Google Business Profile Guide
Published

Multi-Location Review Management: How to Run Reviews Across Every Location

19 min read
Multi-Location Review Management

One location with a review problem is a bad week. Forty locations with a review problem is a brand problem, and it usually goes unnoticed until the group rating has already moved.

Review management across many locations is not the single location job repeated. The work changes shape. Volume stops being something a manager handles between other tasks. Consistency becomes a real risk, because every reply is published under your brand name by somebody you may never have met. And the rules get sharper, because a policy breach at one location can reach every profile you own.

This guide covers the operating model, the platforms that matter, response times that survive contact with reality, the templates worth having written down, and the mistakes that cost brands their ratings. If you only manage Google, we have a narrower guide on managing Google reviews for multiple locations.

What multi-location review management involves

Four jobs, running continuously, across every location and every platform you appear on.

The jobWhat it means at scaleWhere it breaks
CollectingAsking every customer, at every location, in a way that is consistent and within the rulesOne location stops asking and quietly falls behind the others
MonitoringKnowing a review landed, at which location, within hours rather than weeksNobody is watching the platforms outside Google
RespondingA reply that sounds like your brand and like that location, inside an agreed timeThe hard ones sit unanswered because nobody is sure who owns them
LearningSpotting the pattern across locations rather than reading reviews one at a timeEveryone reads their own reviews and nobody reads them all

Most brands do the first three badly and skip the fourth entirely. The fourth is the one that pays, because reviews are the cheapest operational feedback you will ever get. If three locations all get complaints about waiting times in the same month, that is an operations problem presenting itself as a marketing problem.

Why one location is everyone’s problem

This is the part that makes multi-location different, and it is underplayed almost everywhere.

A single location manager who decides to get clever with reviews can create consequences that reach the entire group. Not just their own profile.

  • Review gating. Asking happy customers for a public review while routing unhappy ones to a private form breaks Google’s Maps user generated content policy. Some review tools still ship this as a feature, which is how well meaning managers end up doing it.
  • Incentivised reviews. Offering anything in exchange for a positive review, as opposed to neutrally asking for feedback, is against platform policy and against the FTC’s 2024 rule on consumer reviews in the United States.
  • Staff and family reviews. Common, well intentioned, and a clear policy breach. It also tends to be obvious, because the accounts have no other review history.
  • Buying reviews. Rare at corporate level, not rare at all at location level when a manager is under pressure about their rating.

When a platform notices a pattern, the response is rarely surgical. Reviews get removed in bulk, which drops a rating overnight. Profiles can be suspended, which starts a reinstatement process with no published turnaround time. If the profiles sit in one account or group, the investigation does not politely stop at the one location that caused it.

The practical rule: every location follows the same collection policy, and the policy is written down somewhere they have actually read.

This is worth more than any tooling decision. A brand with mediocre tools and a clean collection policy is in a far better position than one with a good dashboard and a few locations quietly gaming their ratings.

Who replies: central, local, or both

Every multi-location review programme has to answer this before anything else. There are three models and the third is where most brands land once they have tried the other two.

Comparison of three review response models for multi-location brands: fully central, fully local, and hybrid with central policy and local voice
The three models. Most networks past about 20 locations end up hybrid.

Fully central

Head office replies to everything. The voice is consistent and the quality is controlled. It works well up to a point, and the point is usually volume. It also produces replies that locals can tell were not written locally, because the reply cannot reference the member of staff the reviewer named.

Fully local

Each location replies for itself. The replies are specific and genuine when they happen. The problem is the variance. Some managers reply within the hour and write something human. Others have not opened the profile since the handover. You will not find out which is which until you look.

Hybrid, which is what usually works

Central owns the policy, the templates, the escalation rules and the reporting. The location owns the voice and the detail. WiserReview put this well: centralise the policies, decentralise the personality.

In practice that means head office decides how fast a one star review gets answered and what is never said in public, and the location decides how to say thank you for mentioning Sarah on the front desk.

Which platforms actually matter for you

Every guide lists Google, Yelp, Facebook and TripAdvisor and leaves you to work out the rest. The honest answer is that Google matters for everyone and the second platform depends entirely on what you sell.

