You would often want to split your users based on certain conditions within a flow to personalize the experience or split your users randomly to experiment. MoEngage offers the following ways to split your users:
Way to Split
Description
Check User Attribute
This stage allows you to split your users based on who meets the defined user properties check.
Has Done Event
This stage allows you to split your users based on who meets the defined actions-based check.
Conditional Split
This stage allows you to divide users based on a combination of their user properties and actions, resulting in a maximum of 25 different paths.
Campaign-Based Conditions
These are custom stages that are displayed after you add Action stages in the flow. This allows you to split users based on conditions related to the previously-added campaigns.
A/B Split
This stage allows you to split users randomly based on the defined allocation percentage and experiment on them.
Intelligent Path Optimizer
This stage allows you to split users randomly using the Merlin AI intelligence of MoEngage and help you experiment on them.
The following is the use case for splitting users randomly:
Want to split users randomly to find the channel type that works for my user base? Add an A/B split stage to help you achieve the same.
When you want to divide your users based on the Entry Criteria, add a Conditional Split or a Has Done Event Stage and select the option under the Flow Entry Events Category. When this option is selected, MoEngage considers the event that allowed the user to enter the flow while deciding the path for the user.
Perform the following steps to add condition stages:
On the MoEngage Canvas, click +.
In the Add stage dialog box, do one of the following:
In the search bar, search for the stage directly. From the search result, click the option you want to add. For example, if you search for “conditional”, the Conditional Split option is displayed.
On the All stages tab:
Under Conditions & controls, click the category to explore the condition stages you can add:
Category
Stages Available
AI TOOLS
Intelligent path optimizer
USER CONDITIONS
Check User Attribute
Has done event
SPLIT USER PATH
Conditional Split
A/B Split
Action conditions
Has received push notifications
Has dismissed push notifications and other campaign-based conditions
Click the stage you want to add.
The five frequently used stages are displayed under the Frequently Used section on the All stages tab. The list of available stages changes dynamically based on your workspace. Click the required stage to add it to the flow.
On the Conditions & controls tab, click the stage you want to add. The selected stage is added to the canvas.
Enter a name for the stage to manage and identify the condition.
Branches
Users are evaluated against each branch in sequence. All users are evaluated first against Branch 1, then Branch 2, and so on, until the default branch.
Click the pen icon beside the branch name to change the name of the branch or path. This name helps you identify the path or condition. Note: Ensure the display name does not exceed 20 characters.
Use one of the following:
User property
In the list, select a user property.
Provide the property details. The filter is the primary filter condition.
Click + Filter to add more filters to the primary filter condition.
User behavior
In the list, select one of the following:
Has Executed
Has Not Executed
In the list, select an event. Events include any flow entry conditions.
Click + Attributes to add filters to the primary condition.
The default Constant value helps you complete the filter with specific static values.
Use Event attribute (applicable for all types of flows), User attribute (applicable for all types of flows), and Business event attribute (applicable only for Business Event Triggered flows) to link this primary condition with the event-trigger or business event trigger conditions and define dynamic checks.
Click to add another filter with the primary condition.
Click AND or OR to define the relationship between filters.
In the list, select the type of user attributes.
Add additional conditional checks and link them to previous user actions with Linked Conditions.
Click +Add to include more branches as part of the conditional split stage.
Branches are evaluated in sequence. Branch 1 has the highest priority, which decreases as you move down the list. The Default branch has the lowest priority.
For example, when there is a conditional split where a user matches both branch 1 and branch 2 conditions, they move down Branch 1 because of its highest priority. So, in a conditional split, where the user matches more than one branch’s criteria, the order of the branches matters.
To delete a branch within Conditional Split or to duplicate the branch, hover over the list of branches mentioned on the left side.
Click Duplicate to create a copy of a branch. The new branch is automatically named “Copy of Branch <x>”
Similarly, click Delete to remove a branch from the stage.
In this step, define the time window for evaluating the condition by setting a start point (using the timeline slider) and an end point (using the wait duration).How It Works: The evaluation window consists of a start point and an end point (calculated as: when the user enters the stage + wait duration).
