"Example Auto Repair" is a fictional business. The format below shows the evidence report Ghost Reviews prepares; every name, date, and review is invented for illustration.

Evidence report — Example Auto Repair

Prepared by Ghost Reviews for the business owner · Private — prepared for you only, not for publication.

Summary

3 ratings flagged, posted between March 14, 1:38 a.m. and March 15, 1:51 a.m.
Pattern: two textless one-star ratings 26 minutes apart in the middle of the night, from accounts with no other reviews, followed by a vague no-specifics one-star the next night.
Each flagged rating is matched below to a relevant Google policy it may implicate, with reasoning.
These are documented signals, not verdicts. Google alone decides every removal.

What we looked at (and what we didn't)

Source: public Google reviews for the business — a deep pull of the most recent 100 reviews, focused on negatives. Method: a pattern screen plus a human read of every flagged review. We used no private data, and we make no claims about who any reviewer is or why they posted.

Rating timeline

Oct 2025 – Feb 2026 · 4.6★ average · 11 reviews · no 1★

Mar 14, 1:38 a.m. · ★ 1 — no text — no account history

Mar 14, 2:04 a.m. · ★ 1 — no text — no account history (26 min later)

Mar 15, 1:51 a.m. · ★ 1 — "Low quality work. Avoid." — near-zero history

Mar 16 – Apr 2026 · back to 4–5★ reviews from established accounts

The visual that matters: months of a normal pattern, then a one-night anomaly, then normal again.

The flagged ratings (3)

Rating 1 of 3T—— R——★☆☆☆☆March 14, 2026 · 1:38 a.m.

Review text: — none —
Account history: No other publicly visible Google reviews
Cluster context: Posted 26 minutes before Rating 2
Official Google policy — exact quotation: Fake engagement. “Fake engagement is content that does not represent a genuine experience.” Read Google's policy. Checked July 27, 2026.
Reasoning: A textless rating from an account with no other publicly visible review history, posted at 1:38 a.m. inside a same-night cluster, is one documented signal. Taken together with Ratings 2 and 3, the pattern is consistent with potentially coordinated inauthentic activity, but it is not proof and Google makes the decision.

Rating 2 of 3K—— M——★☆☆☆☆March 14, 2026 · 2:04 a.m.

Review text: — none —
Account history: No other publicly visible Google reviews
Cluster context: Posted 26 minutes after Rating 1
Official Google policy — exact quotation: Fake engagement. “Fake engagement is content that does not represent a genuine experience.” Read Google's policy. Checked July 27, 2026.
Reasoning: The second textless one-star of the night, again from an account with little publicly visible review history, arrived 26 minutes after the first. The timing and account-history pattern are documented signals that can be consistent with coordinated activity, but they are not proof on their own.

Rating 3 of 3J—— P——★☆☆☆☆March 15, 2026 · 1:51 a.m.

Review text: "Terrible shop. Low quality work. Avoid."
Account history: One other review (posted the same week)
Cluster context: Third 1★ in a ~24-hour window
Official Google policy — exact quotation: Fake engagement. “Fake engagement is content that does not represent a genuine experience.” Read Google's policy. Checked July 27, 2026.
Reasoning: The only worded review in the cluster names no service, vehicle, staff member, or date. That lack of detail is not proof of a policy violation; together with the timing and near-zero account history, it is a signal worth presenting to Google for its decision.

Filing these yourself (free, if you'd like)

Every flagged review can be reported to Google directly from your own profile, free, using the policy reason matched on its card. Two honest expectations if you go that route: Google decides on its own timeline, and it may answer an initial report with "no violation found." That decision can be challenged through Google's separate review-appeal process, which has its own evidence requirements and timelines. Preparing and tracking that follow-through is part of our optional filing service.

The full report maps every flagged rating to the policy it may implicate, with the evidence. We never hold the evidence hostage; what we sell is the drafting, the filing, and the follow-through.

One honest limitation

Google says no to some well-evidenced challenges, and nobody outside Google can predict which. Anyone who guarantees removals is not telling the truth.

This report identifies signals, not verdicts. Removal decisions are Google's alone. Want this for your own profile? Start with the free scan. Free evidence-report requests are currently paused; the scan remains available when capacity is open.