Why reviews drive local choice
Local search collapses a decision into a glance. The results page (SERPThe page of results a search engine shows.) for a local query shows a map with a handful of businesses, each with a star rating and a review count next to the name. Before anyone has read a word about the business, the rating has already sorted the options. Google's help centre names reviews among the inputs to prominence, one of its documented local ranking factors, and says in plain terms that high-quality, positive reviews can improve visibility. Whatever the exact weight, the practical effect is that a business with few, old or poor reviews is less likely to be shown and less likely to be chosen when it is.
The second half of the effect happens after the click. People open the profile (GBPLocal listing on Google Search and Maps.) and read the most recent reviews, not the best ones. They look for a reply under the negative review to see how the business behaves when something goes wrong. They look at the dates to see whether the business is still active. A profile with a high rating from reviews left three years ago reads as a business that stopped caring. Recency, volume and replies together drive the conversion rate (CVRShare of visits or clicks that convert.) from profile view to call or visit, and that conversion is where presence work pays.
Reviews also carry the experience and trust signals (E-E-A-TExperience, Expertise, Authoritativeness, Trust.) that search engines describe wanting to see, because they are evidence from people who were there. A review that describes a specific job, a specific member of staff or a specific problem solved is worth more to a reader than any claim the business makes about itself. That is why the routine below aims at genuine, specific reviews from ordinary customers rather than at a number.
Asking for reviews within the rules
Most businesses that lack reviews lack them because nobody asks. Most businesses that get into trouble over reviews do so because they asked in a way the platform or the regulator prohibits. The table separates the two.
| Practice | Allowed in general | Not allowed in general | Why |
|---|---|---|---|
| Asking after a job or purchase | Asking every customer, in person, by message or on a card, with a direct link | Asking only customers you expect to be positive | Google's help centre prohibits selectively soliciting reviews; regulators treat suppression of negative feedback as misleading |
| Incentives | None | Discounts, prizes, free items or entries in exchange for a review | The help centre prohibits incentivised reviews; the CMA and FTC both treat undisclosed incentives as a consumer protection issue |
| Who leaves the review | The customer, from their own account | Staff, family, the owner, or a review written for the customer | Platforms prohibit reviews from people with a conflict of interest; regulators treat fake reviews as unlawful |
| Bulk requests | A steady request sent to each customer after service | Collecting reviews on a shared device in the shop, or sending a burst to a backlog | The help centre documents that review stations and bulk solicitation can trigger removal |
| Gating | Asking for feedback and for a review at the same time, visible to all | A form that routes happy customers to the platform and unhappy ones to a private box | Gating produces a misleading picture; regulators have acted against it |
| Buying reviews | Never | Any paid service that supplies reviews | Fake reviews are the clearest breach of both platform policy and consumer law |
The rules in the table are general direction, not legal advice; the specifics differ by country and the regulators update their guidance. The FTC's business guidance and the Competition and Markets Authority pages are the places to check, and a business running any scheme beyond a plain request should ask a professional. Before asking customers by text, card or in person, the review request launch checklist turns the table into lines to tick, from gating forms to staff incentives.
What works, and is allowed, is simple. Ask every customer at the moment of satisfaction: when the job is signed off, when the bouquet is collected, when the appointment ends. Give them a short link to the review form, which the profile tutorial shows how to generate. Say why it matters to a small business. Then stop. A consistent ask produces a steady stream of reviews, and a steady stream is what the platforms treat as natural and readers treat as credible.
Responding to reviews: positive, negative, fake
Every reply is written for the next reader, not for the reviewer. That principle settles most questions about tone.
Positive reviews get a short, specific reply within a day or two. Thank the person, mention the thing they mentioned, and sign off. A reply that repeats the same sentence under every review reads as automated; a reply that mentions the sofa they chose or the leak that was fixed reads as a business that remembers its customers. Do not use positive replies to sell.
Negative reviews get the careful reply. Read it twice, check the facts internally, then write. Thank them for the feedback, acknowledge the specific problem without arguing the details in public, say what has been done or will be done, and offer a direct route to resolve it. Keep it short. Readers forgive a business that had a bad day and handled it well; they do not forgive one that attacks a customer. Never reveal private details about the customer in a reply, even when the review is unfair. If the review is a genuine complaint, resolve it off the platform and, once resolved, it is acceptable to ask the customer whether they would consider updating their review; it is not acceptable to make the resolution conditional on it.
Fake or suspicious reviews are the ones from people who were never customers, from competitors, or from accounts that leave identical reviews across unrelated businesses. Flag them through the platform's reporting route, citing the specific policy, and keep a record. Reply publicly and calmly: "We have no record of a customer by this name or of this job; please contact us directly so we can look into it." That reply tells readers the review is disputed without making an accusation. Platforms remove reviews at their own discretion, often slowly, so the plan cannot depend on removal; it depends on enough genuine reviews that one or two fakes do not define the rating.
