Quick answer
An in-app message is a message shown inside an app while the person is using it, as a banner, modal, card or image. It needs no notification permission but reaches only people who open the app, so it suits guiding active users rather than bringing back lapsed ones.
What is an in-app message?
An in-app message is a message shown inside your app, on its own screen or as an overlay, while the person is using it. Unlike a push notification, which arrives outside the app on the lock screen or notification shade, an in-app message only appears when the app is open. That is both its strength (the person is already there and paying attention) and its limit (it cannot reach someone who has not opened the app).
Typical uses are announcing a new feature, guiding a first-time user to the next step, asking for a rating at a good moment, and offering an upgrade when someone reaches a limit. Related terms are in-app notification, in-app popup and in-app campaign; the word "message" usually means a designed piece of UI, not an operating-system notification.
Types of in-app messages
Messaging tools give you a small set of formats. Firebase In-App Messaging, for example, offers four built-in templates, which are a good map of the common types:
| Type | What it looks like | Best for | Watch out for |
|---|---|---|---|
| Banner | A slim strip at the top or bottom | Light news, status, reminders | Easy to ignore |
| Modal | A centered dialog with a title and a button | One clear action, such as try the new feature | Interrupts what the person was doing |
| Card | A larger layout with image, text and two buttons | A choice: accept or later | Needs good copy for both buttons |
| Image-only | A single designed image | Campaign art, seasonal offers | Text is baked into the image, so it is harder to localize |
Outside those, many apps also use tooltips and coach marks (pointing at a button), full-screen takeovers (for major launches) and inline cards inside a feed or settings page. The rule of choosing: the more it interrupts, the rarer it should be and the more important the message.
In-app message vs push notification
| Question | In-app message | Push notification |
|---|---|---|
| When can it appear? | Only while the app is open | Anytime, even with the app closed |
| Who can receive it? | Anyone who opens the app | Only people who allowed notifications |
| Permission needed? | No | Yes, for example on Android 13 and higher |
| Best for | Guiding, announcing and nudging active users | Bringing back users who are not in the app |
| Main risk | Interrupting a task | Being switched off for good |
The permission point is documented. On Android 13 (API level 33) and higher, a newly installed app's notifications are off by default until the app requests the POST_NOTIFICATIONS permission and the user grants it, and the user can revoke it at any time. An in-app message needs none of that, because the user is already in your app.
A reach example
Sample data: an app with 10,000 monthly active users wants to announce a new feature during one week. The rates below are illustrative.
| Step | Push | In-app message |
|---|---|---|
| Starting audience | 10,000 monthly active users | 10,000 monthly active users |
| Can receive it | 62% have notifications on: 6,200 | 70% open the app in that week: 7,000 |
| Actually delivered or shown | 95% delivered: 5,890 | 85% pass targeting and are shown: 5,950 |
| Acted on it | 8% open: about 471 | 12% tap: 714 |
| Share of all users who acted | 4.7% | 7.1% |
The two channels reached almost the same number of people (5,890 and 5,950), but they are different people: push can reach someone who has not opened the app for weeks, and in-app messages cannot. In this sample the in-app message wins on engaged users, and push wins on lapsed ones. That is the decision rule: use in-app to talk to people who are here now, and push to bring back people who are not. Run a lapsed-user message as push and a feature announcement as in-app.
How to write and time a good in-app message
- One goal per message. One screen, one button. Tell the person what they get and what to tap.
- Trigger on behavior, not on time. Show the "try saved filters" message after a person applies the same filter for the third time, not on the first launch.
- Cap the frequency. For example, at most one message per session and one campaign per week. Remember that people dismiss what interrupts them.
- Make dismissing easy. Modal and image-only messages should have a visible close control; Firebase's Modal and Image-only templates include an X button.
- Ask for permissions in context. Android's guidance is to request the notification permission when its purpose is clear, such as after a user taps an alert bell or follows an account. An in-app message that explains the benefit before the system dialog is a common way to do it.
- Define the success metric before sending. For example: share of shown users who use the feature within 7 days, compared with a group who did not see the message.
Common mistakes with in-app messages
- Showing it at launch. A message that covers the first screen is dismissed before it is read. Wait for a moment of natural pause, such as after a task is completed.
- Stacking campaigns. If three teams each send one message a week, a user sees three interruptions. Keep one shared calendar and a cap per person.
- Targeting everyone. A message about a feature that 40% of users already use should go only to the 60% who have not. Use a rule such as "did not use the feature in the last 30 days".
- No holdout group. Without a group that does not see the message, you cannot tell the effect of the message from what would have happened anyway.
What to measure
| Metric | Formula | Sample |
|---|---|---|
| Impression rate | shown ÷ eligible | 5,950 ÷ 7,000 = 85% |
| Tap rate | taps ÷ shown | 714 ÷ 5,950 = 12% |
| Feature adoption | users who used the feature ÷ users shown | measured over 7 days |
| Dismiss rate | closed without action ÷ shown | rising values mean annoyance |
| Retention effect | retention of shown vs holdout | compare over 30 days |
Tap rate alone can mislead: a very prominent message will have high taps and low value. Always pair it with what happened afterwards.
Measuring in-app messages with TapCub
Imagine you built the feature announcement above in your own messaging tool and now want to know whether it worked.
- TapCub does not send in-app messages or push. It measures. Send an event such as
inapp_shownandinapp_tappedfrom your app with the SDK, with the message name as a property. - Build a funnel. In All-platform analytics › Behavior › Behavior analysis, make a funnel from
inapp_showntoinapp_tappedto the feature's own event, so you see where people drop. - Compare shown with not shown. In All-platform analytics › Users › Segments & tags, create a rule segment for "did
inapp_shownin the last 7 days" and one for "did not". Apply them in analysis to compare feature use and retention. - Hand the segment on. If a webhook is configured, TapCub emits segment membership changes, which your own messaging tool can use to decide who gets the next message.
- Check app health. After a campaign, App analytics › Quality & attribution › Stability shows whether a new screen coincides with a drop in the crash-free share.
Sources
- Firebase Documentation, What can you do with Firebase In-App Messaging?.
- Android Developers, Notification runtime permission.
Was this explanation clear?
We use your answers to fix and expand this page.
Couldn’t send. Please try again in a minute.
Thanks for the feedback. We’ll use it to improve this page.You’ve already sent feedback on this page. Thanks.

Data Analyst
Explores traffic sources, conversion funnels, and retention, using clear metric definitions and concrete examples to explain what the numbers mean.