Event Based Push Notification¶
Event Based Push Notification is a function that delivers push notifications depending on the actions of users in the service.This allows you to trigger push notifications based on the behavior of users on your site.
This is useful for the following use cases.
You want to capture changes in user information (ex. member status) and send notifications.
You want to send a notification to users, who did "Add to Cart" and left the site without completing the purchase and recommend them to complete the purchase in an hour from their leaving.
You want to send a push notification in an hour to users, who started signing up but didn't complete it.
Create Event Based Push Notification campaign¶
To create an Event Based Push Notification, follow the steps below.
- Move to the Event Based Push Notification's creation screen.
- Enter campaign information
- Select Target
- Enter delivery settings
- Enter messages
- Press LAUNCH to start the campaign.
Move to the Event Based Push Notification's creation screen.¶
Log in to the dashboard, click "Marketing" -> "Event Based Push Notification" in the left menu and go to Event Based Push Notification List screen.
Note
If you do not find the "Event Based Push Notification", please contact our CS Team.
After the list page is displayed, click the "CREATE NEW" button in the upper right corner to go to the screen for Event Based Push Notification's creation screen.
Enter campaign information¶
Enter the title of the campaign, the campaign goals, the campaign hypothesis, and the results.(The campaign goals, the campaign hypothesis, and the results are optional.)
Select Target¶
Select the target of the campaign.
Note
If you select "Add Cart at least once" as the target and "Add Cart" as the trigger, the push notification will be delivered when "Add Cart" is executed for the first time, not when it is executed twice.
This is not the case if segmentation takes a long time.
Enter delivery settings¶
Enter the delivery settings for the campaign.
Enter the trigger conditions. You can specify the event or property that will be the trigger for delivering the push notification.
It is not possible to specify user profile's change as a trigger condition.
You can choose whether to deliver the message immediately after the specified origin event is executed or after a certain amount of time has passed.
However, it may take several minutes to several dozen minutes depending on the congestion of the server, even if you have set up immediate delivery.
The elapsed time can be specified from 15 minutes as minimum to 7 days as maximum.
Set the delivery period.
Once a campaign is saved, the delivery period cannot be changed. If you want to change it, please duplicate the corresponding campaign.
Set the delivery time range. To prevent delivery at unintended times such as midnight or early morning, you can set the time when delivery is allowed.
If an a trigger condition's event for push notifications is executed outside of the set delivery time range, push notifications will be delivered when the next delivery time range comes.
If the campaign delivery period has passed while you are waiting for the next delivery time range, the push notification will not be delivered.
Once a campaign is saved, the delivery time range cannot be changed. If you want to change it, please duplicate the corresponding campaign.
In order to stop delivery to unintended users, you can select an event to cancel delivery.
Depending on the timing of the event to cancel the delivery, such as just before the delivery or triggered time close to cancel, the delivery may be made as is.
To prevent push notifications from being delivered multiple times a day, you can set the frequency of delivery.
If you register events that may be executed many times a day as trigger conditions, push notifications may be delivered in large numbers.
Enter messages¶
Enter the message content for the push notification.
Press LAUNCH to start the campaign.¶
By starting the campaign, push notifications will be delivered when the targeted user tracks the event of the trigger in the delivery settings.
Campaign Measurement¶
You can check how effective the Event Based Push Notification was by confirming the number of openings and conversions.
Delivery Result¶
- Deliveries
The number of requests from the Repro server to APNs and FCM that were successfully accepted among push notifications forusers with push notification permission ON.Aggregates the number of unique users on a daily basis.
- Opens
Number of total users who have opened the campaign on a daily basis. This includes both direct and influenced opens.
- Influenced Opens
Number of users who have activated the app within 1 hour from the delivery of the notification.
- Direct Opens
Number of users who have activated directly from the push notification.
- Conversions
The number shows the daily total of users who have achieved the goal designated in the creation page within one hour of “opening” the Event Based Push Notification.
“Opening” can mean two things: influenced opens and direct opens.
- Conversion Ratio
Number of conversions divided by deliveries.