The community management guide in the social media section covers replies and moderation on social platforms in more depth; the same principles apply to reviews with a stronger emphasis on the permanent, public record. When the criticism appears in a local news story rather than a review, the guide to responding to negative or inaccurate coverage covers separating errors from fair criticism and asking the reporter for a correction.
Review platforms beyond the search listing
The search listing is the review platform that matters most for most local businesses, because it is the one shown at the moment of choice. It is not the only one, and the mix depends on the trade.
- General review sites that aggregate ratings across businesses carry weight in some countries and trades and little in others; the test is whether customers in the area actually use them.
- Trade-specific platforms matter for trades, home services and professionals, where customers look for vetted tradespeople or for solicitor and dentist reviews on sites built for that purpose.
- Hospitality and travel platforms are where restaurants, hotels and attractions are judged, often ahead of the search listing. The hospitality marketing guide covers how those reviews and booking platforms shape a venue's plan.
- Social platforms with recommendation features matter for businesses whose customers find them through friends and neighbourhood groups.
- Marketplace and booking platforms carry reviews that live with the booking, which matters for anyone who takes bookings through them.
Choose two or three that customers in the trade actually read, claim the profiles, and include them in the routine. Spreading requests across platforms thinly is worse than concentrating on the one or two that are read. Product reviews on an online shop are a separate discipline, covered by the reviews and social proof guide in the e-commerce section; this guide is about the business, not the catalogue.
Monitoring and alerts
A reputation routine needs an early warning system, because a bad review that sits unanswered for a month does more damage than one answered the same day. The minimum is notifications turned on for every claimed profile, routed to a person who checks them daily, with a named deputy for holidays. Beyond that, a weekly check of each platform catches reviews the notifications missed and the questions people ask on the profile, which the help centre documents can be answered by the business and by the public.
Multi-site businesses and busy trades usually move to a monitoring tool that pulls reviews from several platforms into one inbox, with alerts for low ratings and for keywords such as "refund", "rude" or "never turned up". The tool is not the point; the response time is. A simple spreadsheet log of each review, its platform, its rating, the date, who replied and whether a fix followed is enough for a single site and makes the next section possible.
Watch for patterns rather than individual reviews. One complaint about waiting time is noise; three in a month is a staffing problem showing itself in public. Track the rolling rating and the review rate per month, not the lifetime average, because the recent figures are what readers see first and what change first when something goes wrong.
Turning reviews into content and fixes
Reviews are the cheapest customer research a local business will ever get. Read a quarter's worth together and two things fall out. The first is a list of operational fixes: the recurring complaint about parking, the member of staff who is named in every positive review and should be training the others, the service nobody mentions because nobody knows it exists. Fix those and the next quarter's reviews change.
The second is content. The phrases customers use to describe the work are the phrases prospective customers search for, and they belong in service pages and area pages. Specific reviews, quoted with permission or displayed through the platform's own embed, make the strongest testimonialsA named customer's account of a specific result in their own words, used in marketing as proof and treated as an endorsement that must be genuine. because readers can check them. Photos customers attach to reviews show real work rather than stock imagery. The local content and community marketing guide covers where that material goes; the profile tutorial covers the posts and the question-and-answer section where some of it belongs.
A plumbing firm rebuilds its rating after a bad winter
Embermantle Heating, a four-van plumbing and heating firm, finished a cold winter with a rating of 3.6 from 58 reviews after a run of missed appointments during a staffing gap. The owner started a routine: every engineer sends the review link from the van when the job is signed off, the office replies to every review within a working day, and the Monday meeting reads the previous week's reviews. The eleven negative reviews from the winter each got a short reply acknowledging the delay and naming the change made, a dedicated scheduler. Over the next five months the firm added 83 reviews, 74 of them five stars, and the rolling ninety-day rating reached 4.8. Calls from the profile rose from around 90 a month to just over 150, with no change to the listing itself and no advertising. The scheduler was paid for within the first month.
When reviews that were on the profile have gone, the diagnosis for disappearing Google reviews separates filtering, deletion by reviewers and profile changes before any appeal.
To review the whole routine rather than one problem, the customer reviews and reputation audit scorecard rates asking, platform coverage, replies, display and the handling of themes on every platform the business is judged on, and ranks what to fix first.
Common mistakes
- Asking only when things go well. Selective solicitation breaches the platform rules and produces a rating readers do not trust.
- Running a review drive, then stopping. A burst followed by silence looks unnatural and leaves the profile stale within a year.
- Arguing in public. The reply to a negative review is read by hundreds of prospects; the reviewer is the least important reader.
- Ignoring positive reviews. No reply tells the next reader that nobody is watching.
- Treating fakes as an emergency. Flag, reply calmly, keep the record, and let genuine volume do the work.
- Collecting reviews on a device in the shop. The help centre documents review stations as a cause of removal, and the reviews rarely survive.
- Never reading reviews together. Each one is an anecdote; a quarter of them is a diagnosis the business is paying nothing for.
The local presence hub sets reputation beside community work; the profile tutorial is where the review link, the posts and the question-and-answer section are set up.
