Mobile app reputation management: your replies are half of it

Mobile app reputation management, also called app store reputation management, means looking after your app's public record on the App Store and Google Play: its rating, reviews and the replies you publish underneath. Customers write the reviews. You write the replies.

App reputation is the rating, the review text and the replies: the replies are the half a team writes. Our 2026 pilot study of 100 top Google Play apps on US charts found two separate gaps. Apps tended to answer everything or nothing. Publishing replies did not mean answering the review. This guide is about managing both.

Coverage is a decision; quality is a second one.

Key takeaways

• App reputation is ratings, review text and replies. The replies are the half your team writes. Read them as closely as you read complaints.
• Coverage and quality need separate decisions. In our 2026 pilot study of 100 top Google Play apps on US charts, apps tended to answer everything or nothing. Setting up replies did not ensure useful answers.
• A reply count cannot show what you published. In our 2026 pilot study of 100 top Google Play apps on US charts, 42.3% of all replies in the 30-day window, to reviews of any star rating, repeated another reply from the same app exactly.
• Write a standard and require evidence. Measure a pass rate, make wrong facts and wrong language automatic failures, and check each language separately.
• Read quality beside rating movement on replied reviews. Google describes how responses can help ratings (Play Console). Movement alone cannot tell you whether the reply was accurate.
• Diagnose a rating drop before replying. The guides linked below cover recovery, rating prompts and store mechanics in full. This guide is about how you check the replies you publish.

In September 2026, Den Markov published a breakdown of the rubric Google's app quality audit reportedly applies to Google Play apps. He describes a score called perceived quality from reviews. It grades review text rather than the rating, looking at recurring complaints and praise. It also considers the gap between recent sentiment and the historical rating, and whether replies are personal or templated.

The write-up's case-study app had a 100% reply rate: every review had a reply. It scored medium on reviews, and the author notes that its replies were templates.

The reviews score in the reported Google Play audit, and the templated case study at Medium.

Google has not published this rubric.

Its focus on review text fits the stores' changes over the past two years. That gives teams another reason to look beyond the rating when managing app reputation.

What app reputation management covers: reviews, ratings and your developer replies

App reputation management covers your app's public record on the App Store and Google Play: ratings, reviews and the replies you publish underneath. It sits alongside app store optimization. ASO works on the listing users see before installing, including the title, keywords and screenshots. Reputation management works on the record they read beside it. The two connect because ratings and review sentiment affect search ranking and install conversion.

Our 2026 pilot study of 100 top Google Play apps on US charts shows why this needs separate decisions. Of 76 apps with at least 30 low-star reviews, 24 answered none and 18 answered 90% or more. Replying looked like a yes or no choice. A team had set up a way to answer reviews, or it had not.

Quality is the next decision.

Counting replies cannot tell you whether they name the complaint, use the right language or state a confirmed fact. Customers write the review half of the record; your team writes the reply half. Each reply stays under its review unless you edit or delete it.

Review analysis remains the foundation: find recurring complaints, pass them to the people who can fix them, and use confirmed facts in your answers. Then read what you published.

The study covers Google Play, with apps drawn from US charts that skew large. It cannot establish the pattern for the App Store or show cause and effect. It gives us a reason to measure both coverage and quality in our own work.

Reply rate on Google Play: most low-star reviews get no answer, and many answers say nothing

Most low-star Google Play reviews get no answer, and a reply can still leave the complaint unanswered.

In our 2026 pilot study of 100 top Google Play apps on US charts, 11.5% of low-star reviews with real text had a reply a week later. That coverage gap comes before speed: a faster workflow cannot help a review that never reaches someone.

But turning replies on does not settle quality. Repeated text can fill a reply count while leaving the reviewer's question open. Track whether complaints reach your team, then check what the answers say. The 2026 app review reply benchmark covers the method, app and game breakdowns, language gaps and limits. Use it to compare coverage and quality separately.

Four ways a review response fails, ranked by how much damage it does

Replies fail through wrong facts, wrong language, broken team instructions or poor hygiene, in that order of severity. That is how support leads tend to rank the damage. Their question is less "what is our average score?" and more "did we publish something that embarrasses us?"

A wrong fact is the worst failure. The reply invents a refund window, mentions a feature missing from that platform, or gives a plausible but false explanation of how the app works. It is rarely a deliberate lie. It is a confident guess, but a reader has no way to tell the difference.