Slider Position
Configuration Fields
Evaluation Window
Old UI Equivalent
A fixed time before users enter this stage
Enter the fixed time to evaluate before users reach this stage.
Enter duration (number).
Select unit: mins, hours, days.
Max: 30 days
Enter the fixed duration to wait after users enter this stage
Enter duration (number).
Select unit: mins, hours, days
Max: 30 days
Start: Time before reaching the stage.
End: Time after reaching the stage.
Example: User reaches stage at 2:00 PM - “5 mins before” + “10 mins after” - Evaluates: 1:55 PM to 2:10 PM
”In the past”
When users enter this flow
Enter the fixed duration to wait after users enter this stage
Enter duration (number)
Select unit: mins, hours, days
Max: 30 days Note: Start time is automatically set to flow entry time (max 30-day lookback)
Start: When the user enters the flow
End: When the user reaches the stage + wait duration
Example:
User enters flow on Jan 1, reaches stage on Jan 5 at 2:00 PM - “5 hours after” - Evaluates: Jan 1 to Jan 5 7:00 PM
If the user enters on Jan 1 but reaches the stage on Feb 15, it evaluates from Jan 16 only
”Since entering the flow”
When users reach this stage
Enter the fixed duration to wait after users enter this stage
Enter duration (number)
Select unit: mins, hours, days
Max: 30 days Note: Start time is automatically set to when the user exited the previous stage
Start: When the user exits the previous stage
End: When the user reaches this stage + wait duration
Example: User exits previous stage at 2:00 PM, reaches this stage at 2:30 PM - “2 hours after” - Evaluates: 2:00 PM to 4:30 PM
”Since exiting the last stage”
Old Navigation
Click the pen icon to edit the condition.
Do the following:
Field
Description
Stage name
Type to modify the condition’s name to manage and identify the condition.
Evaluate the following conditions
In the list, click an option to choose the start of your evaluation window:
Since entering the flow This defines the start time to be from the time the user has entered the current flow. MoEngage evaluates the condition from the time the user entered the flow to decide whether the user meets the condition. However, the maximum look-back is 30 days. For example:
If a user enters the flow on January 1 and reaches the condition stage that’s defined to evaluate “Since they entered the flow” on February 15, MoEngage can evaluate only from January 14 because the maximum look-back it has is 30 days.
If a user enters the flow on January 1 and reaches the condition stage that’s defined to evaluate “Since they entered the flow” on January 15, MoEngage can evaluate from January 1, that is, the time they entered, because this falls within the max limit defined.
Since exiting the last stage This defines the start time to be from the time the user has moved from the previous stage in the flow to the next stage.
In the past This defines the start time to be from the set timeframe in the past. Set the time in minutes, hours, or days. MoEngage evaluates the condition for the time you mention with this option. For example:
If a user enters the flow on January 1 and reaches the condition stage that’s defined to evaluate “In the past for 30 Days” on January 15, MoEngage will evaluate from December 14, effectively for the defined 30-day period.
If a user enters the flow on January 1 and reaches the condition stage that’s defined to evaluate “In the past for 30 Days” on February 20, MoEngage can evaluate from January 21, effectively for the defined 30-day period.
Condition is satisfied if the user
In the Display name box, type or paste the name of the branch or path. Note: Ensure that the display name has a maximum of 20 characters only.
Use one of the following:
User property
In the list, select the user property.
Provide the property details. The filter is the primary filter condition.
Click + Filter to add additional filters to the primary filter condition.
User behavior
In the list, select one of the following:
Has Executed
Has Not Executed
In the list, select one of the events. Events include any of the flow entry conditions as well.
Click + Attributes to add additional filters to the primary condition. The default Constant value helps you complete the filter with specific static values. The Event attribute (applicable for all types of flows), User attribute (applicable for all types of flows), and Business event attribute (applicable only for Business Event Triggered flows) option, as shown above, helps you link this primary condition with the event-trigger or business event trigger condition and define dynamic checks. To know more, refer to Linked Conditions
+ Filter
Click to add another filter with the primary conditions.
Click AND or OR.
