# Notification prompt after sending a message: 5-10% opt-in vs 1-2%

- Outcome: Won
- Page type: Retention (App)
- Elements: Paywall design
- Test run by: [LinkedIn](https://liftarchive.com/practitioners/linkedin/)
- Commentary by: LiftArchive (our analysis, not theirs)
- Original source: https://growth.design/case-studies/linkedin-retention-triggers
- Page: https://liftarchive.com/test/gd-linkedin-contextual-notification-prompt/

LinkedIn asks for notification permission right after a user sends a message, offering to alert them when that person replies. LinkedIn's product team reported 5-10% opt-in for this prompt, against 1-2% for generic ones, about 5x. The case study was published by Growth.Design.

## Result

**about 5-10% opt-in vs 1-2% for generic prompts (about 5x)**

- Notification opt-in: about 5-10% for the contextual prompt vs 1-2% for generic prompts, roughly 5x.
- Growth.Design's headline rounds this to +500%. Treat it as a range, not an exact lift.
- No figures on retention, replies or later engagement are shared.

## Control

Generic notification permission requests that could appear at any time, such as at login, promising alerts in general terms.

## Variant

Right after the user sends a message, a prompt names and shows the person they wrote to and asks a simple yes/no question about being notified when that person replies. The system permission dialog comes after.

## Why it likely won (our reading)

- Timing: the user has just put effort into a message and wants to know when the reply arrives, so the ask fits what they want right now.
- Naming a real person likely feels more concrete than a vague promise of alerts.
- The ask is framed as a benefit (knowing when someone replies) rather than a feature list of sounds and badges.
- A yes/no question feels like a normal conversation and may be easier to agree to than a system "Allow" dialog.

## Caveats

These are LinkedIn product-team figures relayed by Growth.Design, with no sample size or test design given. It may not have been a split test. The 1-2% baseline comes from a different prompt shown at a different moment, so timing and copy are mixed together.

## How to run this test

- Hypothesis: If we ask for push permission right after a user does something that has a natural follow-up, and name that follow-up, then more users will opt in because the value of the notification is clear at that moment.
- Primary metric: Push notification opt-in rate (users who allow / users shown the prompt)
- Guardrail: Notification opt-out or disable rate
- Guardrail: Day-7 and day-30 retention
- Guardrail: System dialog denial rate
- Who should test it: Apps where users take actions that lead to a later event: messaging, social, marketplaces, delivery, habit and fitness apps. It helps less if your notifications are mostly promotional, or if there is no clear moment when an alert serves the user.

---

Credit: LinkedIn ran this test and owns the result. Summary and commentary by LiftArchive (https://liftarchive.com/test/gd-linkedin-contextual-notification-prompt/).
Usage: Free to use, quote and build on, with credit and a link: name the practitioner or organization that ran the test and link to its LiftArchive page. The results belong to the practitioners who ran and shared them; the summaries and commentary are LiftArchive's.