All posts
PlaybookJul 8, 2026 · 10 min

When You Should NOT Reply to an App Review

Should you reply to every app review? No. Here are the cases where a reply cheapens the moment, feeds a troll, or creates real legal risk.

RA

The Argus Team

Reply Argus

No, you should not reply to every app review. Reply to most of them: the data is one-sided that responding lifts ratings and rescues unhappy users. But there's a small, specific set of reviews where a reply does nothing useful, cheapens the moment, or hands you real risk. Knowing which is which is the difference between a reply program that reads as human and one that reads as a bot dutifully stamping every card.

This is the contrarian half of the advice nobody gives. Everyone tells you to respond; almost no one tells you when silence, or a private channel, is the sharper move. Here's the judgment framework.

So should I reply to every app review?

Start from the default, then learn the exceptions. The default is yes-ish: replying is usually the right call and the numbers back it hard. Google reported at I/O 2019 an average lift of 0.7 stars for developers who respond, and Hassan et al., studying over four million reviews, found users were roughly six times more likely to raise their rating after a developer replied (4.4% versus 0.7%). If you want the full case for the default, [does replying to app reviews raise your rating?](/blog/does-replying-to-app-reviews-raise-your-rating) walks the studies.

But "usually" is not "always," and the studies measure averages, not every individual review. The reviews where a reply backfires don't show up in an average lift; they show up one at a time, as a reader who thinks "why did the company argue with a grieving customer in public," or a five-star fan who feels processed by a form letter. The goal isn't a 100% reply rate. It's replying where a reply helps and having the discipline to skip, delay, or take private the ones where it doesn't.

The reviews where a reply hurts

Five patterns where the reflex to respond works against you. Read them as a filter you run before you hit publish, not as reasons to go quiet in general.

  • The troll fishing for a reaction — a review written to provoke, not to be resolved. Replying is the oxygen it wants. A calm one-liner can be fine, but arguing feeds a public fight that other shoppers will scroll straight to.
  • The happy five-star a canned reply would cheapen — genuine warmth answered with "Thanks for your feedback! We're glad you're enjoying the app 🎉" reads as a mail-merge. Either say something specific and real, or let the five stars stand on their own.
  • The legal-risk complaint — accusations of a data breach, injury, medical harm, a chargeback, or anything a lawyer might touch. A public reply can be read as an admission and becomes a permanent, discoverable record. Route these to a private channel.
  • The one already being handled in support — if the same user has an open ticket, a public reply duplicates the conversation and can contradict what support just told them. Close it where it started.
  • The policy violation you should report, not answer — spam, hate, an off-topic ad, a competitor's plant. Flag it to the store instead of dignifying it. Sometimes a short neutral reply plus a report is right; a full engagement rarely is.

Legal-risk reviews: this is not legal advice, but the reflex is

If a review alleges harm, a breach, or names a specific dispute, do not litigate the facts in public. A reply that says "that never happened" or "we did issue your refund" can confirm details, waive a defense, or read as an admission later. Acknowledge briefly if at all, invite them to a private support email, and loop in whoever handles legal before you type anything specific. The public record is forever and it is discoverable.

The five-star that a form reply makes worse

This one surprises people. A genuine, specific five-star review is a small gift, and answering it with a template is like replying to a handwritten thank-you note with a printed slip. The reader who left it can tell. So can the next shopper reading your reply history, who now sees the same three cheerful sentences under twelve different reviews and quietly downgrades how real your team feels.

The rule: if you reply to a five-star, say something only that review could earn. Name the feature they called out. If you can't, a reply adds nothing, and there's no penalty for letting a happy review sit there being happy. Replying to positives is optional in a way replying to a fixable one-star is not.

The offline mode saved me on a 14-hour flight to Tokyo. Whole trip planned with no wifi. Chef's kiss.

Reply

Fourteen hours with no wifi is exactly the case we built offline mode for, so this genuinely made our week. Safe travels, and if you hit anything in Tokyo where offline could go further, tell us.

That reply works because it could only have been written for that review. Swap it for "Thanks for the 5 stars, we're so glad you love the app!" and you've spent a real moment on nothing. Specificity is the whole point, and it's why blasting a template across every positive review is worse than replying to none.

Does ignoring a troll actually work?

Mostly, yes. A review engineered to bait you wants a reply, and a public back-and-forth is a stage you're handing over for free. The reader you're actually writing for is never the troll; it's the next hundred shoppers. And a measured non-answer, or a single unbothered line, tells them more than a paragraph of defense ever could.

