Data, SEO & AI Search · Field guide

How to get more Google reviews

Ask every pleased customer once, at the moment the job finishes, with a link that takes one tap. That is most of it. The part almost nobody plans for is that reviews decay: 74 per cent of people look only at the last three months, so the number that decides whether you get called is how many you collected last month, not how many you have collected ever.

97%of consumers read reviews before choosing a local business
74%look for reviews written in the last three months
88%would use a business that replies to all of its reviews
47%would use one that never replies to any of them

The short answer

How do you get more Google reviews?

Ask every satisfied customer, once, at the moment the job is finished, in the channel they already use, with a link that takes one tap. Everything else is detail. The businesses that collect reviews steadily are not the ones with the cleverest campaign; they are the ones that made asking part of finishing a job, so it happens whether or not anybody remembers to run a campaign that month.

Key takeaways

01

Ask at the moment of completion, when the customer is most pleased. An hour later is good, a week later is much worse, and a monthly bulk request to everyone at once is worst of all.

02

Reviews decay. 74 per cent of consumers look for reviews written in the last three months, so a business with 180 old reviews and none this quarter reads as a business with none.

03

Reply to everything, including the complaints. 88 per cent said they would use a business that replies to all of its reviews, against 47 per cent for one that never replies.

04

Never buy, incentivise or filter reviews. They are removable, they breach the platform rules, and a listing that loses a batch is worse off than one that never had them.

The thing nobody expects

Why your review total matters less than you think

The instinct is to treat reviews as a bank balance: get to a hundred, then relax. That instinct is wrong, and the survey data is unusually clear about why.

BrightLocal's Local Consumer Review Survey for 2026 asked how old a review can be before people stop finding it relevant. 74 per cent look for reviews written within the last three months. 32 per cent narrow that to a fortnight. 18 per cent will only be swayed by something written in the last week. The same survey found 97 per cent read reviews before choosing a local business at all, and 41 per cent now say they always do, up from 29 per cent a year earlier.

How fresh a review has to be to count Share of consumers who look for reviews written within each window. They overlap: the same person can appear in more than one row. Written in the last week 18 Written in the last two weeks 32 Written in the last three months 74 BrightLocal Local Consumer Review Survey 2026, a representative panel of 1,002 adults, published February 2026. Australian survey data on review recency at this scale is not published, so read the direction rather than the exact figure.

How recent a review has to be before people count it. The windows overlap, so one person can sit in several rows. The point is the shape: three quarters of buyers are looking at a rolling three month window, not at your total.

Put those two findings together and the conclusion is uncomfortable. Nearly everyone reads reviews, and three quarters of them are effectively reading only the last quarter of yours. A business with 180 reviews, all more than a year old, is presenting almost nothing to most of the people deciding whether to call it.

97 per cent read them, 74 per cent want them recentWhich means the useful question is not how many reviews you have. It is how many you got last month, and whether that number is going to repeat.

A note on currency: BrightLocal's panel is 1,002 adults in the United States, published in February 2026. There is no Australian survey of the same size on review recency, so treat the direction as reliable and the precise percentage as indicative rather than local.

The reframe

Rate beats total, and rate is the part you control

The same shopfront, two very different signals Both businesses below are real shapes we see constantly. One of them looks far better on paper. BUSINESS A 180 reviews in total Collected hard during one push, then nobody asked again. Written in the last three months None What most buyers actually see: an empty recent window The 180 is still there. It is just not doing the work. BUSINESS B 60 reviews in total Two a week, asked for at the moment the job finishes. Written in the last three months Twenty six What most buyers actually see: a business people use now A third of the total, and far more of the effect. Totals are a vanity number. The rate is the number that shows up in a decision, and it is the one you can change this month.

Two businesses with very different totals. Within the three month window most buyers actually look at, the one with a third as many reviews is showing far more. The total is history; the rate is what is visible.

This is the single most useful shift in thinking we can offer on the subject. Stop counting reviews and start counting reviews per month.

A rate is a much better target than a total, for three reasons. It is achievable this week rather than this year. It survives contact with a busy month, because two a week is a habit and two hundred is a project. And it keeps working: a business holding a steady rate is permanently visible in the window that matters, while a business with a big total is slowly disappearing from it.

Work out your realistic rate first

Count the customers you finish with in a month who were genuinely pleased. A sensible starting target is that a quarter to a third of them leave a review once you actually ask. Set the target from that number rather than from what a competitor's listing shows.

Attach the ask to an existing step

The request has to live inside something that already happens: the final invoice, the job completion message, the discharge summary, the handover email. Anything that requires somebody to remember a separate task will run for three weeks and stop.

