What Is a 'Review Drive'? The Manipulation Pattern Where the Customers Are Real

A review drive is what our scoring model calls it when a business's staff prompt real customers, in a compressed burst, to leave five-star reviews that name a specific employee. The reviewers are typically genuine, unlike a bought-review campaign, but the directed, clustered pattern still distorts the rating record enough that we flag it under its own verdict rather than folding it into ordinary manipulation.

#What is a review drive, exactly?

It's a distinct pattern in our four-signal model, separate from the burst, low-activity-account, duplicate-text, and rating-mismatch markers described in how the score works. The tell is a spike of five-star reviews landing on the same day or two, a meaningful share of which name a staff member by name, well above the business's normal posting pace. That combination looks like an organized ask rather than reviews arriving on their own schedule.

It isn't the same thing as a fake-review campaign. Review-drive reviewers generally show up as real people who visited the business; there's no evidence of bought accounts or copy-pasted text behind the spike. What's being flagged is the solicitation pattern itself: staff directing customers toward a specific, named review at a specific moment, which produces a rating record that doesn't reflect how customers would have reviewed the business on their own timeline.

#How common is this pattern in the data?

As of August 2026, 15 of the 528 places we've completed analyses for, 2.8%, carry our Review Drive verdict rather than Clean, Suspicious, or Likely Manipulated. It's a small share next to the 89 places flagged Suspicious and the 417 that come back Clean, but it's concentrated almost entirely in full-service restaurants and bars, the kind of business where a manager can plausibly tell a floor shift to ask happy tables for a review before they leave.

Watermark, a waterfront restaurant and bar in Manhattan, scored 53 out of 100 after our analysis flagged 495 of its 999 examined reviews. The evidence log shows nine separate spikes, one alone put 78 five-star reviews on a single day against a baseline of under six per day, and 57 of those named a staff member. IHOP in Hollywood scored 52 after a similar pattern: 22 five-star reviews in one day against a roughly two-per-day baseline, half naming a server. Applebee's Grill + Bar in New York shows the same shape at a score of 48. None of these reports concludes that staff were told to solicit reviews; they show what the review-timing and name-mention pattern looks like, and link the full evidence so anyone can read it directly.

#How does a review drive score differently from a fake-review campaign?

Our model weights review-drive spikes lower than an unexplained burst, 0.5 against a burst's full weight of 1, because a named, staff-adjacent spike carries a plausible innocent reading, a good night, a popular event, a team that happens to be well-liked, that raw burst timing alone doesn't. Verdicts run Clean below a score of 40, Suspicious from 40, Likely Manipulated from 70; a Review Drive verdict applies specifically when the flagged reviews are dominated by this staff-solicitation pattern rather than by the markers more consistent with bought or fabricated content, thin reviewer accounts with no history, or repeated near-identical phrasing.

That distinction matters for how a reader should treat the rating. A Likely Manipulated verdict raises real doubt about whether the reviews reflect genuine customer experiences at all. A Review Drive verdict says something narrower: the positive reviews likely came from real visits, but the timing and prompting mean the published rating is probably higher, and more front-loaded with recent five-star activity, than an unprompted review pattern would have produced.

#Is asking customers to name a staff member in a review against the rules?

Yes, as of April 2026. Google rewrote its Business Profile rating-manipulation policy that month to explicitly bar two practices: setting staff quotas for reviews, and directing customers to name a specific employee. Our breakdown of that policy change covers exactly what changed and what businesses can still do. A customer who decides, unprompted, to praise a server by name is fine under the policy and always has been; what's now explicitly out of bounds is staff being told to request that mention, which is the same behavior our review-drive signal is built to catch independent of Google's policy language.

The FTC's Consumer Review Rule sits alongside this at the federal level, though it's aimed more squarely at incentivized and fabricated reviews than at solicitation timing on its own. A staff-directed review drive that stops short of paying for reviews or scripting their content sits closer to a platform-policy violation than a federal one, which is part of why we score it as its own category instead of folding it into Likely Manipulated.

#What should a business do instead of running a review drive?

The businesses in our data carrying this verdict aren't necessarily doing anything a court would call illegal. They're running review requests the way a lot of hospitality management training still teaches, ask the table before they pay, mention who served them, get the name in there for the employee's own recognition. Google's April 2026 rewrite specifically targets that habit because it produces exactly the clustered, named-staff pattern our model flags.

#Can I check whether a listing I'm evaluating shows this pattern?

Yes. Our analyzer runs the same four-signal model, plus the review-drive check, against any public Google Maps listing and returns the verdict with the underlying evidence in seconds, the same report format shown for Watermark, IHOP, and Applebee's above.

#Frequently asked questions

#Is a Review Drive verdict worse than Suspicious?

Not necessarily. It's a different category describing a different pattern, staff-directed solicitation of named reviews from real customers, rather than a position on a single severity scale. A Review Drive listing's reviewers are typically real visitors; a Suspicious or Likely Manipulated verdict raises more doubt about whether that's true.

#Does a review drive mean the five-star reviews are fake?

Usually not in the sense of being fabricated. The reviewers generally show signs of having actually visited. What's flagged is the timing and staff-directed prompting, not evidence that the reviews themselves are invented.

#How many reviews need to name a staff member to trigger this signal?

There's no single public threshold; the model weighs the share of a burst's reviews that name an employee against the business's own baseline posting rate, alongside how sharply the burst breaks from that baseline.

#Can a restaurant fix a Review Drive verdict?

Stopping the staff-directed solicitation and letting new reviews accumulate at a normal pace over time will change what a fresh analysis finds, since the model reads the whole review history rather than a single snapshot.

#Is asking for reviews at all against the rules?

No. Automated, evenly timed requests sent to every customer, staff uninvolved in who gets asked or what they're asked to say, are fine under both Google's policy and our model.