Notification CTR

Troubleshooting low notification Click-Through Rates

Click-Through Rate (CTR) is measured by (clicks / delivered) * 100% and represents how many notifications were clicked based on how many were delivered to the push servers who then send it to your subscribers. It does not count:

  • swiping away the notification
  • clicking "Dismiss"
  • clearing the notifications

Best Practices

CTR for push notifications varies widely among different platforms (Android, iOS, Web), verticals (news, games, travel, e-commerce, etc), type of notification (message from a friend, promotion, informative update), country, etc. Not unlike email, the variance is so wide as to make averaging a bit difficult.

To help with maximizing clicks on push, we regularly publish blog posts and best practices on how to utilize iZooto to get the maximum engagement from users.

A common best practice to increase CTR and user engagement is to send subscribers messages based on topics they are interested in. If your site publishes a wide range of topics and sends more than 1 notification a day, your users likely would not like to know about every push you send.

Websites can easily segment users based on their behaviour on the website using what we call User Properties. This will allow you to group users by the different topics you offer so you can set up Segments and target users with the notifications they care about.

Troubleshooting Low CTR

There are a couple of reasons that the push notifications CTR might be dropping. Review the below questions for common issues and solutions.

When did you notice the drop?

It is important to check what specific timeframes you are comparing?

  • What date did you start seeing a decrease in clicks?
  • What dates in the past are you comparing with?

What metrics are you specifically tracking?

Are you checking the reports for clicks or CTR?

  • Clicks would be the actual click events where CTR could change based on how many devices is the notification being sent to. Are you tracking only specific types of messages?
  • For example, are you only tracking marketing messages or transactional messages or are you tracking all messages?

How are you gathering and reading the analytics data?

What reporting tools are you reviewing and how is the data generated?

What changed?

Once you have a clear picture of the issue, you may notice that some internal changes might have occurred.

Changed code on site

Did you or your dev team change the way iZooto is implemented on your app or the website?
Was another push service added to the app/website?

iZooto is not active on all pages of the website

If you removed iZooto's code or Service Workers from your website or app, you will see a decrease in user stats and clicks. Make sure iZooto is active and working on all pages of the website for optimum performance.

Another Service Worker

Web Push Notifications require Service Workers to run on the browser so the user can receive the push message. Adding multiple Service Workers on a website (for example a service worker for PWA) can cause issues with push being received and CTR.

Please see our Service Worker's guide.

If you have already fixed this or removed the other service worker, then users will need to return to the site for our service worker to be added again.

Changed sending patterns

Have you or your team started sending more or fewer notifications since the drop?
Perhaps you started sending more API messages.

Did you change any of the segments being targeted? For instance, did you use to send notifications to all users, but around the date of the drop, you started sending to fewer users within the segments.

Too many notifications

This can be different depending on the platform and type of notifications you are sending, but if you are blasting users with notifications every few hours with content they may not be interested in, that usually leads to low CTRs.

Boring Notifications

Data shows adding flare to the notification gets more clicks. This can be emojis, images, icons, or just engaging words.