ReviewProof guides
By the Ingenious AI team · Published 14 November 2026 · Reviewed 14 November 2026
Run the free review audit See pricing
What these guides cover
Every guide below answers one real question about proving reviews are real: how ReviewProof imports reviews and computes its risk signals, what a provenance certificate and an evidence pack actually show, how the verified-experience widget decides what it may display, and how the FTC fake-review rule (16 CFR Part 465) shapes day-to-day review practice for local businesses and agencies.
Signals and detection
- How can you prove a review is real? — A review is provable when it carries a verified-transaction link, a public provenance certificate and a chain of custody. Here is how to check and publish that proof.
- How do you detect fake reviews in your own profile? — Fake-review detection on your own data: gating, incentives, duplicates, bursts and AI wording, each traced to the imported row that triggered it.
- What is review gating, and how do you stop doing it? — Review gating means steering happy customers to public reviews and unhappy ones to a private channel. Why it is prohibited, how it reads in your data, and how to fix the ask.
- What counts as an incentivised review? — Discounts, gift cards, free items and refunds offered for reviews: what counts as an incentive, how it appears in review text, and how to audit for it.
- What does duplicate review text tell you? — Identical review text appearing more than once: copy-paste campaigns, shared scripts and honest coincidences — and how duplicate detection separates the questions.
- What is a review burst, and when is it innocent? — A burst is an unnatural pile-up of reviews on a single day. How ReviewProof detects it, the innocent causes to rule out, and what to document when it is not innocent.
- Can you tell if a review was written by AI? — What AI-written reviews tend to sound like, what the signal in ReviewProof actually measures, and why it is labelled a pattern, never a proof.
- How is the ReviewProof risk score calculated? — The exact formula, the weights for gating, incentives, duplicates and AI signals, the burst addition — and how to read the number without misusing it.
- What is a reviewer fingerprint? — A salted hash of a reviewer's handle and source — pooled across businesses as counts only — that spots repeat flagged reviewers without exposing anyone's identity.
- What is the difference between a verified and an UNLINKED review? — Verified means a transaction link sits behind the review. UNLINKED means none was imported — the review stays in the audit but out of the verified widget.
FTC and evidence
- What does the FTC fake-review rule actually ban? — 16 CFR Part 465 in plain language: fake and AI-generated reviews, bought sentiment, insider reviews, suppression and fake independence — and the record that answers each one.
- Why does keeping negative reviews protect your business? — Deleting, gating or burying negative reviews is suppression — a violation in itself. How a kept-negatives record becomes your strongest evidence instead.
- What should an evidence pack for a review dispute contain? — For a platform dispute, an FTC question or due diligence: the rows, the signals, the transaction links, the moderation log — assembled as one exportable record.
- What is a chain of custody for a customer review? — Imported, transaction-linked, analysed, moderated — every event logged, nothing silent. How a review's chain of custody turns a claim into a record.
- What is a review provenance certificate? — The public page behind a review: rating, source, date, transaction reference, moderation state and chain of custody — what it shows, and what it refuses to claim.
- What does transaction-backed review provenance mean? — Attach a controlled transaction reference to a review — hash and masked form only, raw reference never stored — and publish a result anyone with the reference can verify.
- How do you audit your own reviews, step by step? — Gather the rows, import, read the five signals, check the links, moderate with a log, export the pack — the complete self-audit a business can run on itself this week.
- How should you moderate reviews without suppressing them? — Kept, published or flagged internally — every decision logged, nothing silently deleted. A moderation model that survives scrutiny.
Workflow
- How do you import reviews into ReviewProof by CSV? — The columns the importer reads, the verified flag and transaction reference, plan caps per import — and the file habits that make the analysis honest.
- Paste import or file upload: which should you use? — Both routes parse identically. The real differences are volume, error-checking and habit — how to choose per situation, and when each one bites.
- How do you read a ReviewProof risk report? — Totals, flagged count, risk score, the four signal counts, burst status, verified share and average rating — what each line means and what to do about it.
- How does the verified-experience widget decide what to show? — Only reviews that are both transaction-linked and kept or published. How the widget rule works, what it excludes, and why the exclusions are the feature.
- How should you use AI-drafted replies to reviews? — Rule-based reply drafts that ship with a disclosure label, differ by rating, and must be reviewed before sending. How to use them without deceiving anyone.
- How do you ask customers for reviews without breaking the rules? — The safe ask has three properties: same invitation to every customer, same public destination regardless of sentiment, nothing of value attached. Wording that works.
- How do you write a review-request email that does not gate? — Subject lines, the one-link rule, where the private feedback link may sit, and the phrases that quietly convert a request email into a gating machine.
- How do you audit reviews across multiple locations? — Per-location files, per-location scores, one shared standard: why group averages hide the one location that needs help, and how the Agency workflow runs.
- How can an agency offer review audits to clients? — A deliverable a client can verify: their file, their signals, their evidence pack. Scoping, running and handing over provenance audits without mixing clients' records.
- How do you use the reviewer-fraud graph in practice? — Pooled fingerprints, flagged-elsewhere counts, per-merchant analysis: the monthly graph check that catches travelling reviewers your own file cannot see.
- What should a monthly review-provenance routine include? — Thirty minutes a month: import the period, read five numbers, work the flagged rows, check the graph, export the pack. The routine that keeps proof current.
- How does the AI summary of your reviews work? — A deterministic extract over verified reviews — count, average, frequent themes — labelled AI-generated wherever shown, and built from nothing else.
- How do you respond to a negative review you can prove is real — or isn't? — Acknowledge, do not argue facts in public, move specifics offline — and let the provenance record, not the comment thread, carry your evidence.