Wrong language comes next. An English reply under a Turkish review suggests you did not read it. A Polish reply with two Czech words is harder to catch: it looks close enough to pass a quick check. But it reads as broken to a native speaker.

A reply can also break your team's instructions. These might say never admit fault, never promise a date, route billing to the help center, or sign off with the team name. Write those rules down. Breaking one may not make the reply wrong for the reviewer. But it can create a problem your legal or support lead finds six months later.

Hygiene comes last: wrong store names in Google Play replies, missing signatures, replies four times as long as the review, or leftover template markers. These errors are cheap to catch and fix, but still make a public reply look careless.

Reply detail with the offending sentence highlighted in the published reply, the failed criterion named critical beside its reason, and the checks that passed listed underneath

Share of graded replies that answer the review, by star rating.

In our 2026 pilot study of 100 top Google Play apps on US charts, 73.8% of graded replies to one to three star reviews answered what the reviewer raised. This is an AI grade, not validated against human raters. In an internal check, a person disagreed with it on 1 reply in 10, so read that share as an upper bound. Quality shares describe the graded replies, not all of Google Play.

A reply being present was no guarantee it addressed the complaint.

A passing reply answers the specific review. Here is a two-star review of a mobile game from the pilot. It is paraphrased with small details changed so the original text cannot be found through search:

It's the usual build-a-base strategy game, like ones I've played many times already. It kept my interest for ten hours or so. The action side missions are good fun.

Here is the reply, also paraphrased:

Glad the action side missions worked for you. We understand that the base-building side felt like something you'd played before and lost its pull after ten hours or so. The weaker parts of the game are on our list, and we'll keep working on them.

The reply names both the praise and the complaint in the reviewer's terms. It accepts the criticism without arguing and would not fit a different review unchanged. Run that paste test on your own replies.

Averages hide problems that are obvious in individual replies. A high pass rate tells you little about the replies that failed. Read those replies to understand and fix the failures. Your checks should make them easy to find.

Some checks need no model. Plain rules can check the store name, signature or phrases your policy forbids. Checking whether a reply answers the concern, invents a commitment or stays in the reviewer's language needs a system that reads the text. Whatever method you use, make wrong facts and wrong language automatic failures. A weighted average can hide one critical defect behind five things that went well.

Make each verdict quote the words it judged. "This reply mixes languages" asks you to trust a claim. Highlighting two Czech words inside a Polish sentence gives you evidence to inspect, even if you cannot read either language. If a check cannot point to specific text, record it as unmeasured, never as a pass or a failure.

You can start these checks by hand. Write five or six checks that define a good reply for your team, and mark the critical ones. Read your last ten replies in a row. If you cannot tell them apart, neither can a reader. Grade twenty against your list and record the pass rate as your baseline.

Sort failures by what broke, with wrong facts at the top. Within each group, prioritise replies under one-star reviews. Fix them in that order: an invented fact under a one-star complaint is both likely to be read and hard for a reader to forgive. For graded examples of replies to one-star complaints, see how to respond to negative app reviews.

How to measure review response quality and its effect on your rating

Measure reply quality as a pass rate against your written standard, and measure reply effect as rating movement on replied reviews below five stars.

Reply quality asks whether the reply met your standard, with evidence quoted from its text. Reply effect asks whether the review's rating moved after you replied. Google describes the link between responding and rating updates (Play Console). Our Google Play pilot measured coverage and reply text; it cannot show that a reply caused a rating change. Keep that limit when reading your own results too.

Measure upward movement on replied reviews below five stars, where there is room to improve. A review base dominated by five stars distorts that comparison.

Compare movement after AI replies with movement after human replies. Read the result alongside quality to judge whether automation is helping reviewers as well as saving time.

AppReply Performance dashboard showing reply effect for AI vs human, automation score, and review coverage

Read both measures together to decide where to look. If quality rises and ratings move, keep going. If quality rises while ratings stay flat, check whether your standard measures what matters to reviewers. If ratings move while quality falls, read the replies to see what helped and which rules they broke. Either measure alone leaves you without enough context to act.

The app reputation KPIs worth tracking, and the ones that lie

Track low-star reply coverage, reply speed, repeated replies, quality pass rates, unmeasured checks, stale replies and reply effect. The rating gives you context but cannot explain the work by itself; the table shows what each KPI tells you and the trap to avoid.

