How do reviews feed AI recommendations?
Reviews are one of the outside sources an assistant checks before naming a business. Your website says you're good. Reviews are other people saying it.
For a buying question, assistants usually run a few web searches, read the results and look at the shortlisted businesses in more detail. Review platforms and map listings show up often in those searches, so their contents end up in the answer. You'll see it in phrasing like "highly rated for fast turnaround" or "customers mention clear pricing".
Google's own AI answers sit on top of Google Search, and Google says review count and score factor into local ranking. ChatGPT search leans on Bing, which has its own listings and pulls in reviews from other sites. Either way, reviews are part of what gets read.
What do assistants actually take from a review?
More than the star rating. A 4.9 with no detail is less useful to an assistant than a 4.7 with fifty reviews that describe the job.
- The rating and the number of reviews, as a quick signal of trust.
- How recent they are. A great rating from four years ago says little about today.
- The words customers use. "Replied in an hour", "fixed the leak same day" and "clear fixed price" become the reasons an assistant gives.
- The services and places mentioned. A review that names the job and the town backs up what your pages say.
- How you reply. Calm, specific replies to bad reviews show a business that deals with problems.
Which review platforms matter most?
The ones your buyers already use and the ones that show up when you ask the assistants. For most local businesses that starts with Google.
Google Business Profile reviews feed Google Search and Maps, and so Google's AI answers. After that it depends on your sector: Trustpilot for many online brands, Checkatrade for trades, Clutch for agencies and B2B services, and sector directories for professional firms. Our guide to the directories AI assistants cite in the UK covers each one.
The quickest way to know is to ask. Put your buyers' questions to ChatGPT, Gemini and Perplexity and note which review sites they cite. Those are the profiles worth your time.
How do you earn more reviews the right way?
These are the steps we set up for clients, in order. None of them take long once the habit is there.
- Run your buyers' questions through the main assistants and list the review sites they cite.
- Claim those profiles and make your name, address, phone number and services match your website exactly.
- Pick one moment to ask: when the job is finished and the customer is happy, not weeks later.
- Send a short, direct link to the review form by email, text or WhatsApp. One tap, no hunting.
- Ask every customer, not just the ones you expect to be happy. Choosing who to ask breaks Google's rules.
- Suggest what to mention without scripting it: the service, the town and what stood out.
- Reply to every review within a few days, good or bad, in plain words.
- Re-run your questions every month and note whether the assistants describe you differently.
What about bad reviews?
A few bad reviews won't stop an assistant naming you. A pattern of the same complaint, left unanswered, might.
Reply to each one with the facts and what you did about it. If a review breaks the platform's rules, report it through the platform. Don't argue in public, and don't try to bury bad reviews under a wave of new ones from friends. Assistants read the whole picture, and so do buyers.
What should you never do with reviews?
Anything fake or bought. It breaks the platforms' rules and, in the UK, the law.
Google bans fake reviews, paid reviews and review gating, where you only ask people you know are happy. The Digital Markets, Competition and Consumers Act 2024 makes fake reviews and hidden incentives for reviews illegal for UK businesses. Platforms remove reviews they think are fake, and a profile that suddenly loses a batch looks worse than one that never had them.
Do reviews matter more than your website?
No. They work together. Your website gives assistants the facts, and reviews give them a reason to trust those facts.
A business with great reviews and a vague website often gets named for the wrong thing. A business with a clear website and no reviews gets read but not trusted. You need both, and the details in each should match.