In this article8 sections · 6 min
Why small businesses are the easier target
Many owners assume fake-review attacks only happen to big brands. They do not. A smaller business is often the easier target, because it has fewer reviews to absorb the hit, and a rating that slips even slightly can decide whether a customer calls you or the business listed below you. Knowing how to recognize an attack is now part of protecting the business, the same way you protect the shop front.
Telling an attack from a bad stretch
Real unhappy customers leave fingerprints. They name a date, a product, a staff member or an order number — something you can look up. A coordinated attack usually cannot, because the person writing was never inside your business. Look for these signals together, not one at a time:
- Several low ratings landing inside the same afternoon or overnight.
- Reviewer accounts with no photo, no history, or a burst of reviews created the same week.
- Complaints that stay vague, with no order, date or name you can trace.
- Repeated phrasing or oddly similar word choices across unrelated accounts.
- Glowing reviews for one particular competitor sitting in those same profiles.
- A sudden drop that starts right after you won a contract, opened a location or moved above a competitor in search.
One of these proves nothing. Several together is a case. Motive is not something you can show a moderator; pattern is.
The first day: move before the evidence does
Attackers edit or delete the moment they notice attention, so the first hours matter: a review that changes before anyone has recorded it is evidence that cannot be got back. Get the profile in front of a specialist quickly — the free audit records what is there while it is still there — and keep your own booking and job records for the dates described within reach.
Two things not to do in this window. Do not argue in public replies — a defensive owner is the reaction the attacker wanted. And do not have staff go after the reviews from their own accounts. Coordinated flagging looks exactly like the fake-engagement pattern the policy prohibits, and it gives a moderator nothing new to act on.
The first week: one case, not six complaints
This is where most businesses lose without knowing why. They flag six reviews one at a time. Each is scanned by an automated check, each reads on its own as an ordinary opinion, and all six survive.
Handled properly, they are one connected incident. A fake review rarely violates a policy in its wording. It violates the policy in its origin — the timing, the account histories, the services that were never delivered — and the origin only becomes visible when the reviews are placed side by side. Putting that picture together is specialist work, and it is what our team does for you.
A first refusal is common, and it is where most owners stop. Our team keeps it in hand until Google reaches a final decision. Google decides at every stage; what we control is how complete the picture is when it gets there.
Is legal action worth it?
Sometimes, and rarely as the first move. A letter from a lawyer to a competitor you can identify can end an attack quickly when the evidence is tidy and the conduct is obvious. The problem is identification. Most attackers use throwaway accounts, and unmasking them means asking the platform for account data through a court, which costs money and takes time.
Courts also treat opinion generously. A review that calls your business unprofessional is an opinion; a review that accuses you of a specific false fact is a different matter. The practical route for most businesses is a cease-and-desist letter to a competitor you can name, sent alongside the removal case rather than instead of it. Reserve anything more for an attack that is sustained, measurable in lost business and clearly attributable. Get advice from a lawyer in your own jurisdiction before relying on any of this.
Rebuild the rating while the case runs
Cases take as long as the platform takes. Your average does not have to wait. An average is arithmetic: a handful of low ratings moves a profile with fifty reviews a long way and a profile with several hundred very little. The fix for the number is volume from genuine customers, and it starts on day one.
Ask at the moment the customer is happiest — usually right after the job finishes or the order arrives, not weeks later in a marketing email. Make the link one tap. Ask every week rather than in one panicked burst. A profile with a steady inflow also makes the next attack less effective, because the same six reviews barely register against a deep history.
Making the profile a harder target
- Turn on review alerts so a new review reaches your inbox in minutes, not weeks.
- Keep the profile complete and current — photos, posts, answered questions — so it reads as monitored.
- Note your rating and review count weekly, so you can show exactly when a drop began.
- Keep transaction records tidy enough to produce proof for any date on request.
If this has happened once, assume it can happen again, particularly before your next busy season.
What this comes down to
- Pattern is evidence. Motive is not.
- Act in the first hours, before reviews are edited or deleted.
- Treat the cluster as one incident, handled together through to a final decision.
- Legal action fits a sustained, identifiable attack, not one anonymous post.
- Genuine review volume restores the number faster than any removal can.
If you are in the middle of one of these now, every policy-violating review becomes an open case, followed through to the platform’s final answer, while you run the business. The audit is free and it tells you what is realistically removable before any case is started.
About this guide
Written by the Erase Trax review team: specialists in review policy, working with businesses in the United States, the United Kingdom, Canada, Australia and the United Arab Emirates. Each guide is written against the platforms’ own published policies. Last reviewed .
- We never need your login
- We never buy, write or incentivize reviews
- The platform makes the final decision
