For a local business, the Google listing is the shop window. Most people decide between you and the place next door by reading a handful of reviews, and they never tell you they did. The good news is that getting reviews is not down to luck. It is about asking properly and asking at the right time.
1. Timing is everything
Asking at the till, while the customer is putting their card away, gets almost nothing. Asking right after a moment of genuine satisfaction gets a lot. The best moment in a loyalty programme is obvious: just after someone redeems a reward. They have received something free, they feel good about you, and they have their phone in their hand.
2. Make it one tap
Every intermediate screen loses part of your audience. Link straight to the review box, not to the listing, not to a landing page. If someone has to search for your business to leave the review, most will not finish.
3. Ask the people who already like you
Your regulars are the ones who will write something specific and credible. A first-time visitor writes "nice place". Someone who has been coming for six months writes about the staff, the usual order, the detail that makes you different. That second review is worth more to the reader.
4. Answer all of them, especially the bad ones
Answer thinking about the person who will read it in six months, not the person who wrote it. That person has already made their mind up. The future reader has not.
- Answer within 48 hours. A late reply reads as obligation.
- Do not argue the facts in public, even when you are right. A business defending itself point by point always looks worse than the complaint.
- Acknowledge, say what you changed, offer to continue privately. In that order. What convinces the reader is the "we have changed X", because it is the only part that affects them.
- Never ask for it to be deleted. It is the single most damaging reply.
A listing with nothing but five stars breeds suspicion. A handful of three and four star reviews, answered well, makes the good ones more believable.
Star filtering: why we do not do it and neither should you
There is a tactic many tools sell that is worth naming so you can rule it out. You first ask the customer, inside your own system, how many stars they would give. Anyone answering four or five is sent to Google; anyone answering one, two or three is diverted to a private form that never reaches the listing.
It sounds clever and it does raise the average. The problem is that Google policy on local content prohibits soliciting reviews selectively, and in the United States the FTC has penalised the practice as deceptive reviews. Applied across hundreds of businesses it stops being a debatable trick and becomes a risk to every one of those listings.
The alternative that does work is not asking everyone at once: it is asking at the moment the experience has genuinely been good. You are not filtering opinions, you are choosing the instant. The difference sounds subtle and it is the line between complying and not.
How many reviews you actually need
The usual question is how many are needed to rank. It is the wrong question, twice over.
First, because people choosing where to eat do not compare numbers, they compare impressions. A business with 40 reviews at 4.6 with recent opinions is more convincing than one with 300 at 4.8 whose last review is two years old. Freshness weighs as much as volume.
Second, because the number that matters is relative to your street. If the place opposite has 25, your target is not 500: it is passing 25 and continuing. Look at the three listings you genuinely compete with, not at a national chain.
A sustainable pace for a neighbourhood business is two to five new reviews a month. Enough for the listing to look alive, slow enough that it does not look bought. Fifty reviews in a week after two years of silence is a pattern that shows.
How to ask without it sounding like a favour
- The reason. "It helps people in the area find us" works better than "leave us a review", because it explains what it is for.
- The effort. Saying "it takes thirty seconds" removes the real barrier, which is not laziness but not knowing what you are getting into.
- The direct link. To the writing box. Every extra screen costs you people.
And one thing to avoid: do not dictate the content. Suggesting what to write produces reviews that resemble each other, and that is easy to detect.
What Google will not tell you
- Reviews disappear, often with no visible reason. Automated filters remove legitimate opinions, especially from accounts with no history. It is not personal and there is usually no way to recover them.
- A review written on the shop wifi can be removed. Several reviews from the same network in a short window look coordinated. If you ask in the shop, it is better that people write later, on their own connection.
- Replying does not raise your score, but it changes what the reader decides. It is one of the few things about your listing you control completely.
What not to do
- Do not pay for reviews. Beyond being against Google policy, bought reviews read like bought reviews and they damage the ones you earned.
- Do not offer a discount in exchange for a review. Incentivising the review itself is a different thing from asking at a good moment, and it is the line Google policy draws.
- Do not ask everyone on the same day. A burst after months of silence is the most visible pattern there is.
- Do not write them yourself. Reviews from accounts with no history, written in the same voice, get filtered and take your credibility with them.
- Do not chase the same customer twice. If they did not leave one after a good experience, a second reminder is not going to help and it costs you goodwill.
Automating it: the real jump
Everything above works done by hand and almost nobody does it by hand for more than a fortnight. Asking at the right moment means noticing the right moment, and in a busy service nobody is watching for it.
What changes the numbers is tying the request to an event that already happens: a reward being redeemed, a visit milestone, a customer reaching a level. The event fires, the request goes out, and nobody at the counter has to remember anything.
Two rules if you automate it. Cap how often the same person can be asked, so an active customer never gets it twice. And keep the request tied to something genuinely positive, because the moment it fires on every transaction it stops being a good moment and becomes noise.