A business with eleven reviews and a business with a hundred and eleven are not in the same market, even if they sell the same thing on the same street. The second one gets shortlisted without a phone call. The first one has to earn a conversation before it can earn a sale. That gap is not a marketing problem and it is not a mystery. It is a process problem, and most owners lose it in the same three places: they never ask, they ask at the wrong moment, or they ask everyone at once and then go quiet for a year.
I founded BrandAmplifi, which works in online reputation and search visibility, so I spend more time than most people looking at what separates a review profile that compounds from one that stalls. What I have come to believe is unglamorous. The answer to how to get more customer reviews is almost never a clever tactic. It is a small, repeatable habit installed at the one point in your operation where the customer is most likely to feel something.
Find the moment of relief, not the moment of payment
Most businesses ask at checkout, because checkout is where the software lives. That is the wrong moment. At checkout the customer is thinking about money, or about leaving. The right moment is what I call relief — the instant the thing they were worried about stops being a worry.
For a roofer, relief is not when the invoice clears. It is when the crew leaves and the driveway is clean and the homeowner walks the perimeter and sees nothing wrong. For a dental practice it is not the front desk. It is the follow-up call two days later when nothing hurts. For a marine service shop it is the first time the boat runs right after the repair. If you sell to businesses, relief is the week after go-live, when the thing you installed has survived contact with real use.
Write down your relief moment. Then ask whether anyone in your business is present at it. Very often the answer is no — the person who created the relief has already driven to the next job, and the only person still in contact is a billing system. That is the actual reason your review count is flat. Nobody is standing where the goodwill is.
The ask itself: one human, one sentence, one link
The strongest ask I know of has three parts and takes about eleven seconds. A person who was actually involved says something specific about the job. They say the review helps other people in the area find them. Then they send one link, immediately, while the customer is still holding their phone.
Specificity is what makes it work. "Leave us a review" is a chore. "You were patient with us on the permit delay — if you have a minute, a review helps other homeowners in the neighborhood find someone they can trust with that stuff" is a favor between two people who just did something together. The second version also, quietly, tells the customer what to write about. Not by scripting them — by reminding them what the story was.
Two things I would treat as rules. First, never ask a customer to mention a specific keyword or phrase, and never hand them draft text. It is transparently coached, the platforms can see the pattern, and a page full of reviews that all use your service description reads as fake even when it isn't. Second, do not chain the ask to an incentive. A discount for a review does not buy you a review; it buys you an obligation, and the customer will discharge it with four stars and eight words. Worse, on most major platforms it violates the terms you agreed to.
Pace beats volume
Here is the part that surprises owners. Getting fifty reviews in one week is worse for you than getting fifty over five months. A cluster looks like a campaign. A steady trickle looks like a business that keeps serving people. Recency is doing quiet work too — a reader scanning your profile is really asking whether you are still good now, and thirty reviews from three years ago answers a different question than eight from this quarter.
So set a rate, not a target. If you close forty jobs a month, ask on every one and expect a modest fraction to convert. That is a handful of new reviews a month, which crosses a hundred in a year or two without a single stunt. It also means a bad month cannot sink you, because the denominator keeps growing. This is the same discipline I associate with steady operating rhythm rather than heroics — the habit beats the push, which is a theme I have written about in the context of what private equity discipline teaches an owner-operator.
Why bought reviews destroy the thing you are building
The temptation is obvious and the pitch is always the same: you need social proof to compete, buying it gets you there faster, everyone does it. I want to be direct about why this is a bad trade even in pure self-interest terms.
A review profile is an asset because it carries information other people cannot easily manufacture. The moment you buy reviews, you have converted an asset into a liability with a delayed trigger. Platforms improve their detection continuously, and removals are not gentle or partial — profiles get filtered, badges get pulled, and the loss lands at whatever moment the platform chooses, which will not be a convenient one. You have also blinded yourself. Fake praise is noise in the one feedback channel that tells you where your operation is leaking.
And it is legible to humans. Bought reviews cluster in time, come from thin accounts, praise in general terms and never mention anything inconvenient. Anyone who reads carefully — a prospective customer comparing three vendors, a bank, an acquirer doing diligence — can spot it. In my own work evaluating businesses, an implausibly clean review profile makes me look harder, not less hard. It is a signal about how the owner handles pressure, and that signal transfers to everything else they tell me.
Responding to negative reviews without arguing
A one-star review is not the emergency. The reply is where reputations are actually made or lost, because the reply is the only part a future customer reads as evidence of how you behave when things go wrong.
My structure is boring on purpose. Acknowledge the specific thing that happened, without the word "but." Say what you are doing about it, if anything. Move the conversation off the platform by giving a direct way to reach a real person. Stop. Do not relitigate the timeline, do not correct their memory, and do not explain your policy. You are not writing to the reviewer — that relationship is already strained. You are writing to the person reading over their shoulder six months from now.
Two harder cases. When the review is unfair or factually wrong, respond once, calmly and briefly, correcting the record without adjectives, then let it stand. Readers are surprisingly good at recognising an unreasonable customer, but only if you have not matched their tone. When the review is fair and the failure was real, say so plainly. A visible, unhedged admission does more for a local business's credibility than a page of five-star praise, because it is the only entry on the page that could not possibly have been bought.
Respond to the positive ones too, briefly, and not with the same sentence every time. Identical replies are the tell that a template is running the account. Two lines that reference the actual job take fifteen seconds and read as a business where someone is home.
Recovering from a bad stretch
Every business eventually has a month where a crew turns over, a supplier fails, or a system change breaks something quietly, and three bad reviews arrive together. The instinct is to go on offense and gather good reviews fast. Resist that for a week.
Read the three reviews for the shared mechanism first. Bad reviews arriving in a cluster usually describe one operational failure from three angles — a scheduling gap, a handoff nobody owns, a communication step that was dropped when someone left. Fix that, then resume asking at your normal rate. If you fix nothing and simply out-volume the complaints, you have bought yourself an average and kept the defect, and it will surface again.
Do watch what the pattern is telling you about demand. Reviews and referrals are the same underlying asset seen from two directions, which is why I think about them together with how customers actually find you in a dense local market. And once you have a year of honest reviews, you have a small, real dataset about your own operation — the kind of thing worth reading properly before you buy tooling for it, which is closer to the first-ninety-days approach I take to AI in a small business than to any reputation software purchase.
The version of this that lasts
If I had to reduce online reputation management for a small business to one instruction, it would be this: put the ask in the hands of the person who did the work, at the moment the customer feels relief, and then do it every week for two years without drama. That is it. No campaign, no incentive, no purchased volume.
What you end up with is not just a number next to your name. It is a public, dated, hard-to-fake record of how you treat people when nobody made you — including the times you got it wrong and said so. We tend to talk about reputation as something to protect. It is more useful to treat it as something you deposit into, in small honest amounts, at a rate you can sustain when business is bad as well as good.