How to Reduce Churn with Effective Onboarding Campaigns

Hand drawing a green line on a printed journey map from one box to an exit box, showing a user who activates leaving a Braze onboarding Canvas.

To reduce churn with onboarding campaigns in Braze, treat the Canvas as a route to one activation event, not a sequence of messages. Set that event as the conversion and as the exit, branch the rest with Action Paths, and stop messaging users who have already activated. The sections below show how the documented Canvas mechanics support that design.

Key Takeaways

  • Choose one activation event and use it as the conversion event, the exit event and the Action Paths trigger.
  • A user who performs the exit event while in a Delay step stays in that step until the delay ends.
  • Total Exits by Exit Criteria updates only after a user has fully left, so count activation with the conversion event.
  • An Action Paths step sorts users by their first qualifying action, and the Everyone Else group is the natural silent path.
  • Match property names and data types between the entry event and the exit criteria, and test in a development environment.
  • Add a control group so you can see what the Canvas changed.

What should an onboarding campaign achieve if the goal is less churn?

An onboarding campaign has one job: get a new user to the first action that shows the product is useful to them. We call that the activation event. Every step in the Canvas should serve it.

Churn that starts in the first weeks is usually an activation problem, so measure activation and nothing softer. Opens and clicks show that a message was seen. They don't show that the user came back.

Braze documents conversion events as the way a Canvas measures its own efficiency. Name your activation event as the conversion event before you draft a single message.

How do you reduce churn with onboarding campaigns in Braze Canvas?

Build the Canvas around the activation event and let the messages follow. This is the order we use when we set one up as a Braze implementation partner.

  1. Choose one activation event and record it as the Canvas conversion event.
  2. Add the same event as an exception event in the exit criteria, so activated users leave the journey.
  3. Place an Action Paths step after the first message to split users who act from users who don't.
  4. Write a different second message for the silent path. A repeat of the first message adds volume and nothing else.
  5. Set re-eligibility and frequency capping in the Entry Audience window before launch.
  6. Add a control group so you can see what the Canvas changed.

Braze documents that exception events are added directly to the Canvas entry rules and remove users when they perform a specific action. The Entry Audience window is also where you control re-eligibility and frequency capping. Turning on rate limiting or frequency capping eases the marketing pressure on your users and helps you avoid over-messaging them.

Why does the exit event matter more than the message copy?

Exception events keep your messaging relevant. That matters for churn more than a clever subject line does.

A user who activates and then receives a get-started reminder learns that the programme isn't paying attention. That is a small signal, and it adds up across a whole cohort.

The exit event should be the same action you chose in step one. One decision, made once, drives measurement, targeting and suppression.

What happens to a user who exits while waiting in a delay?

Braze documents a detail here that changes how you read your reports. When a user performs the exit event, Braze marks them to exit straight away and they don't advance to later steps. If they're in a Delay step, they stay in that step until the delay ends.

The documented example uses a Delay step of 30 days. A user who performs the exit event on the first day is marked to exit at once but doesn't fully leave the Canvas until the delay ends, 29 days later. The Total Exits by Exit Criteria figure in Canvas analytics only updates after a user has fully exited.

So don't use the exit count as your activation count. Use the conversion event for that, and read both against the same time window.

MeasureWhat it recordsDocumented behaviour
Conversion eventUsers who performed the activation eventWindow runs three days beyond the Canvas's maximum duration; minimum window is five minutes
Exit criteria totalUsers who have fully left the CanvasUpdates only after the user has fully exited
Action Paths stepWhich group a user moved throughEvaluation window up to 31 days; first qualifying action sets the route

Braze also recommends a conversion window that at least matches the longest path in the Canvas. If your onboarding path runs for several weeks, a short window will undercount activation.

How do Action Paths separate activated users from at-risk ones?

An Action Paths step holds users for an evaluation window and sorts them by what they do. The default window is one day. You can adjust it from seconds to weeks, up to a maximum of 31 days.

With ranking off, a user advances through the group matching the first qualifying action they perform after entering the step. A user who does nothing falls through the default Everyone Else group at the end of the window. You can also tick a box on any action group so that its users exit the Canvas when the window ends.

We treat the Everyone Else group as the at-risk group. These are users who received the first message and did nothing in the window you chose. They get the second message, and the activated group gets nothing more.

Don't change the ranking toggle after launch while users are already in the step. Braze documents that the path state recorded earlier in the window may reflect the previous setting. Build a new step instead.

How do you match exit criteria to the entry event?

Braze documents a pattern for exits that depend on the event that started the Canvas. Its example is an abandoned checkout Canvas, where a user exits only when they purchase the exact item they abandoned. A user who buys something else keeps receiving the reminders.

The same pattern fits onboarding wherever the entry event carries a property, such as the plan or product a user chose. The user should exit when they complete the action for that choice, not for any choice.

