Spoke
Google reviews
For most local businesses this is the biggest reputation asset there is, and most run it as an occasional push rather than a process. What the signals are, what a working system looks like, and which widely sold tactics are illegal.
Google documents prominence as a local ranking input and reviews as part of it. How much volume, recency and rating each count for is not published, so the specifics here are practitioner consensus and the text says so.
Most businesses treat reviews as a campaign. Somebody notices the count is low, there's a push, thirty arrive in a fortnight, then nothing for a year.
That's worse than a steady trickle on every dimension that matters. It looks engineered to a reader, it decays immediately, and the burst itself is the shape that attracts scrutiny.
The alternative is boring, and it's the whole job: ask everyone, always, the same way.
How do you get more Google reviews?
Build the ask into the moment work finishes, for every client, with no screening step. A short request sent the same day, by whoever the client actually dealt with, with a direct link to the form, converts far better than a generic message sent later by a system nobody recognizes.
Volume comes from consistency, not effort. A firm closing forty matters a month that asks every one of them will pass a competitor running an annual push, usually inside two quarters.
What the signals look like
Google documents local ranking as relevance, distance and prominence, and on reviews it's unusually direct: more reviews and better ratings can help your local ranking. Both the count and the score are named.
What it does not publish is the weighting — how much, against what, decaying how fast. So everything below this line is practitioner consensus.
One distinction most advice blurs. Google ties reviews and ratings to ranking, and ties replying to standing out rather than to ranking. Reply anyway, for the reasons in responding to reviews — but claiming it lifts your position is an overclaim the documentation won't support.
- Recency. A steady flow of recent reviews seems to matter more than a big historic total. It's also the first thing a human checks.
- Volume, with diminishing returns. The gap between six reviews and sixty is enormous. The gap between two hundred and three hundred isn't.
- Rating, non-linearly. Poor to good is decisive. Very good to perfect is nearly nothing, and perfect invites suspicion.
- Text, not just stars. Reviews that actually describe something are more useful to a reader, and they're the part a machine can quote.
- Replies. Specific, visible replies get read by prospects and signal a business that's running.
A process that works
Ask at the right moment. Right after the thing that created the goodwill — not three weeks later when the feeling's faded and they've forgotten your name.
Ask every single client. No screening, no "how did we do?" pre-filter routing unhappy people somewhere else. That's review gating, it's the practice the FTC rule addresses, and it's built into several commercial tools as a selling point.
Make it one tap. A direct link to the form. Every extra step loses a big share of people, and the ones you lose are disproportionately the happy-but-busy.
Send it from a person. A request from whoever they dealt with substantially outperforms a branded automated message.
Ask once, follow up once, stop. Past that you're generating irritation, not reviews.
Make it somebody's job. A process with no owner dies within a month of the person who cared taking a vacation.
What you must not do
This rule isn't advisory. It's been in force since October 2024, it reaches these practices, and it carries civil penalties:
The FTC's Rule on the Use of Consumer Reviews and Testimonials, 16 CFR Part 465, took effect on October 21, 2024. It was published in the Federal Register on August 22, 2024 at 89 FR 68034.
The rule reaches buying positive or negative reviews, insider reviews without disclosure of the relationship, company-controlled review sites, certain review suppression, and fake indicators of social media influence. The FTC can seek civil penalties. Some summaries date it August 14, 2024, which is the announcement, not the publication.
| Do | Don't |
|---|---|
| Ask every client, same schedule | Screen by how happy you think they are |
| Offer an incentive, if any, regardless of rating | Condition anything on the review being positive |
| Let staff honestly review other businesses | Have employees review their own employer as clients |
| Respond to criticism publicly and carefully | Buy removals, or mass-flag honest negatives |
| Use a tool that asks everyone | Use a tool whose selling point is filtering |
That employee one catches well-meaning firms constantly. A staff member leaving a glowing review of their own employer without saying they work there sits squarely inside the insider-review provision, and nobody involved thinks of it as fraud.
Why bursts backfire
A burst produces a cluster of same-dated reviews, often with similar phrasing, often from accounts with no other activity. Any reader who scrolls can see it, and it's the exact signature automated filtering looks for.
Here's what filtering actually does, because it confuses everyone. Reviews that vanish usually aren't deleted — they're withheld. The person who wrote it still sees it on their account. You watch your count fail to move. No notification, no appeal.
That's why "we got thirty and only eleven showed up" is such a common and baffling experience.
Steady flow sidesteps the whole thing.
What it's worth
Reviews convert. Somebody choosing between two firms reads them, and the one with recent, specific, well-answered reviews wins calls that never appear in any ranking report.
They also buy you ad placement. Review score and count are named inputs to Local Services Ads position, so this makes your paid spend work harder too.
That's the honest case. The ranking contribution is real and undocumented in its specifics. The conversion contribution is large, direct, and reason enough to do this even if the ranking effect were zero.
More in Reviews & reputation
Responding to reviews
Your reply to a bad review is read by every prospect who scrolls that far, and it's often more persuasive than the praise above it. For a law firm it's also where a confidentiality breach is most likely to happen.
Fake reviews & removal
What to do about a review that's fraudulent, from a competitor, or just untrue — and the far more common situation where your real reviews are being filtered out and nobody told you.
Reddit & communities
Forum threads turn up constantly in search results and AI answers, which makes them a reputation surface whether you show up or not. They're also where faking it fails most publicly.
Glassdoor & employer reputation
Employer-review sites matter for hiring. The widely repeated claim that they're becoming a search ranking factor has nothing behind it. Both halves of that are worth saying out loud.
Yelp & directories
The platforms beyond Google each serve a different audience. Between them, they're where your old address, your old phone number and your old firm name are still being published to anyone who looks.
Review schema
Marking up reviews you collected about yourself, on your own site, is specifically not eligible for rich results. That single rule explains most of the review markup in the wild, and why it does nothing.
Multi-location reviews
What works for one office falls apart at twelve. The hard parts are knowing which office a review belongs to, giving each one an owner, and stopping head office writing all the replies.