Similar to the step above, you can add additional conditional checks and link them to previous user actions with Linked Conditions.
Keep evaluating for the next
Turn it on to check if the condition is met for the set time duration from the start time defined in Evaluate the following conditions.
Type or use the up and down arrow keys to set the time duration.
In the list, select the minutes, hours, or days to set the evaluation duration.
Use the + icon to add more branches as part of the conditional split stage. Branch 1 has the highest priority, and the priority reduces as you progress down. Default branch has the least priority. For example, when there is a conditional split where a user X matches both branch 1 and branch 2 conditions, the user moves down Branch 1 as that has the highest priority. So, for a conditional split, the order of branches plays an important role where the user moves provided they match more than one branch’s criteria.
The default branch does not have any conditions.
You can add personalized and dynamic conditional checks based on the previous user actions with Linked Conditions.
Users navigate to the default branch when they do not meet the conditions applied in the conditional split.
The order of branches will influence which branch path the user takes in a conditional split, because the priority of the branches decreases from the first branch to the last branch or the default branch.
Users meeting the primary branch will leave as soon as they meet the criteria. Meanwhile, in the other branches, they will leave only after the completion of the evaluation period.
Users navigate the branch and follow the branch after meeting the evaluation criteria.
When you edit a published flow having a conditional split, the following are applicable:
Change the display name.
Change the branch evaluation conditions.
Cannot add a new branch.
Cannot delete a branch.
After adding a Check User Attribute stage, perform the following steps to configure it:
New Navigation
This is an Early Access feature. To enable it for your account, please contact your MoEngage Customer Success Manager (CSM) or the Support team.
Field
Description
Stage name
Enter a name for the stage to manage and identify the condition.
Condition(s)
Define the user property condition by selecting a user attribute from the list The default Constant value helps you complete the filter with specific static values.
Use the User attribute (applicable to all flow types) and Business event attribute (applicable only to Business Event Triggered flows) options to link this condition to other user attributes or the business event trigger condition, and to define dynamic checks. To know more, refer to Linked Conditions.
+Filter
Click to add another filter with the primary condition.
Click AND or OR to define the relationship between filters.
Select the type of user attribute from the list.
Evalutation Window Configuration
Specify the maximum duration to wait before evaluating the condition in Evaluation Ends:
Enter the time duration (number)
Select the time unit: minutes, hours, or days
How it works:Once a user reaches this stage:
MoEngage checks the evaluation window to see if the condition is met.
If the condition is met, the user immediately moves to the “Yes” path.
If the condition is not met, MoEngage waits up to the specified duration before moving the user to the “No” Path.
In the meantime, if the user attribute is updated, they are evaluated to move users in real-time.
Old Navigation
Click the pen icon to edit the condition.
Do the following:
Field
Description
Stage name
Type to modify the condition’s name to manage and identify the condition.
Condition is satisfied if user
In the list, select one of the attributes. The default Constant value helps you complete the filter with specific static values. The User attribute (applicable for all types of flows) and Business event attribute (applicable only for Business Event Triggered flows) options, as shown below, help you link this condition with other user attributes or the business event trigger condition and define dynamic checks. To know more, refer to Linked Conditions.
+ Filter
Click to add another filter with the primary filter.
Click AND or OR.
In the list, select the type of user attributes.
Keep evaluating for the next
Turn it on to check if the condition is met for the set time duration.
Type or use the up and down arrow keys to set the time duration.
In the list, select minutes, hours, or days to set the evaluation duration.
For example, a user came to the event condition evaluation at time T (after viewing a product page), and you want to track whether your user has added the product to the cart and made the purchase within 2 hours. MoEngage Flows will check for the next two hours if the condition is met.If yes, it will immediately move to the yes branch. If the condition is not met at the end of 2 hours, it will move to the no branch. Marketers can accordingly decide the subsequent flow paths. The system checks if the actions have been taken in the past since the user entered the journey. The system will wait 2 hours if the condition is not met. After 2 hours, it will move to the yes or no branch based on the current status.The evaluation starts when the user reaches the event condition. For certain use cases, marketers might want to check whether some actions have been taken since their users have entered the journey. Marketers can use the option accordingly.