This only works if the property names and data types in the entry event match those in the exit criteria. Braze says to verify this and to test every component in a development environment before launching.

How long should the gaps between onboarding messages be?

Braze lets you extend a Delay step up to two years (730 days). Its own onboarding example adds an extended delay before a message that nudges users who haven't started a session. The documentation gives you the range and not the answer.

We start with short gaps in the first days, when intent is highest, and widen them as the user stays silent. The widening is a judgement, so test it against a control group instead of copying our shape.

Braze also documents that changes to a user's attributes or contact information during a delay are reflected when they advance. A user who fixes their profile mid-journey can therefore receive a message on a channel they didn't have at entry. Plan the copy with that in mind.

What does an activation-first Canvas look like in practice?

Take a hypothetical subscription app. This is an illustration of the design, not client work and not a result from any programme.

The Canvas enters users when they sign up. The activation event is the first time they complete the core task of the product. The first step is an in-app message, because Braze recommends building in-app messages for onboarding Canvases to get a strong first impression.

An Action Paths step follows, with an evaluation window long enough to cover a normal first session and the day after. Users who complete the core task go through a group set to exit. Everyone Else receives an email that names the one task they haven't done, with no general product tour.

After a further delay, users still silent get one last message. They can then move to a separate retention Canvas through a Send to Destination step, which the Braze release notes list as generally available. The exit criteria carry the same activation event throughout, so nobody who activates mid-delay moves on to a later step.

Three decisions carry this design. The activation event is chosen once, the silent path gets different content, and the handoff goes to a separate Canvas with its own goal. If you change any of the three, the report you read afterwards changes meaning.

How does this connect to the rest of the programme?

Onboarding exit logic is one part of a wider retention system. Our guide to Braze onboarding best practices on entry, exit and QA covers the launch checks that sit around this design.

Measurement is the next question. Our post on key onboarding metrics in Braze explains what to track once the activation event is set. For feedback from users who didn't activate, see using feedback surveys in Braze to improve onboarding.

When onboarding ends, the work moves to retention. Read reducing churn with retention automation tactics in Braze for the next stage.

The timing of those later messages is covered in triggering retention messages at the right time. For users who have already left, see building win-back campaigns in Braze.

Two further pages sit beside this design. Our Braze services page describes how we build behaviour-triggered journeys, and our case study on interactive re-engagement for lapsing users covers one approach to the retention stage.

What should you check before you launch?

Run through a short list before the Canvas goes live. Test in a development environment, as Braze advises, and never in production.

  1. Confirm the conversion event, the exit event and the Action Paths trigger all name the same activation action.
  2. Check that event property names and types match between entry and exit.
  3. Set the conversion window to cover the longest path.
  4. Add a control group.
  5. Remember that Canvas steps move users only forward, so there's no loop back to an earlier message.

As a Braze implementation partner, CustomerIK recommends deciding the activation event first and writing the messages second. Teams that reverse that order tend to ship a Canvas that reports well on opens and says little about churn. This post doesn't cover consent or unsubscribe handling, so check those settings against your own policies before launch.

Frequently Asked Questions

1. What is an activation event in a Braze onboarding Canvas?

It's the first action that shows a user has found value, such as completing the core task of your product. The Canvas pages we read don't define it, so you choose it. The Canvas then uses that event as its conversion and its exit.

2. Why do users stay in a Canvas after they perform the exit event?

Braze marks the user to exit straight away, but a user in a Delay step stays in that step until the delay ends. They don't move on to later steps. Exit counts in analytics update only after they've fully left.

3. How long can an Action Paths evaluation window be?

The default is one day. You can adjust it in seconds, minutes, hours, days and weeks, up to a maximum of 31 days. The limit may change, so check the current documentation.

4. Can a user take two different paths from one Action Paths step?

With ranking off, no. The first qualifying action a user performs after entering the step sets their route. A second matching action doesn't move them to another group.

5. Should I send onboarding messages to users who have already activated?

Our recommendation is no, and the exit criteria are how you enforce it. Braze documents exception events as the way to remove users when they perform a specific action.

Sources

CustomerIK is a Braze implementation partner working across onboarding, technical integration, marketing operations and customer data management.

If you're deciding which action should count as activation in your Canvas, let's talk.

Preferences

Privacy is important to us, so you have the option of disabling certain types of storage that may not be necessary for the basic functioning of the website. Blocking categories may impact your experience on the website.

Accept all cookies
Accept all cookies

These items are required to enable basic website functionality.

Always active

These items are used to deliver advertising that is more relevant to you and your interests.

These items allow the website to remember choices you make (such as your user name, language, or the region you are in) and provide enhanced, more personal features.

These items help the website operator understand how its website performs, how visitors interact with the site, and whether there may be technical issues.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.