Restaurant Review Management: One Guest, Three Review Clocks

Restaurant review management is usually explained as if it were a Google problem with a side of Yelp. For a restaurant that also takes delivery orders, it is not. The same guest, on the same night, can rate the food in a delivery app, rate the room on Google, and rate your ordering app in an app store. Three surfaces, three sets of rules, three response windows, and almost always three different people inside the business who never compare notes. This is the version of the playbook that starts from that fact, and it applies whether the review lands on Google, the App Store, Google Play or a delivery app.

Why restaurant review management is not just Google review management

A single-site restaurant with delivery typically collects feedback in at least three places. Google Business Profile carries the public rating that shows next to your name in Maps and in the local pack. The delivery apps carry per-order ratings that decide how your store looks inside the app where the order is actually placed. And if your group ships its own ordering app, the App Store and Google Play carry a rating that most restaurant marketers never look at until it falls below four stars.

These are not tiers of the same thing. They answer different questions for different people at different moments. Someone deciding where to eat tonight reads Google. Someone already inside a delivery app, scrolling a list of twenty nearby kitchens, reads the in-app rating and never sees your Google profile at all. Treating the second audience as an afterthought means conceding the surface closest to the purchase.

The same structural point applies to any multi-surface business, which is why review management across locations and apps treats the platform inventory as the first step rather than the last. For a restaurant group the inventory is short but the ownership is messy: marketing owns Google, the delivery manager owns the apps, and the product team owns the app store listings.

The three response clocks, and what each platform actually allows

The most useful thing you can do before writing a single response is to write down what each platform permits and how long you have. These are not similar numbers, and a service level agreement that assumes they are will miss the tightest one.

On Google, replies are public and moderated. Google’s guidance on reading and replying to reviews, read on 22 September 2026, states that an approved reply "will be publicly posted under the customer’s review" and that it "will appear like your business replied, and your personal name won’t be shown". It also warns that replies "usually take up to 10 minutes to review, but sometimes a review can take up to 30 days". That moderation delay is worth knowing before you promise a manager that a crisis reply goes live immediately.

On DoorDash, the window is explicit and short. The DoorDash merchant help article on customer reviews, read on 22 September 2026, says merchants have "up to seven days to respond to new feedback, though responding within 24 hours is recommended", and that most customers leave feedback within 24 hours of their order. It also states that only verified DoorDash orders count and that "a maximum of one review per order is factored into your rating".

On Uber Eats, responses run through Uber Eats Manager and reach the customer directly. Uber Eats describes the feature, read on 22 September 2026, as responding to reviews from Uber Eats Manager or the mobile app, with the customer notified by email of the reply. That is a different shape of conversation from a public Google reply: it is closer to a service recovery message than to a statement for future readers.

• Google Business Profile. Public reply, moderated, usually visible in minutes but occasionally up to 30 days. Written for the next thousand readers, not only the reviewer.
• DoorDash. Seven days to respond, 24 hours recommended, one review per verified order counted toward the store rating.
• Uber Eats. Reply from Uber Eats Manager or the app; the customer is notified by email. Effectively a direct message.
• App Store and Google Play. Public developer responses on your own ordering app, usually owned by a product team rather than by operations.

The practical consequence is that a single blanket target such as "respond within 48 hours" is wrong in both directions: too slow for DoorDash, and too rigid for Google where a considered public reply beats a fast one. Build the target per platform, the way the review response time SLA playbook sets it for Google, then hold each owner to their own clock.

What Google permits you to ask for, and what gets reviews removed

Restaurants are the industry where incentives are most tempting, because the currency is cheap: a free dessert, a discount on the next visit, an entry into a draw. Google’s position on this is not ambiguous. Its prohibited and restricted content policy, read on 22 September 2026, states that merchants must not "Offer incentives - such as payment, discounts, free goods and/or services - in exchange for posting any review".

“Contributions to Google Maps should reflect a genuine experience at a place or business.”

— Google, Prohibited and restricted content policy, read 22 September 2026

The same policy is equally clear about what is allowed, and this is the sentence most restaurant owners have never read: Google permits you to "Solicit or encourage the posting of content that does represent a genuine experience, without offering incentives". Asking is fine. Paying is not. The line sits exactly at the incentive, not at the ask.

