A Case For Personalized Push Notifications.
September 18, 2017

Being a news addict, I have (only) 6 news apps installed on my mobile phone. All of them offer breaking news push alerts and for a reason: push works for news organizations. According to this study by Newscycle Solutions, push notifications increase engagement significantly:

  1. An average user spends 153 seconds (about 2.5 minutes) longer in an app session when directed to the app via an article push notification.
  2. App opens from push notifications represent more than 21% of all opens.
  3. Almost 70% of respondents to Newscycle’s survey indicate that they have opted to receive pushes from their favourite media apps.

Data compiled this spring by Urban Airship found that publishers who push at least one notification to their app users every day have app retention rates more than twice as high as those for publishers that don’t.

But, according to the same Newscycle survey, 50% of the users remove push because they receive too many messages, see the graph below.


graph why users remove push

And that’s exactly what I personally do not like about push messages: I often get the same news pushed by 4 different apps. And I can only “personalize” the frequency. Result: I deactivated push for 3 of my 6 news apps.

So why do online publishers not add personalized push messages next to “breaking news” push alerts? Some examples:

  1. A push alert if a new development about an interesting article/video I read or watched 2 weeks ago.
  2. When Ajax, my favorite football team, is not just mentioned in an article but is the main subject of it.

The main reason why this functionality is not available is that it is very hard to implement:

  1. Most news organizations relate old content with a new story they publish. But the old story is not related with the new story.
  2. They use keywords to match content, which is inaccurate and takes up much time of their editorial staff, resulting in inaccurate related content or none added at all.
  3. And because most tagging solutions do not offer a qualitative score of the relatedness between the topic (“Ajax”) and the article/video, my second use case is not trivial to implement either.

Klangoo’s Magnet supports personalized push. Using semantic understanding of content along with machine learning user profiling techniques Magnet helps in offering your users a complete story-telling experience along with a personal touch.

And if you offer more flexibility, a more personalized experience, I am sure the push of your competitors will be de-activated, and you will be top of mind for your app users.

If you want to know more, reach out to us on Twitter.

Klangoo NLP
Subscribe to our Investors' newsletter to download the Investor Relations Brief & receive exclusive content!