Key takeaways
- Review management is five jobs, not one: ask, route, reply, escalate and use. Replying is the easiest of the five and the one most teams stop at.
- Invite every customer who reaches a defined trigger event, with the same link. Screening for happy customers first is what Google's policy describes as selectively soliciting positive reviews.
- Google prohibits incentives for posting, changing or removing a review. The FTC's consumer reviews rule, in force since 21 October 2024, adds civil penalties for knowing violations of its fake-review and review-suppression bans. Follow whichever rule is stricter. This is general information, not legal advice; confirm the details for your situation with counsel.
- Write escalation triggers before you need them. A hard review is never answered with a threat or an accusation you cannot prove.
- Measure the system, meaning invitation coverage, reply time and theme decisions closed, rather than a target star rating that pushes staff toward gating.
- Yelp's content guidelines say businesses should not ask customers for reviews, so keep Yelp out of any request flow.
01
What is review management, and what does a complete system include?
Review management is the set of routines that decide how customer reviews are requested, where they land, who answers them, which ones escalate and what the business learns from them. A complete system covers five jobs: ask, route, reply, escalate and use. Most teams only do the third.
That imbalance is understandable. Replying is the step platforms make easiest: Google lets a verified business answer its reviews directly, so the work lands on whoever holds the login and happens whenever they notice a notification. Asking depends on someone remembering. Routing, escalating and using depend on nobody in particular, so they quietly do not happen.
This guide is the operating layer around those jobs. It does not repeat the platform walkthrough in How to Get More Google Reviews Without Breaking the Rules, the reply and escalation craft in A Review Reply Should Solve the Next Step, Not Win an Argument, or the analysis method in Review Mining Should Preserve the Complaint, Not Just Harvest the Phrase. It shows how those pieces connect and where the rules draw hard lines.
| Stage | The question it answers | Owner | What failure looks like |
|---|---|---|---|
| Ask | Which customers are invited, when, and through which link? | Whoever owns the customer moment, such as the service lead or front desk | Reviews arrive in bursts, only in weeks when the owner pushes |
| Route | Where does each review land, and who sees it within a day? | One named queue owner plus a backup | Reviews are noticed only when someone happens to open the profile |
| Reply | What does the public answer say, and how soon? | A trained responder | Copy-pasted thanks on every review, silence on the difficult ones |
| Escalate | Which reviews need someone beyond the responder? | A manager, with counsel for legal threats | Staff argue in public, or a safety claim sits unread for a week |
| Use | What changes because of what customers said? | Operations and marketing together | Reviews are admired in meetings and never acted on |
02
Which platform and legal rules shape a review management program?
The rules differ by platform, so a single “send everyone to review us” script tends to break at least one of them. Start from what each source says, read in October 2026, and design the workflow to the strictest of them.
- FTC staff guidance says a business may respond publicly to a negative review, may contact the reviewer to resolve the problem and may ask a satisfied customer to update a review.
- The same guidance says the rule does not itself ban offering an incentive to take down a negative review, but that doing so could be an unfair practice under the FTC Act. Google's policy is stricter and prohibits it outright. In practice, follow Google.
- The rule does not stop a business from sorting the reviews on its own site by helpfulness or rating, though the FTC's own Q&A cautions that arranging reviews so negative ones are hard to find could still be unfair or deceptive. It does bar claiming that a review section represents all or most reviews when some were suppressed because of their rating or sentiment.
| Source | What it says | What it means for the workflow |
|---|---|---|
| Google Maps user-generated content policy | Merchants may not offer incentives such as payment, discounts or free goods or services for posting a review, or for changing or removing one. They may not discourage or prohibit negative reviews or selectively solicit positive ones, and the policy lists staff review quotas and pressure on the premises as problems. | Ask everyone, never screen first, offer no reward, set no staff quota. |
| Google Business Profile help | A business must be verified before it can reply to reviews. A review you believe violates Google's content policies can be flagged. Customers can be sent to a Google link or a QR code to leave a review. | Replying and flagging need a verified login. Flag against a named policy, not because the review stings. |
| Yelp content guidelines | Businesses should not ask customers to write reviews. | Leave Yelp out of invitation flows entirely. |
| FTC Rule on the Use of Consumer Reviews and Testimonials (16 CFR Part 465) | In effect since 21 October 2024. Covers fake or false reviews, buying positive or negative reviews, insider reviews, company-controlled review sites, review suppression and fake social media indicators. Courts may impose civil penalties for knowing violations. | Write down that no incentive depends on the rating. Never threaten a reviewer. |
03
How do you ask for reviews without gating them?
Review gating means screening customers by satisfaction and sending only the pleased ones to the public platform. Google's policy language names the practice when it bars merchants from selectively soliciting positive reviews. The safest design is one where selection cannot happen, because the invitation depends on an event and never on an opinion.