Two further restrictions catch restaurants specifically. Content "based on a conflict of interest" is prohibited, which covers reviews from current or former employees, so the staff-writes-a-five-star habit is a policy breach rather than a clever tactic. And posting "from multiple accounts by or at the request of one person" is fake engagement. A table of five phones passed around by a manager is the textbook example.

Routing unhappy guests to a private form while sending happy ones to Google is review gating, and it is a separate problem from incentives: it produces a rating that does not describe the restaurant. When a review genuinely breaches policy, the removal route is narrow and procedural, and reporting fake Google reviews for removal covers what actually works and what wastes a week.

Delivery app feedback is operations data, not reputation data

The biggest change a restaurant can make is to stop treating delivery ratings as marketing and start treating them as a kitchen and packing report. A one-star Google review about slow service is a reputation event. A run of delivery complaints naming the same missing item across a fortnight is a station problem, and no amount of well-written responses will fix it.

Because DoorDash counts one review per verified order, delivery feedback is tightly coupled to individual tickets in a way Google reviews never are. That coupling is the opportunity: every complaint can be traced to an order, a time, a station and often a specific menu item. Tag the feedback by cause rather than by star rating, and the pattern appears within a month.

1. Tag every delivery complaint with a cause: missing item, wrong item, temperature, packaging, portion, wait time, or courier issue.
2. Separate what the kitchen controls from what the courier controls, and stop counting courier delays against the kitchen’s scorecard.
3. Count causes weekly rather than reading reviews individually. One complaint is noise; the same cause eleven times is a line check.
4. Take the top cause to the pre-shift briefing with the actual guest wording, not a summary of it.
5. Re-count the same cause four weeks later. If it has not moved, the fix was a conversation rather than a change.

The same tagging discipline applied to public reviews is what review sentiment analysis is actually for: not a mood score, but a count of causes you can act on. Sentiment analysis that produces a number nobody changes their shift around is decoration.

A weekly operating rhythm that survives a busy service

Restaurant review management fails for a boring reason: it competes with service. Any process that requires a manager to open four dashboards during a Friday dinner rush will be abandoned by the second week. The rhythm has to fit the shape of the week instead.

Daily, one person clears the delivery queue, because that is the clock that runs in hours. Twice a week, a manager writes the Google replies, because those are read by future guests and deserve more than a rushed sentence. Monthly, someone counts causes and rating movement across every platform together. That cadence is realistic for a single site and scales to a group with the ownership model set out in multi-location review management.

For the actual wording, do not invent it during a rush. Keep a short set of approved openings and let the manager add the specific detail. Responding to negative Google reviews covers the structure that holds up in public, and the same structure works with a softer register for a direct delivery reply where only the reviewer will read it.

Groups running several restaurants should also decide, in writing, who may reply on behalf of a site. A regional manager answering for a kitchen they have not visited produces the generic apology that readers discount instantly. Delegate the reply to the site and keep the review of tone central. ReviewMankey’s response and approval workflow exists for exactly this split: drafts prepared centrally, approval and the local detail kept where the service actually happened, across Google, the App Store and Google Play.

The five numbers worth tracking for a restaurant

Most restaurant reputation dashboards track the star rating and nothing else, which is the one number that moves slowest and explains least. These five say more, and all of them can be counted without a tool.

• Response rate per platform. Not an average. Google and DoorDash will diverge, and the average hides which one you are neglecting.
• Median time to respond per platform. Median, not mean, so one holiday week does not distort the picture.
• Review velocity. New reviews per week per site. A rating that looks stable while review velocity collapses is a profile going quiet, not a profile doing well.
• Top three causes by count. Taken from your own tags, refreshed monthly, and compared against the previous month.
• Share of reviews mentioning a named staff member. The cheapest leading indicator of service quality a restaurant has, and the one guests volunteer without being asked.

If you want the wider context for why these matter, the review management statistics page collects the published consumer research on how reviews affect local decisions, each figure with its named source. Use it to set expectations with an owner, not as a substitute for your own numbers, which are the only ones that describe your restaurant.

Restaurants that get this right are rarely the ones with the most sophisticated tooling. They are the ones that decided who owns each surface, wrote down the clock for each one, and counted causes instead of stars. Everything else is easier once that is true. If you want the industry version of this with the platform set laid out, the restaurant review management use case covers how the pieces fit together across Google, delivery apps and app stores.