How to get more Google reviews without the tactics that get you penalized

What is wrong with the usual advice?

It describes review gating. Survey customers first, send the happy ones to Google and route the unhappy ones to a private form. It works, it is widely sold, and it is against Google's policies because the resulting rating is not a representative sample of anything.

The tactic survives because it sounds reasonable when described gently. Nobody says we suppress negative reviews. They say we identify promoters, or we route feedback appropriately, and the software that does it is marketed as reputation management rather than as what it is.

In the United States there is a second layer. The FTC's rule on consumer reviews and testimonials addresses suppressing negative reviews and incentivized reviews that are not disclosed, and it carries real penalties. That moves this from a platform policy question to a legal one for anyone selling into the US market, which is most of the companies we work with.

So the honest starting point is that the fast methods are the prohibited ones. What follows is slower and it does not put your rating at risk.

What exactly is not allowed?

The specifics matter, because several of these are things well-meaning teams do without realizing. None of them are grey areas.

That last row surprises people. A well-intentioned push where twenty customers review from the same connection in one afternoon looks, to an automated system, exactly like a purchased batch. The reviews disappear and the effort is wasted.

So what actually works?

Asking everyone, quickly, in the channel the relationship already lives in, at the moment the value is obvious. That is the entire method. It is unglamorous and it compounds.

The specific question is the underrated one. A prompt like tell others what problem we solved for you produces reviews that mention the service and the situation, which is what a future buyer is actually scanning for. A generic prompt produces four words and a star rating.

For B2B this runs differently than for a restaurant. You have fewer customers, you know them, and a request from the person who did the work is far more effective than an automated sequence. Volume is not the goal. Twenty specific, credible reviews outperform two hundred that say great service.

• Ask every customer, not a filtered subset, because the filter is the violation
• Ask close to the moment the work delivered something visible, not weeks later
• Ask in the channel you already use with them, since a new channel adds friction
• Send the direct review link rather than instructions for finding you
• Ask a specific question, because what was the service like produces nothing usable
• Make it one person's job, or it happens for a fortnight and then stops

Do Google reviews even matter for B2B?

Less than for a local consumer business, and more than most B2B teams assume. They are rarely the first touch and they are frequently the last check before someone books a call.

The realistic role is verification rather than discovery. A buyer has found you some other way, is mostly convinced, and does a quick search on the company name to see whether anything is obviously wrong. An empty profile is not damning. A profile with two reviews from 2023, one of them unhappy and unanswered, is.

There is a second reason that matters more here. Reviews are part of how a company is described across the web, and that consistency feeds how AI systems summarize you. A business with a complete profile and a handful of specific reviews is easier for any system to describe accurately than one with almost nothing attached to its name.

So the goal for a B2B firm is not a high volume. It is a profile that is complete, current, and specific enough to be worth reading.

What do you do about the bad one?

Reply to it, in public, once, without arguing. The reply is not for the person who wrote it. It is for the next twenty people who read it while deciding whether to contact you.

The instinct is to correct the record, and it reliably makes things worse. A defensive reply reads as defensive no matter how accurate it is, and a long one signals that the complaint touched something. Acknowledge, state briefly what changed, offer to take it offline, stop.

A negative review that is answered well does less damage than most people fear. Buyers discount a perfect five star profile and read a mixed one more carefully. What actually costs you is a complaint sitting unanswered for a year, because the message that sends is that nobody is watching.

If the review is genuinely fake or violates policy, report it and expect the process to be slow and mostly automated. We have written separately about what can realistically be removed and what cannot, because that expectation is where most of the frustration comes from.

Can any of this be automated?

The reminder can. The judgment cannot, and this is the specific place where automating too much creates the problem you were trying to avoid.

A scheduled request after a milestone is fine and it is the difference between a program that runs and one that does not. What must not be automated is who gets asked, because a rule that only asks customers with a high satisfaction score is review gating implemented in software, and it is a violation whether a person or a system applied it.

So automate the trigger and the send. Keep the recipient list unfiltered. If someone objects that this will bring in negative reviews, that is the objection, and the answer is that a rating built by excluding them is not a rating.

Keep the record too. Who was asked, when, and through what channel. If a batch of reviews is ever questioned, being able to show an ordinary, unfiltered process is the whole defense.

How long does this take to show up?

Months, and it is one of the few marketing activities where the slow version is the only version. A profile that goes from three reviews to thirty in a week is the shape that gets flagged.

Plan for a steady trickle rather than a campaign. A handful a month, arriving from different people at different times through normal channels, looks like what it is. The same number in two days does not.

Set the expectation internally before you start, because the pressure to do a push always arrives around week three when the numbers look unimpressive. That push is what costs businesses the reviews they had already earned.

The compounding argument is the one worth making. Reviews do not expire. A year of asking properly leaves you with an asset that keeps working, and there is no shortcut that produces the same thing without the risk attached.