How to improve the click-through rate of Push messages through A/B testing?

The 618 e-commerce festival is in full swing. An e-commerce app is going to push a global event message to users. The operation has prepared two push texts:

Copy A : We know you better when we stock up in the middle of the year. How can we dare to alert you without big discounts: beauty makeup, outdoor sports, medical and health, all in one stop, click >>

Copywriting B : 10% off the whole site specials, 300-30% off for cross-store purchases, and even more discounts for members, 6.1 Passionate pre-sale is open, grab it now, the price is the end, click >>

Which copy is more appealing to users?

We can use the message push to do an A/B test to determine the push copy. Select 0.5% of active users each, and analyze the user's click data after pushing different copywriting. The actual effect is as follows:

Copy A 's CTR: 3.3%

Copy B 's CTR: 11.4%

Post-mortem analysis shows that there is a promotion plan in which users can intuitively obtain digital value in copy B, which is more attractive to users. Therefore, pushing copy B in full can undoubtedly improve users’ click-through rate and conversion rate.

Apply the A/B test function of message push to support the comparison test of message push for different copywriting, different groups of people, and different times, and select the copywriting with better effect or tag people to automatically reissue in multiple sets of comparison results, intelligent and intelligent Realize the best push test and improve the push click rate.

Operators will use push notifications to improve the daily activity of the application or wake up silent users, let users vote for their preferences, and also allow the operation to achieve the purpose of testing with the least cost, and then achieve the best push effect in the full push.

The A/B test function of the push service can be used by integrating the Huawei push service . The push notification experiment can conduct group tests on the title, copy, click action of the push notification, and even the comparison test of the push notification by different groups at different times . Based on the user attributes, audience groups and other conditions output by Huawei analysis, the target users of the experiment can be divided into different groups according to the proportion, and each group can be added with different policies. Then, according to Huawei's analysis service, user data of different groups, such as click-through rate, conversion rate, etc., are obtained.

Huawei Push Service A/B Test Push Effect

Integration steps

First, before using the push notification A/B test, you need to integrate the Huawei push service. You can refer to the official documentation of the push service for integration.

Operating procedures

After integrating the push service, you need to configure the copywriting on App Gallery Connect. I will explain the operation steps with a specific case below.

If you are an e-commerce app operator, the platform is doing red envelope marketing activities. You want to push coupon arrival messages to all users of the platform. You have set up two different sets of copywriting. In order to test a set of copywriting with the best content (highest open rate), randomly conduct A/B testing on the messages sent by App users .
First log in to the App Gallery Connect website, select "My Projects", select "Growth > A/B Testing" in the left navigation bar, find "Create Experiment" in the upper right corner and select "Notify Experiment" from the drop-down box

Fill in the name and description of this A/B test experiment, here we can name the experiment "618 Festival Promotion Push Copy Test"

Go to the "Select Experiment Target Users" page, and set the experiment screening users and proportions. We choose here

The proportion of participating users in "All users" is set to 5%, and the system will filter out 5% of all users and distribute them equally between the original group and the experimental group.

On the "Configure original group and experimental group" page, we can fill in the two push copy information to be tested here. The copy of the original group is "You have a 5 yuan cash red envelope that has been credited to your account, you can get it in a limited amount, you can get immediate discounts when shopping, and you will not be slow." !" Click on the text to jump to the home page.

On the Select Tracking Metrics page, select the primary and secondary metrics that you want to track in this push. In this test, we want to see the open rates of two different push texts, so we set the main indicator to "click-through rate"

On the "Set Message Options" page, you can select the push time, push validity period, and push importance. Click "Save" to enter the experiment report interface, where you can view the specific push effect of this A/B test.

After the above 5 steps, we have successfully completed an A/B test push. By checking the test results, we can find the most ideal notification copy and message presentation method, thereby improving the open rate and conversion rate of the official push.

Learn more details>>

Visit the official website of Huawei Developer Alliance
Get development guidance documents
Huawei Mobile Services Open Source Warehouse Address: GitHub , Gitee

Follow us to know the latest technical information of HMS Core for the first time~