For your type of business
- How should a restaurant prove its reviews are real? — Table-side asks, till-point incentives and opening-week bursts: the restaurant-specific patterns, and the booking-linked proof that answers them.
- How should a dental clinic handle reviews and provenance? — Anxious patients, front-desk asks and treatment-linked records: soliciting fairly in a clinic, and keeping proof that never touches clinical detail.
- How should a hotel prove its reviews came from real stays? — Checkout asks, review scores in staff targets, and seasonal bursts: tying reviews to stay records and keeping the proof across seasons and systems.
- How should a salon keep its reviews honest and provable? — Chair-side asks, rebooking offers tangled with reviews, and stylist-level pressure: a sentiment-blind ask and appointment-linked proof for salons.
- How should a plumber or trades business prove its reviews? — Doorstep asks, call-out stress and job numbers: linking reviews to completed jobs and removing the ask from the doorstep entirely.
- How should a law firm handle client reviews and proof? — Matter references without matter details, solicitor-level asks, and the insider and confidentiality traps specific to legal practice.
- How should an auto repair shop prove its reviews? — Counter asks while the customer pays, courtesy-car sweeteners, and one-star disputes about bills: invoice-linked provenance for repair shops.
- How should a gym keep member reviews honest? — Free-pass sweeteners, transformation-story pressure and January bursts: membership-linked reviews and sentiment-blind asks for fitness businesses.
- How should a real estate agent handle reviews? — Only the closed deals get asked, testimonials from friends at launch, and six-figure trust: tying agent reviews to completions without outcome selection.
- How should an online store prove its product reviews? — Post-delivery emails, discount-for-review inserts and review imports from marketplaces: order-linked provenance when volume is high and contact is zero.
- How should a cleaning service keep its reviews provable? — Recurring visits, cash jobs, keys-in-hand trust and team members entering homes: visit-linked reviews and asks that never happen at the door.
- How should a veterinary clinic handle reviews with care? — Distraught clients, end-of-life sensitivity and outcome-skewed asks: appointment-linked proof and solicitation designed for a clinic's hardest days.
Problems and triage
- What should you do if a competitor is leaving fake reviews? — Document the cluster, check your own house first, dispute through the platform with evidence — and never retaliate. A triage order that holds up.
- What should you do in the first 48 hours of a one-star attack? — Preserve everything, import fast, separate the real from the fabricated, answer once — the hour-by-hour discipline for a sudden negative pile-on.
- How do you clean up a profile with bought reviews in its past? — Stop the supply, import everything as found, flag with the truth logged, rebuild on linked reviews. Recovery that survives the next audit, not just this one.
- Why are employee reviews of your own business a violation risk? — Staff, owners, contractors and family reviewing without disclosure: why insider reviews breach the FTC rule and how an audit surfaces them.
- What does it mean when one reviewer appears at many businesses? — Active genuine reviewers exist — and so do review farms. Presence, flags, links and wording: the four readings that separate a traveller from a stable.
- What should you do when a customer threatens a bad review? — A refund demanded against a one-star review: why paying is the worst option, what to document, and how the provenance record strengthens your hand.
- Can having too few negative reviews look suspicious? — A 5.0 with hundreds of reviews reads as filtered to buyers, platforms and investigators. What a believable spread looks like, and how it is earned honestly.
- How do you explain a seasonal spike in your reviews? — Peaks that follow your trading calendar are defensible — if the evidence is filed while the season is happening. What to keep, and what a spike that is not seasonal looks like.
- Are reviews from family and friends a problem? — They mean well, they really visited — and they are still undisclosed insiders. Why the favour fails the FTC test and what to build instead.
- How do you audit and clean up years of old reviews? — Historical imports split by era, solicitation judged by the standards of its time, corrections dated from today. Handling an inherited review history honestly.
- Why is a star rating without provenance not enough? — Stars summarise opinion; they cannot show which transactions happened, how reviews were solicited, or what was suppressed. The claim each layer can and cannot carry.
- How should you handle reviews from refunded customers? — A refunded customer can still review — the transaction happened. Separating remedy from review, linking the row honestly, and the trade you must never offer.
Trust and growth
- How do verified reviews turn into bookings? — Provable reviews work at the moment of doubt — booking pages, quotes and proposals. Placing proof where the decision happens, and measuring honestly.
- Where should you show verified reviews on your website? — Booking pages before homepages, certificates one click deep, the widget uncurated: a placement guide for proof that has to survive being clicked.
- How does provable review proof help win agency pitches? — Differentiation a prospect can click: bringing an audit, not adjectives, to the pitch — and the promises a provenance agency must never make.
- How do reviews hold up in due diligence for a sale or franchise? — Buyers discount unverifiable reputations. Era maps, linked shares and dated packs: assembling the review section of a data room that prices your profile as an asset.
- Why audit your reviews before spending on advertising? — Ads send strangers to your reviews at the exact moment they decide. A pre-campaign audit fixes the profile that spend will be judged against.
- How do you build a review-evidence habit across a whole team? — One owner, one rhythm, written rules where the work happens: turning provenance from the founder's project into the team's ordinary practice.
- How do you train staff to never gate reviews? — Most gating is improvised kindness by untrained staff. The one-session training, the replacement phrases, and the audit feedback loop that makes it stick.
- What review system should a new business set up first? — The first fifty reviews set the pattern forever. The day-one setup: standard ask, insider rule, import rhythm, widget — before the first shortcut can harden.
- How can a customer verify that your reviews are real? — The checking journey from the customer's side: widget to certificate to presented reference — and how to invite the check without sounding defensive.