Your sectorGoogleThe one that matters nextAlso worth watching
Restaurants, bars, hospitalityYes, alwaysTripAdvisor, and delivery platform ratings if you deliverOpenTable, Facebook
HotelsYesBooking.com and TripAdvisor, which often matter more than GoogleExpedia, Trustpilot
Healthcare, dental, clinicsYesThe sector directories patients actually use in your countryFacebook, and practitioner level profiles
Home and trade servicesYesCheckatrade, Trustpilot or the equivalent trade directory locallyFacebook, Nextdoor
Retail chainsYesTrustpilot for the brand, Google for each storeFacebook, product review platforms
Professional servicesYesIndustry and association directoriesTrustpilot, LinkedIn recommendations
AutomotiveYesManufacturer and dealer rating systemsTrustpilot, Facebook

Two rules for the list. Monitor everywhere you can be reviewed, because a one star you never saw is worse than one you answered badly. But only run active collection on the one or two platforms that actually influence your buyers, because splitting review requests across five platforms gets you a thin presence on all of them instead of a strong one on the platform people check.

Response times and who handles what

Pick times you can actually hit. An SLA nobody meets is worse than no SLA, because it becomes a number people learn to ignore.

ReviewTarget responseWho writes itEscalate when
5 star, no textWithin a week, or skip itLocation, or automateNever
4 to 5 star with text2 to 3 working daysLocationNever
3 star2 working daysLocationIt names a staff member
1 to 2 star, ordinary complaint1 working dayLocation, using the agreed templateNo resolution after one exchange
Safety, discrimination, legal threat, dataSame day, acknowledge onlyHead office, never the locationAlways, immediately
Suspected fakeReport first, reply secondHead officeAlways, with evidence logged

The bottom two rows are the ones to get right. A location manager replying to an allegation about discrimination or a data breach is how a bad review becomes a bad headline. Those should route to somebody who has spoken to a lawyer before.

Set one number for the whole network, then measure it per location. Brand average response time hides the location that never replies, which is the exact location you need to find.

The five replies worth writing down

Templates get a bad name because they get used badly. The point is not to paste them. It is to give a manager at 7pm a correct starting point so the reply happens at all. Change the specifics every time.

1. Positive, with detail

Thanks [name], that is good to hear. [Specific thing they mentioned] is something the team here works hard at, and I will pass this on to [staff member if named]. See you next time.

2. Positive, no text

Thanks for the five stars, [name]. Much appreciated.

Short is correct here. A long reply to a rating with no text reads as padding.

3. Negative, your fault, fixable

Hi [name], you are right and I am sorry. [What went wrong, plainly]. We have [what changed] so it does not happen again. If you would like to talk it through, I am on [contact]. [Manager name], [location].

Three things make this work: admitting it without hedging, saying what changed, and signing it with a real name. Most brand replies do none of the three.

4. Negative, disputed or unclear

Hi [name], thank you for raising this. That is not the experience we aim for and I would like to understand what happened. Could you contact me on [contact] so I can look into it properly? [Manager name], [location].

Do not argue the facts in public, even when you are right. You are writing for the next reader, not the reviewer.

5. Sensitive, acknowledge only

Hi [name], thank you for contacting us. We take this seriously and it has been passed to our [team]. Please contact us at [contact] so we can look into it directly.

For healthcare, never confirm somebody was a patient. For anything involving an allegation, the public reply exists to show you responded, not to resolve it. Resolve it privately.

Collecting reviews without breaking the rules

More reviews is usually the right goal. How you get there is where brands get into trouble.

What works

  • Ask everyone, every time. Same request, same timing, no filtering by how the visit seemed to go.
  • Ask close to the moment. A restaurant asks that evening. A trade asks on completion. A clinic asks the next day. Guidance on timing and methods is in Google’s own tips for getting more reviews.
  • Make it one tap. A direct review link, or a QR code at the counter. Every extra step loses people.
  • Give each location its own link. Obvious, and still the most common setup error in a new rollout.

What gets you into trouble

  • Gating by sentiment. Covered above. It is the single most common serious breach and it is often built into the tool.
  • Incentives tied to positivity. A prize draw for any honest review is generally acceptable. A discount for a five star review is not.
  • Asking staff, friends or family. Tempting for a new location with no reviews. Easy to spot and it taints the profile.
  • Bulk requests to an old list. A sudden burst of reviews on a profile that normally gets two a month looks exactly like what it is.

If a location is behind, the fix is asking more consistently, not asking differently.

When one location drags the group down

Here is the arithmetic that explains why head office cares about a single branch.

Bar chart showing nine locations averaging 4.6 stars and one at 2.8, pulling the group average down to 4.42
One weak location moves the number everybody else is compared on.