Measure monthly, not cumulatively

Report reviews received this month against last month. A cumulative total only ever goes up, which means it can look healthy while the rate has been zero since March.

Expect the rate to fall, and plan for it

Every process decays. Check quarterly whether the ask is still being sent, whether the link still works, and whether whoever used to do it still works there. Most collapsed review programmes died quietly rather than deliberately.

The method

How to ask, and exactly when

Timing does more work here than wording. The gap between asking at the right moment and asking a week later is larger than the gap between a good message and a mediocre one.

Finish the job properly

The only reliable input is a customer who is actually pleased. Asking an ambivalent customer for a review is asking for a three star, and three stars cost more than they give. If the job went badly, fix it, and do not ask.

Step 1

Ask within the hour

At handover, at checkout, when the work is signed off. This is the peak of goodwill and it falls away fast. In person, by the person who did the work, is the strongest version of the ask there is.

Step 2

Follow with a link, immediately

Whatever was said out loud has to arrive as a tappable link within minutes, while the conversation is still in their head. A link sent tomorrow is a different, much weaker request.

Step 3

One reminder, then stop

A single nudge two or three days later recovers a meaningful share of the people who meant to and forgot. A second and third reminder annoy people and buy almost nothing.

Step 4

Reply when it lands

A reply within a day or two closes the loop, is read by everyone who comes after, and makes the next request easier because the last person had a visibly good experience.

Step 5

On what to actually say: keep it short, make it specific, and do not script the customer. Naming the job ("the bathroom", "Tuesday's session") produces reviews that mention it, and specific reviews are both more persuasive and more useful for local relevance than five stars with no words.

Never suggest what to write or what rating to leave. Beyond the platform rules, a set of reviews that all use the same phrases is obvious to a reader and increasingly obvious to the platforms.

Do ask at completion

In person if you can, by the person who did the work. Goodwill has a short half-life and this is its peak.

Do make it one tap

Your own short link straight to the review form. Every extra step loses people, and asking someone to search for you loses most of them.

Do name the job

"If you have two minutes, a line about the kitchen would help us enormously." Specific prompts produce specific reviews.

Do ask everyone pleased

Not just the delighted ones. Cherry-picking who gets asked is the first step towards filtering, which is against the rules.

Do not offer anything

No discounts, no entries into a draw, no free anything. It breaches the platform terms and the reviews are removable.

Do not screen first

Asking who is happy before deciding who gets the link is review gating, and it is explicitly prohibited.

The channel

Which method actually gets a response

There is no universal winner here, and the honest answer is that it depends on how you already talk to customers. What matters far more than the channel is that the request goes out in the one they are already reading.

ChannelWorks well whenWatch out for
Asked in person, then textedThere is a moment of handover: a trade, a clinic, a salon, a workshopIt dies the moment the person asking leaves or gets busy
Text messageYou already have mobile numbers and permission to use themConsent still applies, and a text nobody agreed to is a complaint
Email, sent individuallyThe relationship is longer and email is how you already workBulk sends to everyone at once read as marketing and convert poorly
A card, a QR code or a link on the invoiceYou want something that keeps working with no one driving itIt is passive, so it produces a trickle rather than a rate
Inside your booking or job systemThe software can send it automatically at completionAutomation is only as good as its trigger: check it monthly

Whichever you choose, the link itself deserves five minutes of attention. Google provides a short review link from your business profile, and using it directly, rather than sending people to search for you, is the difference between most people finishing and most people abandoning.

If you want the profile itself in order before you start driving people to it, our guide to Google Business Profile optimisation covers the categories, fields and photos that decide what a visitor sees when they arrive.

The half nobody does

Replying, including to the ones that hurt

Replies are the cheapest advantage available in this whole subject, because most businesses still do not do it. BrightLocal's 2024 survey found 88 per cent of consumers would use a business that replies to all of its reviews, against 47 per cent for one that never replies at all.

That gap is not really about the reviewer. Almost nobody rereads a reply to their own review. The audience is the next person, who is reading the whole page and forming a view about what you are like to deal with when something goes wrong.

88%would use a business that replies to all of its reviewsBrightLocal Local Consumer Review Survey
47%would use one that never replies to anyBrightLocal Local Consumer Review Survey
74%look for reviews written in the last three monthsBrightLocal Local Consumer Review Survey
31%will only use a business rated 4.5 stars or better, up from 17 per centBrightLocal Local Consumer Review Survey

For a negative review, the reply is a piece of writing for an audience of future customers, and it has three jobs: acknowledge the specific thing, say what you did about it, and move the argument offline. It does not have to concede that the customer was right.