Coverage and quality need separate rows because one can improve while the other fails.

The top three rows cover day-to-day operations. The next four check reply quality, including gaps hidden by an average. The last measures the outcome. Read that outcome alongside quality so you can see both what changed and whether the replies met your standard.

App store rating drops and review bombing: read before you reply

When your app store rating drops, diagnose the cause before you reply.

A release regression clusters around a version; an outage hits versions together. A pricing change shifts complaints toward money. Review bombing brings a burst of similar reviews unrelated to app quality. Read a short window and line it up with releases and service incidents.

Reply with the confirmed cause and next step, then update affected replies when a fix ships. Report policy violations through the store, while answering genuine complaints within the wave.

A higher reply count is activity, not evidence of recovery. Read app store rating recovery for the full diagnosis, store options and ways to measure whether ratings move after useful replies.

Asking for app ratings: in the app, and in the reply only when the review is happy

Ask new users inside the app. A reply may ask for a rating only when the review is clearly happy, and never ask for five stars.

Praise that also names a problem is still a complaint to answer.

Use the store's native prompt after a successful task. Do not screen for happy users before showing it or offer anything in return. Keep a support route available without making it decide who can rate. For an existing happy review, thank the reviewer for the specific praise before a light request to update the rating.

Google's advice is to avoid that request, so keep it rare. The guide to asking for app reviews covers prompt timing, APIs, quotas and gating rules.

App Store vs Google Play: how developer replies differ on iOS and Android

App Store replies follow storefront country through App Store Connect; Google Play replies follow review language through Play Console. Google Play limits replies to 350 characters, so the confirmed fact and next step need to fit without filler.

Both stores let you edit a published reply.

On Google Play, the date shown is the review's latest edit, so an old answer can end up beneath new words. Read changed reviews again before judging the reply. These differences shape where you measure coverage and how you check quality. Compare storefronts on iOS and review languages on Android. Use how to reply to app store reviews for console steps, permissions, notifications and worked examples on both stores.

Where AppReply fits in mobile app reputation management

AppReply supports app reputation management through app review management: analysing reviews, writing replies and checking reply quality. We built it around both halves of that public record. Coverage gets the reply published; a separate quality check asks whether it meets your standard.

Review analytics covers the reviews: recurring complaints and praise, per app and per language, with alerts when a topic spikes.

Reply Quality covers the replies. It grades what your team publishes on the App Store and Google Play against a scorecard you write. It does this whether an automation or a person wrote the reply. Review analytics and Reply Quality cover App Store and Google Play apps only; neither is available for Samsung Galaxy Store or Huawei AppGallery apps.

Every verdict quotes the words it judged. Failures enter a queue grouped by what broke, with the worst at the top. Reply Quality tracks brand voice per language against your team's own past replies. With a scorecard on, it checks AI drafts before publishing and grades published replies afterward. Nothing is graded until you turn it on.

Reply Quality is available on the Pro and Enterprise plans. The Reply Quality launch post walks through it in detail.

MAX, our reply agent, helps prevent those failures. Before it writes, it reads what the reviewer raised and searches the replies your team already approved. It checks the answer against your live help docs and release notes. Then it drafts in the reviewer's language across more than 100 of them. MAX is included on Pro, and AppReply handles 100,000+ replies a month (observed, automatic replies).

Performance dashboards show whether ratings moved on replied reviews, split by AI and human replies. Read that result alongside reply quality to close the loop. For the full setup, see our AI auto-replies guide. For the basics of ratings, store rules and collecting more reviews, read App Store Reviews 101.

A reputation management plan you can run weekly

Run a weekly app reputation management routine that combines replies, quality checks and rating movement. Give unanswered complaints and failed replies separate places in that routine.

The study found gaps in both; clearing one queue does not clear the other.

Daily: answer new low-star reviews and read reviews behind a spike. Weekly: grade twenty published replies against your written standard, then inspect flagged replies in order of severity. Monthly: compare pass rate and reply effect per language with last month, and re-read replies under reviews edited since. Quarterly: revise the standard, using failed replies to find missing or unclear rules.

Start this week without buying anything. Read ten of your replies in a row. Compare the languages you publish in with those someone at your company can read. Grade twenty replies against a written standard. That baseline will tell you more about the replies shaping your app's reputation than the rating on your listing can.