The rule is short: every customer who reaches the trigger event receives the same invitation, through the same channel, with the same link. A job marked complete in the CRM or field-service system is a good trigger because it is a record, not a staff member's guess about the customer's mood. Staff can still say thank you in person. They just cannot choose who gets the link.
A private route for complaints is worth having. We read Google's policy as allowing a private contact option next to the public link and as disallowing one instead of it. If you are unsure, show both, and let the customer choose.
Repeat customers need a rule too. If the same customer books several jobs a year, decide whether you ask after each job or only after the first and then again after a long gap. Whatever you choose, apply it to everyone, and stop asking anyone who has already reviewed that location. Google also notes that reviews are available only in certain countries and for certain business categories, so confirm your listing can receive them before you build a program around it.
- Pick a send window and keep it constant, so a change in volume means something. A day or two after a service job is a sensible starting point, though we know of no study that sets the ideal.
- Send one reminder at most, then stop.
- Use one link per location. Google's help describes a link or QR code for exactly this purpose.
- Do not hand staff wording for customers to use. Google's policy says merchants should not request that specific content be included.
- Do not attach a reward. A discount or prize for posting a review is an incentive under Google's policy.
- If the invitation goes out by text or an automated call, consent rules may apply. See the TCPA Compliance article before you build that step.
04
How should reviews be routed and replied to?
Platform notifications go to whoever holds the login, so the first routing job is to pull every review into one queue. The minimum record is the platform, location, date, rating, theme tag, owner, status and the time the reply was posted. A shared spreadsheet is enough to start; the discipline matters more than the tool.
Triage by what the review says rather than by its star count. A three-star review that describes a billing error needs a different owner from a five-star review that mentions a staff member by name. The windows below are starting points we suggest, not benchmarks from a study. Set yours to what the team can keep every week.
| Review type | First action | Who | Starting window |
|---|---|---|---|
| Positive and specific | Thank them and name the detail they mentioned, with no sales pitch | Responder | 2 business days |
| Negative, describes a real event | Look up the job, reply once publicly with an acknowledgement and a direct contact route, then contact the customer | Responder and service lead | 1 business day |
| Negative, no matching customer | Ask once, publicly, for details so the record can be found; flag it if the review is not about your business | Responder | 1 business day |
| Mentions safety, health, legal action, discrimination or a named employee | Escalate before replying | Manager, counsel if legal | Same day |
| Looks fake, competitor-made or conflicted | Preserve a screenshot and date, flag against the specific policy | Responder and manager | 2 business days |
05
What should a public reply actually contain?
A public reply has two readers. One is the reviewer. The other is the next customer scanning the profile, who will judge you by how the response sounds far more than by the review itself. Write for the second reader: specific, calm, and ending in a next step.
A workable structure runs four beats. Acknowledge the specific issue in one sentence. Add a fact only if it corrects the public record without exposing private details. Give a direct way to continue the conversation offline. Stop. Here is an illustrative reply, written for this article and not taken from any real business: “Thank you for telling us the technician arrived outside the window we gave you. That is not the standard we aim for, and our dispatch manager would like to hear the details. Please call the office number on our profile or reply to the email we sent today.”
Practices that handle health information face extra limits on what a reply may confirm, including whether the reviewer was a customer at all. Local SEO for Dentists and Dental Practices covers the patient-privacy side. For any business, the safe habit is to confirm nothing the reviewer has not already said.
06
When should a review be escalated, flagged or left alone?
Escalation is a list of triggers agreed in advance, because the middle of an argument is the wrong moment to decide who handles legal threats. The table pairs each situation with the move that protects you and the move that creates a second problem.
| Situation | Do | Do not |
|---|---|---|
| Spam, off-topic content, conflict of interest or a paid review | Flag against the specific policy and keep a dated screenshot | Assume it will be removed, or answer with accusations |
| Customer disputes a bill or the quality of work | Resolve it privately first, then say you would welcome an updated review if they feel it is fair | Offer money or a discount in exchange for removal. Google prohibits it, and FTC staff say it could be an unfair practice |
| Threats, harassment or a safety allegation | Escalate to a manager the same day, and to counsel if a legal threat is involved | Reply from the front desk while angry |
| A review names an employee | Let the manager handle it, and keep personnel details out of the public reply | Describe discipline or performance in public |
| A factual error you can disprove | Correct it once, with facts only | Accuse the reviewer of lying. The FTC rule bars false accusations made knowingly or recklessly |
| You are considering legal action against a reviewer | Only after counsel confirms the claim has a basis | Send a groundless legal threat. The FTC rule treats unfounded or groundless legal threats used to remove a review as suppression |
07
How do you use reviews so the business actually changes?
Once a month, tag each review with a theme such as speed, price, communication, quality, staff or billing. Count the themes and read the negative reviews in full. Then give each theme one decision: fix it, test a fix, accept it with a written reason, or escalate it. A theme list without a decision beside each line is just a more organized way of ignoring customers.