If the condition is met, the user will immediately move to the yes branch. If the condition is not met at the end of the evaluation window, the user will move to the no branch. Marketers can accordingly decide the subsequent flow paths.
After adding a Has done event stage, perform the following steps to configure it:
New Navigation
This is an Early Access feature. To enable it for your account, please contact your MoEngage Customer Success Manager (CSM) or the Support team.
Enter a name to identify and manage this condition.
Condition is satisfied if the user
In the list, select one of the following:
Has Executed
Has Not Executed
Select an event from the dropdown. This includes:
Custom events tracked in your workspace
Flow entry events (to check the event that triggered the flow entry)
Click + Attributes to add additional filters to the primary condition.
Use this default Constant value to complete the filter with specific static values.
Use Event attribute (applicable for all types of flows), User attribute (applicable for all types of flows), and Business event attribut (applicable only for Business Event Triggered flows) option to link this primary condition with the event-trigger or business event trigger condition and define dynamic checks. For more information, refer to Linked Conditions.
+Filter
Click to add another filter to the primary condition.
Select AND or **OR **to define the relationship between filters.
In the list, select the type of user attributes.
Add additional conditional checks and link them to previous user actions with Linked Conditions.
In this step, you define the time window for evaluating the condition by setting a start point (using the timeline slider) and an end point (using the wait duration).How It Works: The evaluation window has a start point (determined by slider position) and an end point (always when the user enters this stage + wait duration). Together, they create the complete timeframe for checking if your condition is met.
Slider Position
Configuration Fields
Evaluation Window
Old UI Equivalent
A fixed time before users enter this stage
Enter the fixed time to evaluate before users enter this stage.
Enter duration (number)
Select unit: mins, hours, days.
Max: 30 days
Enter the fixed duration to wait after users enter this stage
Enter duration (number)
Select unit: mins, hours, days
Max: 30 days
Start: [Time before reaching stage] End: [Time after reaching stage] Example: User reaches stage at 2:00 PM - “5 mins before” + “10 mins after” - Evaluates: 1:55 PM to 2:10 PM
”In the past”
When users enter this flow
Enter the fixed duration to wait after users enter this stage
Enter duration (number)
Select unit: mins, hours, days
Max: 30 days
Note:_ Start time is automatically set to flow entry time (max 30-day lookback)_
Start: When user entered the flow
End: When user reaches stage + wait duration
Example: User enters flow Jan 1, reaches stage Jan 5 at 2:00 PM - “5 hours after” - Evaluates: Jan 1 to Jan 5 7:00 PM If user enters Jan 1 but reaches stage Feb 15, evaluates from Jan 16 only
”Since entering the flow”
When users reach this stage
Enter the fixed duration to wait after users enter this stage
Enter duration (number)
Select unit: mins, hours, days
Max: 30 days
Note: Start time is automatically set to when user exited previous stage
Start: When user exited previous stage
End: When user reaches this stage + wait duration
Example: User exits previous stage at 2:00 PM, reaches this stage at 2:30 PM - “2 hours after” - Evaluates: 2:00 PM to 4:30 PM
”Since exiting the last stage”
How Evaluation WorksOnce a user reaches this stage, MoEngage evaluates the window (start point to end point):
If the condition is already met: The user immediately moves to the “Yes” branch
If the condition is not yet met: MoEngage continues checking until the end point is reached
At the end point:
Condition met: The user moves to the “Yes” branch
Condition not met: The user moves to the “No” branch
Old Navigation
Click the pen icon to edit the condition.
Do the following:
Field
Description
Stage name
Type to modify the condition’s name to manage and identify the condition.
Evaluate the following conditions
In the list, click an option to choose the start of your evaluation window:
Since entering the flow This defines the start time to be from the time the user has entered the current flow. MoEngage evaluates the condition from the time the user entered the flow to decide whether the user meets the condition. However, the maximum look-back is 30 days. For example:
If a user enters the flow on January 1 and reaches the condition stage that’s defined to evaluate “Since they entered the flow” on February 15, MoEngage can evaluate only from January 14 because the maximum look-back it has is 30 days.