Nine locations at 4.6 and one at 2.8 gives a group average of 4.42. That does not sound dramatic until you consider that the gap between 4.4 and 4.6 is often the gap between being shortlisted and not.

The instinct is to push harder on review collection everywhere to dilute the bad one. That is slow, expensive and treats the symptom. A location sitting at 2.8 when its siblings are at 4.6 almost never has a marketing problem. Read its last twenty reviews and the cause is usually named explicitly, repeatedly, by different people.

Fix the operation, then fix the rating. In that order.

Getting a fake or policy-breaking review removed

Every multi-location brand collects some reviews it should not have to live with. A competitor, an ex employee, a customer of a different branch, somebody who has confused you with another business entirely. Removal is possible, slower than you want, and only works on specific grounds.

What actually qualifies

  • Not a real experience. A review from somebody who was never a customer, including competitor and ex staff reviews.
  • Wrong business. Meant for a different company, or a different branch of yours. More common than people expect in a network.
  • Prohibited content. Abuse, discrimination, personal information, off topic political content, advertising.
  • Conflict of interest. Reviews from the business, its staff or its competitors.

What does not qualify, however unfair it feels

  • A harsh but genuine opinion about a real visit
  • A one star with no text from a real customer
  • A complaint about something outside your control, such as the weather or parking you do not own
  • A review you simply disagree with

The process that works

  1. Report it on the platform first, using the flag on the review itself, and pick the specific policy it breaks rather than a general complaint. Google documents this in manage customer reviews.
  2. Log it. Date, location, reviewer name, the policy you cited, screenshots. If you escalate later you will need this and nobody remembers it three weeks on.
  3. Reply publicly in the meantime, carefully and briefly. The review is visible while you wait, and a measured reply does more for the next reader than silence.
  4. Wait properly before chasing. Reporting the same review repeatedly from several accounts does not help and can look like manipulation.
  5. Escalate through support if it is clearly in breach and the first report did nothing. Having the log makes this a short conversation rather than a long one.

Set the expectation internally that most removal requests fail and that is normal. The reviews worth fighting are the ones that are demonstrably fake or clearly breach a written policy. Everything else is answered, not appealed.

Rolling this out across an existing network

Starting from scratch on 40 locations that have each been doing their own thing for years is a different problem from setting it up on day one. This order keeps it manageable.

StageWhat you doWhy this order
Weeks 1-2: find out what is actually happeningList every location, every platform it appears on, who has access, and what each one is currently doing to collect reviews.You cannot write a policy until you know which locations are already gating, incentivising or doing nothing.
Weeks 3-4: stop the risky thingsKill any sentiment gating, incentives tied to ratings, and staff review requests. One clear instruction, in writing.This is a risk control and it should not wait for the rest of the programme.
Month 2: agree the model and write it downWho replies, response times, escalation rules, the five templates, and what is never said in public.Everything after this depends on the rules existing somewhere people can read them.
Month 2-3: one inbox and per location reportingGet every location’s reviews into one place and start reporting the five numbers per branch.You need the baseline before you can show anybody they improved.
Month 3: fix collection, worst locations firstGive each location its own link or QR, set the ask into the normal customer flow, and start with the branches that have the fewest recent reviews.Collection compounds, so the sooner the weakest locations start the better.
OngoingMonthly reporting per location, quarterly audit of what locations are actually doing.The audit matters. Policies drift back within a year without one.

The two week audit at the start is the part people want to skip. It is also the part that finds the location quietly running a gating tool nobody at head office knew about.

If you are in a regulated industry

Healthcare, financial services, legal and childcare all carry constraints that generic review advice ignores.

  • Never confirm somebody was a patient or client. Even “thank you for visiting us” can be a disclosure. Reply in a way that works whether or not they were.
  • Do not discuss any detail of the service provided. Not even to correct something factually wrong in the review.
  • Route anything clinical or financial to a named person. This is not a location manager decision.
  • Keep a record. In regulated sectors the reply is part of the compliance trail, and “we responded on the day” is a useful thing to be able to prove.
  • Check whether your regulator restricts soliciting reviews at all. Some professional bodies do.

The safest pattern in these sectors is a short, warm, non specific public reply plus a real private channel. It looks responsive without saying anything it should not.