- Campaign Progress
Results of the deliveries, opens and occurrence of the conversion event on a daily basis: Note that the results represent number of users.
Note
If the app is launched when the notification is tapped within an hour of delivery, it will only be counted as a direct open and not an influenced open.
A/B Testing¶
A/B testing allows you to see what kind of content drives better open rate or conversion rate by delivering different patterns of the message to divided group of users.
Create Message¶
By composing a message with two or more variants, an A/B-Test can be performed.
You can create up to 4 different patterns.
Adding or removing message variants is not possible after a campaign has been saved.
Distribute Users Among Variants¶
Distribution percentages can be individually adjusted when having more than 2 variants making the total 100%.
You can assign equal percentages to Control Group as well as each of the variants by clicking Distribute evenly, while Distribute evenly (without control group) will make percentages even among multiple variants.
Note
Each of the variants' percentage can be anything between 1% and 100%.
Only integers can be set.
Changing the distribution settings is not possible after a campaign has been saved.
The actual number of message delivery may fluctuate depending on the volume of the target user.
Campaign Measurement (A/BTest)¶
Details screen
You will see the results after a while from the delivery. The charts will show the open rate and the conversion rate of each of the patterns.
During the A/B test, the results of the test will be displayed. Here, you will be able to compare the markers (CVR, open rate) that have selected different patterns.
Pattern
Each pattern used to deliver the A/B test, will be displayed here.
Baseline
Here, you will be able to select the baseline pattern that will be used as the standard when comparing each pattern. By changing the baseline, you will be able to compare between a variety of different patterns.
Targeted indicator
An indicator is used when comparing patterns to baselines in A/B tests. You can select a ratio value (i.e. CVR, open rate) as the indicator. Also, the indicators listed can be changed through the editing page.
90% confidence interval
The 90% confidence interval will be displayed when each pattern is compared to the baseline. For the definition of 90% confidence interval, please see the FAQ. Here, we will talk about the most common ways to recognize patterns.
Pattern 1:
When displayed as such, it means that the difference between the pattern and the baseline is significant. We recommend moving onto the winner pattern if enough time has passed since delivery.
Pattern 2:
When displayed as such, it means that the difference between the pattern and the baseline is not significant. Either wait until the results appear or consider moving on to the next campaign if enough time has passed since the delivery.
Pattern 3:
When displayed as such, it means that the difference between the pattern and the baseline is significant. If this is the only window displayed, we recommend determining the baseline pattern as the winner pattern.
Note
Below is an example of when the winner pattern is selected from the A/B test significant difference results. The comparison results of the rest of each pattern will not be displayed.
Campaign Progress
The graphs show patterns of open rates as well as conversion rates for event based push notifications.
Status List Screen
You will be able to check the status of the A/B test for each campaign here, when the operating campaign is either live or not. Icon display conditions are as follows.
Icon |
Display conditions |
---|---|
A significant difference exists, and the results of the pattern exceed the baseline at least once |
|
A significant difference exists, and the results of the pattern are only of ones that fall below the baseline |
|
Only patterns without significant difference |
|
If all patterns can't be calculated |
|
If winning pattern is already selected |
|
When either the targeted indicator or the baseline pattern of the A/B test is not set |
|
ー | When the campaign is not conducting A/B test |
Note
In the instance of a campaign that has a significant difference with the baseline having both exceeding and falling below results, it will be displayed as the results of the pattern with at least one exceeding the baseline.
For questions, please visit the FAQ.
Select the winner¶
Once you select the winner, all after deliveries of the same event based push notification will be using the selected pattern to all target users.
Keeping the pattern with the highest open rate is likely to be effective to improve retention rate. Choosing the one with the highest conversion rate will also be good for increasing overall KPI.
FAQ¶
Is there a limit to the number of Event Based Push Notification that can be sent?
- There is no limit to the number of push notifications you can publish. The number of campaigns that can be published can be increased. Please contact your CS for details.
It will be managed separately from normal push notifications.
How do you measure effectiveness?
- It can be measured in the same way as normal push notifications.
You cannot use Control Groups in an Event Based Push Notification.
Note