If a user enters the flow on January 1 and reaches the condition stage that’s defined to evaluate “Since they entered the flow” on January 15, MoEngage can evaluate from January 1, that is, the time they entered, because this falls within the max limit defined.
Since exiting the last stage This defines the start time to be from the time the user has moved from the previous stage in the flow to the next stage.
In the past This defines the start time to be from the set timeframe in the past. Set the time in minutes, hours, or days. MoEngage evaluates the condition for the time you mention with this option. For example:
If a user enters the flow on January 1 and reaches the condition stage that’s defined to evaluate “In the past for 30 Days” on January 15, MoEngage will evaluate from December 14, effectively for the defined 30-day period.
If a user enters the flow on January 1 and reaches the condition stage that’s defined to evaluate “In the past for 30 Days” on February 20, MoEngage can evaluate from January 21, effectively for the defined 30-day period.
Condition is satisfied if the user
In the list, select one of the following:
Has Executed
Has Not Executed
In the list, select one of the user attributes. The filter is the primary filter condition.
Click + Attributes to add additional filters to the primary condition. The default Constant value helps you complete the filter with specific static values. The Event attribute (applicable for all types of flows), User attribute (applicable for all types of flows), and Business event attribute (applicable only for Business Event Triggered flows) option, as shown above, helps you link this primary condition with the event-trigger or business event trigger condition and define dynamic checks. To know more, refer to Linked Conditions.
+Filter
Click to add another filter with the primary condition.
Click AND or OR.
In the list, select the type of user attributes.
You can add additional conditional checks and link them to previous user actions with Linked Conditions.
Keep evaluating for the next
Turn it on to check if the condition is met for the set time duration from the start time defined in Evaluate the following conditions.
Type or use the up and down arrow keys to set the time duration.
In the list, select minutes, hours, or days to set the evaluation duration.
For example, a user came to the event condition evaluation at time T (after viewing a product page), and you want to track whether your user has added the product to the cart and made the purchase within 2 hours. MoEngage Flows will check for the next two hours if the condition is met.If yes, it will immediately move to the yes branch. If the condition is not met at the end of 2 hours, it will move to the no branch. Marketers can accordingly decide the subsequent flow paths.The system checks if the actions have been taken in the past since the user entered the journey. The system will wait 2 hours if the condition is not met. After 2 hours, it will move to the yes or no branch based on the current status.The evaluation starts when the user reaches the event condition. For certain use cases, marketers might want to check whether some actions have been taken since their users have entered the journey. Marketers can use the option accordingly.
A/B Split allows you to test up to five flow paths variations to determine which is the best-performing option. You can experiment with different channels, different messages on the same channel, and different message frequencies across these flow paths.You can configure the experiment and, after analyzing the performance of each branch on publishing the flow, change the user distribution in favour of the winning branch. This enables most users entering the flow to go forward with the best-performing branch.
Opt for intelligenceYou can configure MoEngage Flows to automatically adjust the user attribution in favor of the best-performing branch as more and more users enter the flow.For more information, refer to Intelligent Flow Path optimizer.
As users enter the flow, each branch’s performance keeps updating on the canvas.
Email (50%) denotes the display name of the branch and the configured user distribution for the branch, which is 50 in this example.
4 Done is the number of flow trips that have gone through this branch.
Converted trips are the number of branch trips that resulted in at least one conversion. 75% is the conversion rate for the branch in the example above.
The Engagement Score is a derived metric reflecting the engagement of the branch. The higher the number, the better the engagement. For more information on how the engagement score is calculated, refer here.
What is a branch?
All the action stages from an A/B split node to the flow exit stage are included in tracking the branch’s performance.
If the branch splits into multiple branches, the resulting branches are considered part of the parent branch.
If a branch merges into another branch at some point, the common flow path is considered a part of both the merging branches.
You can analyze and compare the branches’ performances in the report for your split node. You can click the report option hovering over the stage on canvas or go to the split node stats section in the flow’s detailed stats page and filter on a particular split node from the list.After analyzing the branch performance, you can choose to edit the configured user distribution so that most of the users entering the flow next will go forward through the best-performing branch.
