What is a review burst, and when is it innocent?
By the Ingenious AI team · Published 14 November 2026 · Reviewed 14 November 2026
Run the free review audit See pricing · See a sample provenance report
Short answer
A review burst is a concentration of reviews on one day that is out of proportion to the profile's whole history. ReviewProof flags a burst when at least five reviews share a date and that day holds at least 40% of the import, and adds 15 points to the risk score. Bursts have innocent causes — openings, events, a first bulk import of old reviews — and bought causes; the transaction links and wording around the day tell you which you are looking at.
What is a review burst?
A review burst is a timing anomaly: a single day contributing a share of reviews that normal, uncoordinated customer behaviour would essentially never produce. Timing is powerful evidence precisely because it is hard to fake retrospectively — a bought package delivered over a weekend, a staff push at a launch event, or a genuine one-off surge all leave the same-shaped spike, and the surrounding record is what separates them.
Your own import can manufacture a false burst
The first burst most businesses see is an artefact: they import a spreadsheet in which a past platform export stamped every historical review with the export date, or a migration assigned one day to years of reviews. Before concluding anything, check whether the burst day is a data date rather than a review date — ReviewProof analyses the dates in your file, so the file's honesty about dates is the analysis's honesty. Correct the source data where you legitimately can, note the artefact in your records, and re-import. A burst that survives clean dates deserves the harder look below.
ReviewProof computes every signal from the rows you import, never from a reputation guess about your business. The risk score is 100 × (gating × 1.0 + incentive × 1.0 + duplicate × 0.8 + AI-signal × 0.6) ÷ (2.6 × number of reviews), plus 15 points when a burst is detected, capped at 100. Each flagged review in the report names the imported row that triggered the signal, so a human can open that row, read it in context, and decide. The product labels these patterns signals for human review, not verdicts — and that restraint is deliberate, because a heuristic that overclaims would be its own kind of false evidence.
The innocent surge and the bought surge look different up close
Open the burst day's rows. A genuine surge — an opening, a festival weekend, a mention in local media — shows varied wording, mixed ratings including some mediocre ones, spread across sources, and transaction links you can match. A bought surge shows the opposite cluster: uniform five stars, similar or duplicate phrasing, AI-generation wording patterns, handles with no other trace, and UNLINKED rows throughout. The burst signal itself only opens the door; the rows behind it, which the report puts one click away, are where the verdict — yours, human, documented — actually happens.
After the first audit, the habit that matters is cadence. Re-import on a regular rhythm — monthly suits most locations — so new reviews enter the provenance record while the underlying transactions are still easy to match. Each import logs provenance events, moderation starts from kept (no suppression), and the certificate for any review shows its chain: imported, transaction-linked or not, analysed with its signals, moderated with its decision. A record built gradually like this is worth far more in a dispute than one assembled in a panic the week the dispute arrives.
Mistakes that undo the work
Panicking over an artefactual burst wastes the audit's credibility inside your own team — verify the dates first. The opposite mistake is explaining away a real burst as 'a busy weekend' without opening the rows. And never respond to a bought-looking burst by buying a counter-burst of positive reviews to dilute it: that compounds a provenance problem with a fresh violation and a second, datable spike.
Two design choices keep the detection honest. First, nothing is scored until you import it: there is no scraping of your public profiles and no third-party score blended in. Second, the weights reflect what the FTC fake-review rule (16 CFR Part 465) actually prohibits — how a review was solicited weighs heaviest (gating and incentive markers at 1.0 each), while wording patterns weigh less (duplicates 0.8, AI-generation patterns 0.6). A burst of same-day reviews adds a flat 15 points because timing is visible in the dates you supply. You can reproduce every number in the report from your own file, which is exactly the point.
Worked example: a guesthouse, one weekend, two different stories
A fictional worked example, written to show the numbers and the decisions in practice.
A guesthouse imported 118 reviews and hit the burst flag: 52 reviews dated on a single March day — 44% of the file. Checking the source export showed the March date was the day a previous manager had bulk-copied reviews from a booking site into a spreadsheet: a data artefact, now documented. But the same audit found a second, real pattern the artefact had hidden — 8 reviews on one October day, all five stars, all UNLINKED, four sharing AI-phrase wording. Those eight the owner flagged internally with the reasoning logged, and raised with the platform. One import, two bursts, opposite conclusions: exactly why the signal starts an investigation instead of ending one.
Your five-action checklist
- Check whether the burst date is a true review date or an export or migration artefact
- Open every row dated on the burst day and read wording, ratings and sources
- Count how many burst-day rows carry a transaction link
- Look for duplicate or AI-phrase wording inside the burst specifically
- Document your conclusion and moderation decisions in the provenance record
Frequently asked questions
How does ReviewProof define a burst?
At least five reviews sharing one date, with that date holding at least 40% of all imported reviews. A detected burst adds 15 points to the risk score, capped at 100.
Are bursts always fake reviews?
No. Openings, events, media mentions and data artefacts all produce bursts. The wording, rating spread and transaction links on the day separate a surge from a purchase.
Can I fix a burst caused by bad dates?
Fix the source data if you legitimately hold better dates, note the artefact, and re-import. Never invent dates — a corrected record with a note beats a fabricated one.
What should I do about a bought-looking burst?
Keep the rows, flag them internally with your reasoning logged, and dispute through the platform's own process with the evidence pack as your documentation.
Keep reading
Related guides from the ReviewProof library:
- How can you prove a review is real?
- How do you detect fake reviews in your own profile?
- What is review gating, and how do you stop doing it?
- What counts as an incentivised review?
- What does duplicate review text tell you?
- Can you tell if a review was written by AI?
- How is the ReviewProof risk score calculated?
Start with the tool itself: open the free audit desk, see how ReviewProof works, compare Free, Pro and Agency plans, inspect the sample provenance report, or browse all ReviewProof guides.