What loses is arguing the facts in public, however satisfying it feels and however wrong they are. A reader cannot verify who is correct, so a long defensive reply reads as a business that is difficult, and that is the only impression that survives.

A four point something is more credible than a fiveA perfect rating with no dissent at all reads as curated. A handful of mixed reviews, answered well, is more persuasive than an unbroken wall of five stars.

There is also a floor worth knowing about. 31 per cent of the same survey's respondents said they will only use a business rated 4.5 stars or higher, up from 17 per cent the year before. That threshold has been rising steadily, which means the cost of a run of poor reviews is higher than it used to be, and the fastest way back above it is a steady rate of new ones rather than a fight about the old.

The shortcuts

What never to do, and why it costs more than it gives

Everything in this section is sold to Australian businesses regularly, and all of it is a bad trade.

Buying reviews

They are detectable, they breach the terms, and removal takes the whole batch at once. A listing that drops forty reviews overnight is in a worse position than one that never had them.

Incentives of any kind

A discount, a draw entry, a free coffee. All prohibited, all removable, and all producing exactly the bland five star reviews that persuade nobody.

Review gating

Surveying customers first and only sending the link to the happy ones. This is explicitly against the rules and it is easy to spot from the outside.

Writing them for people

Handing a customer a draft to paste. It shows: a page of reviews in the same voice is one of the more obvious patterns a reader picks up on.

Asking staff and family

Reviews from people connected to the business, often from the same location or device, and they tend to be the first ones removed.

Chasing a competitor's total

Their total was built over years and may be partly bought. Chasing it produces panic campaigns; a steady rate beats it inside a year anyway.

The honest version of this work is slower and it is the only version that holds. It is also, in our experience, the part of local search that most businesses can do better than their competitors without spending anything, because the bar is genuinely low.

If you would rather have the asking, the replying and the profile handled for you alongside the rest of it, that is our local SEO service in Sydney. Because our Snowball SEO platform automates the audits, the tracking and the reporting that other agencies bill by the hour, more of the budget goes into the parts that actually move a local ranking. On how reviews sit alongside proximity and the rest of the signals, see how local SEO works in Sydney.

Sources: Consumer figures from BrightLocal, Local Consumer Review Survey 2026, a representative panel of 1,002 adults published in February 2026, and from the same survey's 2024 edition for the reply figures. That panel is United States based; no Australian survey of comparable size covers review recency, so the direction is reliable and the precise percentages are indicative. The rate-over-total argument and the collection method are our own practice, not a published benchmark.

Anthony Betzis
Founder, Snowball Productions

Anthony founded Snowball Productions to help Australian businesses win the searches that actually bring in work. He writes about the data behind SEO, AI search and the content that converts.

Start here

Want the asking and the replying handled?

Tell us how you finish a job and who your customers are. We will show you where the ask fits in what you already do, and what a realistic monthly rate looks like for your business.

Get your free audit

Good questions

Google reviews FAQs

How do I get more Google reviews?

Ask every satisfied customer once, at the moment the job is finished, in person where you can, and follow up within minutes with a link that takes one tap. Send a single reminder two or three days later and then stop. The businesses that collect reviews steadily are the ones that made asking part of finishing a job rather than a separate marketing task.

How many Google reviews do I need?

Fewer than you think, but more recently than you think. 74 per cent of consumers look for reviews written in the last three months, so a steady rate matters more than a large total. A business with sixty reviews and six from last month presents better than one with a hundred and eighty that are all a year old.

Can I offer a discount for a Google review?

No. Incentivised reviews breach the platform rules, they are removable, and losing a batch leaves you worse off than never having collected them. They also tend to produce bland five star reviews with no detail, which persuade nobody and do nothing for local relevance.

Should I reply to negative Google reviews?

Yes, and quickly. The reply is written for the next reader rather than for the reviewer. Acknowledge the specific issue, say what you did about it, and move the conversation offline. Arguing the facts in public reads as a difficult business regardless of who is right.

How recent do Google reviews need to be?

74 per cent of consumers look for reviews written in the last three months, 32 per cent narrow that to a fortnight, and 18 per cent will only be swayed by something from the last week. Reviews older than a quarter still count towards your rating but do very little in an actual buying decision.

Do Google reviews help my rankings?

They are one of the recognised signals behind local results, alongside proximity to the searcher and the completeness of your business profile. They also decide whether somebody clicks once you appear, which for most businesses is worth as much as the position itself.