First, establish which reviews went and when
Reviews cannot be diagnosed from the total alone. Use whatever record exists, a monitoring tool's export, a spreadsheet log, notification emails or screenshots, to list the reviews that are now missing with their authors, dates and star ratings. If no record exists, ask the team and regular customers which reviews they remember, and start a log today. Then note when the reviews disappeared and what else happened around that date: a review campaign, a new request method, a profile edit, a merge, a move or a re-verification.
The pattern points to the cause. Reviews removed in a group soon after a campaign suggest filtering. Reviews from named staff or relatives suggest conflicts of interest. A scattering of single reviews, some negative, suggests reviewers acting on their own. Reviews that vanished on the day of a profile change suggest the profile. The reviews and reputation guide sets out the rules for asking; this page uses them only to judge whether past asking explains the loss.
Diagnosis table
Diagnosis
| Symptom | Check | Likely cause | Fix |
|---|---|---|---|
| Reviews gathered through an offer or a selective request disappeared together | How those customers were asked and whether anything was offered | Reviews filtered as incentivised or solicited against the rules | Stop the scheme; switch to a plain request to every customer |
| Removed reviews came from employees, relatives or business partners | Authors of the missing reviews against staff and family names | Conflicts of interest | Stop internal reviews; rely on independent customers |
| Single reviews vanished at different times, with no pattern | Whether the authors' profiles still exist; contact with the customers concerned | Reviewer deleted the review or the account | Nothing to reverse; keep a steady flow of new reviews |
| Many reviews vanished on the day the profile changed | Dates of merges, moves, re-verification or duplicate removals | Profile merged, moved or re-verified | Check which profile holds the reviews; request help with evidence |
| Reviews from a recent push are missing or never appeared | Number of requests sent and their timing against the usual weekly flow | A burst of reviews triggering spam filtering | Spread requests over time from each customer's own device |
Fix, verify and prevent each cause
Asking in ways the rules do not allow
The help centre prohibits incentivised reviews and selective solicitation, and consumer regulators such as the Federal Trade Commission and the Competition and Markets Authority treat incentives that are not disclosed and the suppression of negative feedback as consumer protection issues. Reviews gathered through a discount, a prize draw or a request sent only to happy customers are the ones most likely to be filtered.
Fix: stop the scheme at once, and replace it with the plain request to every customer described in the reviews and reputation guide; do not try to replace the lost reviews quickly.
Verify: new reviews from the plain request stay on the profile over the following weeks.
Prevent: have every review request approved against the guide's rules before it goes out.
Reviews written by people close to the business
Staff, owners, family members and business partners have a conflict of interest, and the platform's rules prohibit reviews from them; regulators treat reviews that do not come from genuine customers as misleading. Well-meaning support from people close to the business is still a breach.
Fix: ask staff and family not to review the business, remove any that can still be removed by their authors, and make sure no one writes reviews on customers' behalf.
Verify: the profile's remaining and new reviews all come from people who were customers.
Prevent: put a line in staff guidance explaining why reviews from employees and their families are not allowed.
Reviewers removing their own reviews
A review belongs to its author, who can edit or delete it, and accounts that are closed take their reviews with them. Scattered single losses over time, with no link to a campaign or profile change, usually have this explanation. Nothing is wrong with the profile.
Fix: accept the loss, and where a specific customer is known, ask politely whether the review was removed deliberately; never pressure anyone to repost. Reply to existing reviews and keep asking every customer, as the reviews guide describes.
Verify: losses remain occasional and the monthly flow of new reviews outweighs them.
Prevent: track the review count per month in the log, so occasional deletions are seen in context rather than as a crisis. The opposite symptom, a burst of negative reviews arriving together, has its own diagnosis for a sudden wave of negative Google reviews.
A profile change that moved the reviews
Merging duplicates, moving a profile to a new address, re-verifying or removing a listing can change which reviews appear where; the citations and NAP guide recommends keeping the profile that holds the reviews as the survivor of any merge for this reason.
Fix: check whether the missing reviews now sit on another listing, such as an old-address profile, and if a merge went the wrong way or reviews were lost in a change, gather the dates and evidence and use the help route the Business Profile help centre provides.
Verify: the reviews appear on the surviving profile, or the help request has a recorded outcome.
Prevent: before any merge, move or removal, note the review count on each listing and choose the survivor deliberately.
A burst after a campaign
A sudden rush of reviews, especially from one place, one device or one day, looks unnatural, and the reviews guide notes that the help centre documents review stations and bulk solicitation as causes of removal. Reviews from a burst may be removed or never published.
Fix: stop collecting reviews on a shared device in the premises and stop sending requests to a whole backlog at once; move to a request sent to each customer after their own visit or job.
Verify: reviews from the new routine appear and stay, at a steadier rate.
Prevent: keep requests tied to individual service moments, so volume rises only as fast as the business serves customers.
Prevention: a steady, rule-compliant asking routine
The protection against losing reviews is a flow of genuine reviews arriving at the pace the business actually serves customers. Ask every customer, through their own device and account, at the moment the job is done, with a direct link and nothing offered in return. Never collect reviews in the shop, never route unhappy customers elsewhere, and never let staff or family review the business. Keep a log of every review, with its author, date and rating, so any loss can be dated and matched to an event. Before merging, moving or re-verifying the profile, record review counts on every listing. Report the rolling count and rating each month as a reputation indicator (KPIA metric tied directly to an objective.), alongside replies sent, because the recent figures are what customers read on the results page (SERPThe page of results a search engine shows.). A Business Profile (GBPLocal listing on Google Search and Maps.) built this way loses the occasional review without noticing; one built on a single campaign can lose a season's work in a day. The local presence section and the review steps of the Business Profile tutorial set out the wider routine.
A salon loses a third of its reviews in a week
A hair salon with 156 reviews finds 49 gone in one week of October. The log shows that 44 of them arrived during a two-week September promotion in which clients were offered a 10% discount on their next visit for leaving a review, most typed on the salon's tablet at the till. Another 4 came from the owner's sister and two stylists. The remaining loss is a single review whose author closed their account. The owner ends the promotion, retires the tablet, and asks staff and family not to review. Clients now receive a text with the review link the day after an appointment. Over the next four months the profile gains 61 reviews, none removed, and the rating settles at 4.7 from 168 reviews.
Common mistakes
Appealing filtered reviews instead of changing the asking method. Offering anything in return for a review. Asking staff and relatives to help. Collecting reviews on a device in the premises. Merging profiles without noting which holds the reviews. Keeping no record of reviews, so losses cannot be dated.
