---
name: Rating prompt
slug: rating-prompt
category: pattern
status: published
created: 2026-08-21T00:00:00.000Z
modified: 2026-08-21T00:00:00.000Z
definition: An in-product request to rate or review, usually timed to a moment
  of success and often filtered so only positive responses reach the public
  store.
aliases:
  - name: review request
    source: community
  - name: app rating prompt
    source: community
  - name: review gating
    source: community
  - name: rate this app
    source: community
  - name: how are we doing
    source: community
tags:
  - onboarding
  - platform-registers
relations:
  contrastWith:
    - microsurvey
  variantOf: []
  partOf: []
  seeAlso:
    - nagging
implementations: []
sources: []
demo: inline
exhibit: false
useWhen: the app asks how you like it right after something went well
---

A rating prompt is a request for public credit, and its design is almost entirely a
question of timing. Asked at random it collects the average feeling of everyone who
happens to be holding the app, which is roughly the truth. Asked immediately after
something worked, a route saved, a transfer cleared, a photo exported, it collects the
feeling of a person who has just succeeded, which is a good deal better than the truth.
Nobody has lied. The sample has simply been taken at the top of the wave, and the store
listing that results is a picture of the best moment the product produces.

The second move is the one with a rule attached. Many prompts do not ask for a rating at
all, they ask a screening question first, then route the answer: the reader who says they
are enjoying it is handed the store's own review sheet, and the reader who says they are
not is handed a private feedback form that goes to the team and nowhere else. That is
review gating, and both major app stores forbid it. Apple's review guidelines require that
the rating request use the system API, which cannot be conditioned on the answer to a
preceding question, and Google Play's policy on ratings and reviews prohibits deterring or
discouraging negative reviews. The private form is not the offence; funnelling one branch
into it while the other branch reaches the public record is.

The honest version keeps the diagnosis and drops the filter. Ask the screening question if
it is useful, then offer the same two doors to both answers: everyone can leave a public
review, everyone can send private detail, and the reader chooses which they want rather
than being sorted. Ask once, at a moment the reader can connect to something they did, and
record the answer so it is not asked again, because a rating prompt that keeps coming back
has stopped being a request and become [nagging](/nagging). Watch the copy on the refusal
too: "no, I don't care about supporting free apps" is
[confirmshaming](/confirmshaming) bolted onto a survey.

None of this makes the pattern suspect on its own. Ratings are how a store ranks anything,
and a product with no reviews is invisible, so asking is legitimate and asking well is
craft. What turns it into a [dark pattern](/dark-pattern) is the gate: the moment the
interface's interest in a high average and the reader's interest in an accurate one come
apart, and the interface quietly decides which readers get to speak.