim Griggs, an independent consultant/advisor and former New York Times and Texas Tribune executive. Facebook and INMA have partnered to bring dozens of publishers from around the world into INMA's Readers First initiative. This case study reflects the partnership between INMA and the Facebook Journalism Project to develop excellence among leading global publishers in reader revenue initiatives. 

About Corinne Raguth Tscharner

By continuing to browse or by clicking “ACCEPT,” you agree to the storing of cookies on your device to enhance your site experience. To learn more about how we use cookies, please see our privacy policy.