The five numbers worth tracking per location

MetricWhy it mattersA reasonable target
Average ratingWhat customers compare you onAt or above the local competitors, not an absolute number
Review velocityNew reviews per month. Recency counts for more than totalSteady, with no month at zero
Response rateShare of reviews answeredAbove 90% on anything with text
Time to respondMeasured per location, not averaged across the brandInside the SLA you set, for 9 out of 10
Unanswered negativesThe single most damaging gapZero, always

Two of those deserve a word. Velocity beats total. A location with 40 reviews from this year usually reads better than one with 200 that stop in 2023, because a reader checks the dates. And unanswered negatives is the metric to put in front of people, because it is the one that is both visible to customers and trivially fixable.

At any real scale this is a reporting job rather than a reading job. Our multi location review management software puts every location in one inbox with these numbers per branch, which is the only practical way to find the location that stopped replying in March.

What to automate and what never to

AutomateKeep human
Sending the review request, every timeAny reply to a one or two star review
Alerting the right person when a review landsAnything naming a member of staff
Drafting a reply for a human to approveAnything clinical, financial or legal
Replying to 5 star ratings with no textFirst reply after an operational failure
The monthly reportThe decision about what to do with it

Automatic replies to positive ratings with no text are fine and save real time. Automatic replies to complaints are obvious to readers and make a bad situation worse. If a tool offers AI replies, the right setting is draft and approve, at least until you trust it on your own reviews.

The mistakes that cost the most

  1. Letting one location run its own collection method. This is how gating enters a brand. Audit what each location is actually doing, not what the policy says.
  2. Measuring the brand average only. It hides every problem worth knowing about.
  3. Copy pasted replies across locations. Readers compare locations. Identical replies under different branch names look automated because they are.
  4. Only watching Google. Fine for some sectors. In hospitality and healthcare it means missing the platform that actually drives bookings.
  5. Treating a bad location as a review problem. Covered above and worth repeating, because it is the most expensive mistake on this list.
  6. Letting the hard ones sit. Nobody wants to answer the angry one. Deciding in advance who does is the whole point of the escalation table.

Multi-location review management FAQs

What is multi-location review management?

It is the process of collecting, monitoring, responding to and learning from customer reviews across every location a business operates, on every platform where it can be reviewed. It differs from single location review management mainly in governance: deciding who replies, enforcing one collection policy across people you do not directly manage, and reporting per location rather than per brand.

Should head office or each location reply to reviews?

Most networks past about 20 locations end up hybrid. Head office owns the policy, the templates, the escalation rules and the reporting. The location writes the reply in its own voice. Fully central replies read generic to local customers, and fully local replies vary too much in quality and speed.

How fast should we respond to reviews?

Set times you can actually meet and measure them per location. A workable baseline is one working day for one and two star reviews, two to three days for everything else with text, and same day acknowledgement for anything involving safety, discrimination or legal threats. The exact numbers matter less than hitting them consistently.

Can one location get the whole brand’s profiles suspended?

It can certainly create a problem that reaches beyond itself. Review gating, incentivised reviews and staff reviews all breach platform policy, and when a platform investigates a pattern it does not necessarily stop at the location that caused it. Reviews can be removed in bulk and profiles can be suspended, so one collection policy across every location is a risk control rather than a nicety.

What is review gating and why does it matter?

Review gating is asking how a visit went, then sending happy customers to a public review site while routing unhappy ones to a private form. It breaks Google’s Maps user generated content policy, which prohibits selectively soliciting reviews from people likely to be positive. Some review tools still include it as a feature, which is why it keeps appearing in networks that never intended to break a rule.

How do we fix one location with a much lower rating?

Read its last twenty reviews before doing anything else. A location sitting well below its siblings almost always has a specific, repeatedly named operational cause. Collecting more reviews to dilute the bad ones is slow and treats the symptom. Fix the cause, then rebuild the rating with consistent collection.

Which review platforms should a multi-location brand monitor?

Google for everyone. After that it depends on sector: TripAdvisor and booking platforms for hospitality, trade directories for home services, sector directories for healthcare, Trustpilot for retail and professional services. Monitor everywhere you can be reviewed, but run active collection on only the one or two platforms your buyers actually check.

Should we use AI to reply to reviews?

For drafting, yes, with a human approving before it posts. For positive ratings with no text, fully automatic is fine. For complaints, automatic replies are obvious to readers and make things worse. The useful setting is draft and approve until the drafts are reliably good on your own reviews.

How many reviews does each location need?

Enough to look credible next to whoever it competes with in that town, which is a different number in every market. Check the three businesses currently ranking above it and aim for that range. Recency matters as much as volume, so a steady flow beats a one off push.