Treat counts as signals, not rates. Reviews come from people who chose to write, so ten complaints about parking may be a real problem or may be one lively week. Review Mining Should Preserve the Complaint, Not Just Harvest the Phrase explains how to keep the original wording, which matters more than the tally.
Reuse outside the platform raises its own questions. If you quote a customer on a service page or in an ad, you are making a different kind of claim from a review that sits on Google. A Testimonial Needs Permission, Context, and an Expiry Date covers that step. When you can match a review to a job, record the link in the CRM so service history and feedback sit together. When you cannot, do not guess.
08
What should a review management dashboard track?
Track the health of the loop as well as its output. A star average tells you how customers feel on balance. It does not tell you whether you are asking everyone, answering on time or acting on what you hear.
- Avoid a target rating. A goal like “reach 4.8” invites exactly the filtering Google's policy prohibits, because the cheapest way to hit it is to stop inviting unhappy customers.
- Assign the queue to one person and a backup. That single decision repairs more of the system than any tool purchase.
| Metric | How to calculate it | What it tells you | How it gets misread |
|---|---|---|---|
| Invitation coverage | Completed jobs that received an invitation divided by completed jobs | Whether the ask is automatic or depends on memory | Pushing it past what customers will tolerate, with repeat reminders |
| Reviews per invitation | Reviews posted divided by invitations sent, by channel | Which channel and timing work | Read as a satisfaction score. It is a response rate |
| Median reply time | Hours from review posted to reply posted, by rating | Whether routing works | Fast replies to easy reviews while hard ones wait |
| Reply coverage | Reviews with a reply divided by all reviews, by rating | Whether negative reviews are being avoided | Counting a generic one-line thanks as engagement |
| Theme decisions closed | Themes with a recorded decision divided by themes raised | Whether reviews change anything | Closing items by relabeling them |
| Rolling 90-day rating by location | Average of ratings posted in the window | Direction of travel at each site | Reacting to a swing in a location with very few reviews |
09
How do you set up review management in the first 30 days?
Start with ownership, not software. In week one, name the queue owner and a backup, list every location and platform where reviews appear, and put the escalation triggers from the table above on one page that a manager has signed. Nothing else holds until a named person is on the hook for the queue.
In week two, build the single queue and backfill the last 90 days of reviews into it. Reply to anything unanswered that deserves an answer, starting with the negative reviews that describe a specific event, and contact those customers directly where you can identify them. Do not mass-edit old replies. A rewritten reply history looks odd and fixes little.
In week three, wire the ask to the trigger event: one invitation, one link per location, one reminder, and a log of every invitation so coverage can be calculated. If the invitation goes by text, settle consent first. In week four, hold the first theme review. Read the month's reviews together, tag the themes and leave with at least one decision and an owner for it.
After that the system should run on a fifteen-minute weekly check of the queue and a monthly theme meeting. If it needs more than that to stay alive, it is too complicated for a team that has other work.
Multi-location businesses add one complication: the people who know what happened at a branch are not the people who hold the central login. Keep one central queue, and give each location manager a way to answer a factual question within a day, such as a short form or a named contact. The central responder writes the reply and the local manager supplies the facts. That split is our suggestion, not a platform requirement, and it keeps the tone consistent and the facts right.
010
What usually goes wrong in review management?
The common failures are organizational rather than technical. A well-meaning manager starts a free-coffee-for-a-review promotion without knowing it breaks Google's policy. A branch sets a staff quota, which the policy names as a problem. Someone with the login replies in anger at ten at night. A rating target creeps into a performance review. None of these needs bad intent, only an unwritten process.
Insider reviews are the quietest risk. The FTC rule covers reviews written by company insiders, and FTC staff say a business is not liable merely because an employee happened to receive a standard invitation and wrote a false review on their own. Staff writing glowing reviews because a manager asked is a different matter, and the rule is aimed at it.
Finally, platform policies move. Re-read the Google, Yelp and FTC pages in the sources below each quarter and note the date you did. This article reflects what those pages said in October 2026, and it is general information, not legal advice.
Was this useful? Send it to someone who owns this problem.
Primary sources and further reading
Use the source material to validate details against your own context and current platform configuration.
- Google Business Profile Help: Manage customer reviews
- Google Business Profile Help: Tips to get more reviews
- Google Maps User Generated Content Policy: Prohibited and restricted content
- Yelp: Content Guidelines
- FTC: The Consumer Reviews and Testimonials Rule, Questions and Answers
- FTC: Final rule banning fake reviews and testimonials (press release, 14 August 2024)
- Legal Information Institute: 16 CFR Part 465, Rule on the Use of Consumer Reviews and Testimonials
This field note follows the XenGrowth editorial policy: primary sources where available, visible limitations, material review dates, and no invented first-hand experience.
Stay with the problem
Related field notes
AI and answer enginesAI Search Optimization: What Moves Citations and What Is Folklore
SEO & GEOTopical Map: Build a Cluster Plan Search and AI Can Read, and Know What Topical Authority Is Not