Re-engagement

Re-engagement is the work of bringing back users who installed your app, or customers who bought from you before, before they drop off for good. The usual channels are ads aimed at past users, push notifications, email and in-app messages.

What is re-engagement?

Every app decides for itself what “inactive” means: 7 days without opening the app, a cart left for 24 hours, or a paying subscriber who stopped renewing. Re-engagement goes after exactly that group. It is a different job from user acquisition, where cost is counted per install.

The channels fall into two groups:

• Owned channels: push notifications, email and in-app messages. They cost nothing in media spend, but users have to allow notifications or still have the app installed.
• Paid channels: ads aimed at people who have installed the app (see retargeting). These reach people who have turned notifications off, too.

Re-engagement on the major platforms in 2026

Google Ads. App campaigns come in three types: for installs, for engagement and for pre-registration (Google Ads Help, About the different types of App campaigns, accessed October 2026). App campaigns for engagement target people who already have the app and run on Google Search, YouTube, Google Play and the Display Network. They are used to win back people who have not used the app in a long time, or who installed it but never opened it (Google Ads Help, About App Campaigns for engagement, accessed October 2026).

iOS. AdAttributionKit can measure a re-engagement when someone who already has the app taps an ad. Only taps count; views do not. Since iOS 18, the app showing the ad can pass along a universal link that opens the intended screen directly. The device has 2 days to send its first conversion value update after a re-engagement, and the number of re-engagements recorded is capped per month for each app and per year for each device (Apple Developer, Receiving ad attributions and postbacks, accessed October 2026).

MMPs. Measurement platforms call this reattribution: a user gets assigned a new source if they were inactive long enough and then came back after interacting with an ad. Adjust’s default is 7 days for both the inactivity period and the click-based reattribution window, and users who came from an ad are not reclassified as organic (Adjust Help Center, Reattribution, accessed October 2026).

Measuring results with a holdout group

Re-engagement campaigns can easily take credit for people who would have come back anyway. A more reliable way to measure is to keep a holdout group that gets no ads and no push messages, then compare return rates between the two groups. This approach is known as incrementality testing.

Example (hypothetical)

A shopping app has 200,000 users who have not opened it in 14 days. The marketing team holds back 20,000 of them as a control group. The other 180,000 receive ads and push messages for 2 weeks, at a cost of VND 120 million.

• Campaign group: 9% came back and made a purchase.
• Control group: 6% came back on their own.
• Incremental lift: 3% of 180,000 = 5,400 people.
• Cost per user brought back by the campaign: 120,000,000 / 5,400 ≈ VND 22,200.

Going by platform reports alone, the campaign would claim all 16,200 returning users, which works out to about VND 7,400 each, only a third of the real cost.

Re-engagement vs. retargeting vs. retention

Frequently asked questions

Does re-engagement need deep links?

It should have them. An ad about a particular product has to open that product’s page in the app, not the home screen. See deep linking.

How long should a user be inactive before you try to re-engage them?

That depends on how often people use the app. A ride-hailing app might treat 14 days as a long gap, while 3 months can be normal for a travel app. Take the average number of days between sessions for your loyal users and use that as the threshold.

If you want to build a win-back campaign with a holdout group from day one, talk to the Adtek team through our contact page.