A/B testing your push notification involves comparing two or more different versions of your push message to see which one resonates better with the needs of your website visitors and performs the best in terms of click through rate. In PageSense, you can effortlessly create and test multiple variations of your push notification by changing the title, description, emoji, image, or CTA button color.
For example, you can A/B test your notification by changing only the title while keeping the other attributes the same, or by changing the icon image while keeping the other attributes the same, as shown in the figures below.
Free shipping on Cart! Flat 25% Off on Cart!
Get 25% Off and free shipping on all your cart items today. Hurry up!
Order Now (CTA)
These notifications can then be targeted over your chosen groups of subscribers. The obtained CTR (Click through rate) can then be used to decide which variant works best, so you can continue sending similar notifications in the future.
- You need a minimum of 1000 subscribers to A/B test your push notifications in PageSense.
- You cannot enable this option for the tests that are already launched.
To run an A/B test for your push notification:
1. Under the MESSAGE tab, toggle on the Run an A/B Test button at the top-right corner.
2. Under the ORIGINAL tab, add an appropriate title, message, and URL link for your push notification. You can also choose to add an icon, banner image, and button element to your message on the page. Learn how to add and customize your push message in PageSense.
3. Click the VARIATION 1 tab, then add a new version of the title, message, and URL link of the push notification that you want to A/B test.
4. You can click on the + icon to create up to five versions of your push message. Click the More icon to do any of the following:
Select Rename to edit the name of the new variation.
Select Duplicate to create a quick copy of a variation, then make minor edits to the new variation.
Select Delete to remove a variation.
While running the A/B test, you can also choose to enable or disable auto-send the most successful variation:
In PageSense, your push notification will be sent automatically to the entire target audience equally amongst all the Variations and Original version. For example, if you have 1000 subscribers, a group of 500 subscribers will receive the Original version and another group of 500 subscribers will receive Variation 1.
However, if you wish to A/B test your message on a smaller percentage of your subscribers, and automatically send the most successful variation to the remaining ones, then you can choose to enable the Auto-send the most successful variation, as shown in the figure.
When this option is enabled, the test is run on 30 percent of your subscribers during the first hour of delivery, and a winning variation is determined under REPORTS. Later, you can send that successful notification to the rest of your user base.
A preview of your push messages will be displayed on the right side of the screen based on different devices and browser types.
5. Click Next (at the top right corner) to choose the target audience segments that you want to send out your notification to.
Your push notification will be randomly sent to your chosen group of subscribers, and a winner will be declared based on the CTR (click through rate) obtained for each variation.
6. After you launch and obtain your test results under the REPORTS tab, you can send the winning variation of the Web Push Notification to your users.
Here are a few strategies you can follow before you begin to set up an A/B test for your push notification:
Pick one variable (like CTA, title, image, or description) to test at a time. Don't test too many elements together.
Segment users based on demographics and buying behaviors to run more targeted and action-driven push messages.
Prioritize exactly what you want to test for and once you have a significant result, move on to the next element or test.
Attach a specific business goal, like CTA button clicks, time spent on page, etc to your notification to track conversions and measure the impact of your message among users.