Intelligent Path optimizer allows you to A/B test up to five branches where Merlin AI automatically keeps adjusting the user distribution towards the best-performing path. You can focus on experimenting on which channel to use or what should be the messaging frequency in each branch and let Merlin AI optimize the user distribution.While deputizing Merlin AI to optimize the user distribution, you must also configure the metric you are expecting to maximize in your flow. This can be one of Engagement, Conversion, or both.
Worthy Alternative If you wish to handle the user distribution manually, you can opt for A/B Split. For more information, refer here.
You can add an Intelligent Path Optimizer at any part of your flow where you want to experiment with some worthy alternative options.Make sure you name each branch distinctly to identify them on the canvas. At every stage, you can experiment with up to five flow paths, and in each flow, you add up to five such stages.
Initially, the users get uniformly distributed, and as MoEngage get more performance results, Merlin AI keeps adjusting the user distribution towards the best-performing branch. Based on which metric is configured to be maximized, MoEngage tracks different performance stats of each branch involved in the experiment.
Maximizing conversions
For each branch, MoEngage aims to maximize the ratio of converted branch trips/total trips of the branch. This can be loosely termed as the Branch CVR as well.\
Things to notice MoEngage refrained from using total branch conversions as the maximized metric because that would give preference to users doing multiple conversions on a branch and would also favour a branch having more action campaigns. Both necessarily would not mean that the branch that eventually wins is the most optimized one.
MoEngage tries to find the optimized branch based on the most common user behavior leading to conversion.
Maximizing engagement
Similar to the conversion calculation, MoEngage aims to optimize the total engaged trips instead of total engagement for each branch. To do this, MoEngage gives an engagement score to each branch and tries to maximize that score instead of total clicks or email opens of the branch.
The engagement score for the branch is the ratio between the number of branch trips that have fetched a click from at least one of the action campaigns (BTC) + one-tenth of the number of branch trips that have only fetched email opens (BTO) and Total Branch Trips. Engagement Score= (BTC+ 1/10(BTO))/Total Branch Trips
Help Us The engagement score allows MoEngage to optimize for the best-performing branch, which might not necessarily be the branch with the most clicks or most emails open.If this approach does not suit your use case or expectation, contact MoEngage at product@moengage.com.
Maximizing the combination of engagement and conversion
For each branch, MoEngage combines Branch CVR and Engagement score with emphasis on branch CVR because fetching conversions is more important than generating engagement. So the Maximized metric, in this case, becomes Branch CVR + 1/10( Engagement Score).
Merlin AI distributes maximum users toward the best-performing branch when the user reaches the path optimizer stage.
Keep ExperimentingMerlin AI does not stop distributing users towards low-performing branches even when a branch is identified to be performing better as the best-performing branch may keep changing from time to time as more and more users go through the flow.
Each flow branch’s performance keeps updating on the flow canvas as more and more users go through it.
A branch can be identified by its branch name. Email is the branch name in the above example.
4 done is the number of branch trips in the example above, and these 4 trips constitute 66.67% of the total trips reaching the Intelligent Path Optimizer stage.
Convertrip trip count is the number of branch trips that lead to at least one conversion from one of the action campaigns in the branch.
The Engagement score of the branch is also shown for each branch, regardless of which metric is maximized for the branch.
All action stages, from an Intelligent Path Optimizer stage to the flow exit stage, are included in tracking the branch’s performance.
If the branch splits into multiple branches, the resulting branches are considered part of the parent branch.
If a branch merges into another branch at some point, the common flow path is considered a part of both the merging branches.
You can analyze and compare the branches’ performances in the report for your split node. You can click the report option that appears on hovering over the stage on canvas or go to the split node stats section in the flow’s detailed stats page and filter on a particular split node from the list.
All MoEngage AI products, which were previously referred to as Sherpa AI, are now known as Merlin AI.
The five frequently used stages are displayed under the Frequently Used section on the All stages tab. The list of available stages changes dynamically based on your workspace. Click the required stage to add it to the flow.
The selected stage is added to the canvas.