The nuance: if a nasty review buries a real, answerable claim in the bile, a short factual correction aimed at the audience can be worth it, once, without escalating. The test is who benefits. If your reply informs future readers, publish it. If it only satisfies your urge to win, close the tab. When a review crosses into a policy violation, [reporting it](/blog/how-to-respond-to-negative-app-reviews) is the lever, not a comment war.

A 20-second decision framework

Before you reply to any review, run it through four questions. Most clear all four in a second and get a reply; the exceptions are exactly the ones that should give you pause.

  1. 1

    Step 1 — Is there a real person to help or reassure?

    A fixable bug, a confused user, an honest complaint: reply. This is the meat of the value and where the rating lift comes from. If yes, you're almost certainly replying; keep going only to check for landmines.

  2. 2

    Step 2 — Could a public reply create legal or PR risk?

    Harm, breach, injury, a named dispute, anything a lawyer might care about. If yes, stop. Move to a private channel and get the right person involved before you write a public word.

  3. 3

    Step 3 — Would my reply be generic?

    If the only reply you have is a template that fits any review, especially on a five-star, skip it or make it specific. A form letter under a genuine review costs you more than silence does.

  4. 4

    Step 4 — Am I replying to help readers, or to win?

    For trolls and bait, be honest about the motive. If the reply serves future shoppers, keep it short and publish. If it only serves your ego, don't. Then report the review if it breaks a store policy.

Not replying is not the same as not watching

Skipping a reply is a decision about a review you've already read. It is never an excuse to stop monitoring. The legal-risk complaint still needs a private response, the troll still needs a report, and the five-star you let stand still told you what users love. Silence is a choice, not a blind spot.

Encoding the judgment at scale

Here's where doing this by hand falls apart. The framework is easy for ten reviews and impossible for a thousand a month across two stores in six languages. You can't eyeball every one for legal keywords, template-worthiness, and troll-bait, which is why so many teams default to either blasting templates at everything or going silent.

This is the judgment you can encode. [ReplyArgus](/features) drafts an on-brand reply for every review, grounded in your past approved replies and in the reviewer's own language, but publishing stays approve-by-default. You turn on [rule-based auto-publish](/blog/is-it-safe-to-auto-publish-app-review-replies) only where you trust it: let clean five-stars in a language you speak go out with a light, specific touch, while anything matching a legal keyword, a one-star, or a flagged language routes to a human first. The exceptions in this post become rules, not things you hope to remember at 11pm on a Friday.

Frequently asked

Should I reply to every app review?
No. Reply to most of them, since responding reliably lifts ratings and recovers unhappy users, but skip or take private the small set where a reply backfires: trolls fishing for a reaction, happy five-stars a template would cheapen, legal-risk complaints, issues already handled in support, and policy violations you should report instead.
Is it bad to reply to five-star reviews?
Only if the reply is generic. A specific, warm reply to a five-star is great and reinforces a happy user. A copy-paste "Thanks for your feedback!" under a genuine review reads as a mail-merge and makes your team look automated. If you can't say something only that review could earn, it's fine to let the five stars stand.
Should I reply to a review that threatens legal action or alleges harm?
Not in public, and not with specifics. A public reply about a breach, injury, or dispute can be read as an admission and becomes a permanent, discoverable record. Acknowledge briefly if at all, invite the person to a private support email, and involve whoever handles legal before writing anything factual. This is general guidance, not legal advice.
Does ignoring a troll review actually work?
Usually. A review written to provoke wants a reply, and a public argument is a stage you hand it for free. Write for future shoppers, not the troll: a short unbothered line, or silence, reads as confidence. If the review breaks a store policy, report it rather than engaging.
Does not replying to a review hurt my rating?
Not replying to a specific review where a reply would backfire won't hurt you. What hurts is a low overall reply rate on the fixable complaints, since those are where responses recover ratings. The strategy is high coverage on the reviews a reply helps, and disciplined restraint on the few it doesn't.
Can I automate which reviews get a reply and which don't?
Yes, with rules. Tools like ReplyArgus keep publishing approve-by-default and let you opt into auto-publish by rating, keyword, language, or store, so clean five-stars can go out automatically while one-stars and legal-keyword reviews route to a human. That turns your judgment into a filter instead of a thing you re-decide for every review.

Replying to reviews is one of the highest-leverage things a developer does, which is exactly why the exceptions matter: a bot that stamps every review, grieving customer and delighted fan alike, undoes the trust the good replies build. Keep your coverage high, keep your judgment about the handful where silence or a private channel wins, and let the rest be human. [Start free with ReplyArgus](/signup), no card required, and Argus drafts a reply to every review while you decide, by rule, which ones actually get published.

Try it

Let Argus draft your next reply.

Watch it answer a real review in your voice. 10-day trial, no card to begin.

See the features or pricing.

Keep reading