The Multi-Location Customer Operations Checklist: From Opening Day to Quarterly Review

Before you can operate anything, you need to know what's actually arriving at each location — and who, if anyone, is looking at it. The most common failure in multi-location customer operations isn't a bad process. It's a location that quietly has no process at all, because whoever set up its Google Business Profile left the company, or the survey link was pointed at an inbox nobody checks.

Week one isn't about strategy. It's an inventory: for every location, what channels exist, who owns them, and whether anyone would notice if a channel went silent.

What "connected" actually means

A location is connected when three things are true: someone owns its Google Business Profile with verifiable access (not a personal Gmail that leaves with an employee), every feedback channel it runs — reviews, post-visit surveys, in-store QR codes, direct email — routes to a person or inbox that's actually monitored, and there's a way to tell, at a glance, whether a given location has gone quiet versus whether it's simply had a quiet week.

That last point matters more than it sounds. A location with zero new reviews for three weeks could mean nothing is wrong, or it could mean the QR code sign fell down, the survey link broke, or a manager turned off notifications because they were "too noisy." Week one is when you find out which.

Checklist

• Every location's Google Business Profile has a verified owner with account-level access — not a personal email tied to one employee
• You have a list of every location and can identify, for each one, who currently has admin access to its listing
• Every feedback channel in use (reviews, post-visit survey, QR code, direct email) is mapped to a specific person or inbox
• You've tested each channel end to end at least once — scanned the QR code, submitted the survey, sent the test email — to confirm it actually reaches someone
• You have a way to flag a location that's gone silent (no reviews, no survey responses) versus one that's simply had a quiet week
• Access lists are stored somewhere that survives a manager leaving — not only in that manager's head or personal accounts
• You know, for each location, what happens to a piece of feedback the moment it arrives — does it sit in an inbox, or does someone see it that day

If you can't check most of these boxes yet, that's normal — it's exactly what week one is for. The goal isn't a perfect system by day seven. It's knowing where the gaps are before you build anything on top of them.

Once signals are flowing, the next failure mode is ambiguity: feedback arrives, gets seen, and then sits — because it's not clear whose job it is to respond, or what counts as urgent enough to escalate. Month one is when you assign ownership explicitly, in writing, rather than letting it default to "whoever happens to notice."

The decision that shapes everything else: delegate or centralize

There are two common models for who actually writes replies, and both are used successfully by real multi-location operators. Neither is universally right — but each has a specific way it breaks.

Delegated to the location manager. Each store's manager (or a designated staff member) monitors and responds to feedback for their own location.

• Where it works: managers know the regulars and the context behind a complaint, and can respond with specifics ("sorry about the wait — Tuesdays are short-staffed until we finish hiring") that a centralized team can't.
• How it fails: tone and quality drift between locations — one manager writes warm, specific replies, another copies the same three sentences into everything. Nobody at HQ notices a pattern repeating across five stores, because each manager only sees their own. And when a manager who was good at this leaves, the replacement often isn't, with no handoff of what "good" looked like.

Centralized at headquarters. One person or a small team handles feedback and replies for every location.

• Where it works: consistent tone and quality, someone sees patterns across the whole chain (the same complaint showing up in four locations is a signal, not four separate incidents), and response times don't depend on how busy any one store's manager is that week.
• How it fails: replies read generic because the person writing them has never been to the store and doesn't know the context. Volume becomes a bottleneck as locations are added — a team sized for eight stores doesn't scale cleanly to twenty. And store managers start to see feedback as "not my problem," which quietly disconnects the people best positioned to actually fix what's wrong from the signal telling them it's wrong.

A workable middle path many chains land on: location managers reply to routine feedback within guardrails (an approved tone, a list of things they can't promise — refunds, discounts, policy exceptions), and anything above a defined severity or anything requiring a commitment gets escalated to HQ. That only works if "defined severity" is actually written down — not left to judgment call by call.

What to define in writing this month

• Who is responsible for reviewing new feedback at each location — a named person, not a role that's currently vacant
• Who has authority to publish a reply, and whether there's an approval step before it goes out
• The specific conditions that trigger escalation to HQ (examples: mentions of illness or injury, threats of legal action, the same issue complained about for the third time in a month, any refund request above a set amount)
• A target response time, and what happens if it's missed — not just what the target is
• What a location manager is allowed to offer on their own authority (an apology, a specific fix) versus what needs sign-off (refunds, discounts, exceptions to policy)
• A written rule that no one asks customers to filter who gets invited to leave a review, and no one offers a discount, refund, or gift in exchange for a review. This is a Google Business Profile policy violation and, in a growing number of jurisdictions, an FTC compliance issue — and it's the mistake multi-location operators back into by accident. Once headquarters sets a review-count or rating target, some location, somewhere, will start hitting it by asking only happy customers or offering something in return. The rule needs to exist before the target does, not after someone gets caught.

There's a strong temptation, once feedback is flowing from every location, to immediately rank them — a leaderboard feels like accountability. Resist it for at least a quarter. Comparing locations before you have enough data doesn't produce insight; it produces noise dressed up as insight, and people make real decisions based on it.

Why early comparison goes wrong

A location that opened six weeks ago with twelve reviews and a location that's been running for three years with four hundred don't sit on the same footing — one bad week is a rounding error for the second and a third of the sample for the first. Seasonal effects compound this: a location near a university looks very different in exam week than in July, and if you're only three months in, you don't yet know which weeks are which for that specific location.

The practical harm isn't abstract. A manager whose location looks worst on a young dataset gets treated as underperforming — coached, scrutinized, maybe passed over — when the actual difference might be entirely explained by sample size or seasonality. Meanwhile a manager whose small location got lucky for six weeks looks like a star. Both conclusions are wrong, and both get acted on if you let them.

There's also a behavioral cost. Once managers know they're being ranked, some start optimizing for the metric instead of the thing the metric was supposed to represent — which is exactly the pressure that leads back to the compliance problem from month one.

What to do instead, for the first quarter

Spend the quarter accumulating, not judging. Track volume and rating per location so you know what "normal" looks like for each one individually — not yet against each other. Note anything unusual (a bad week tied to a known cause: a broken AC, a staffing gap) so it doesn't get mistaken for a trend later. By the end of the quarter, you should have enough data per location to tell the difference between a real pattern and a blip — and only then does comparing locations start to mean something.

Once you're past the first quarter, monthly review becomes the steady-state rhythm. The point of a monthly check isn't to catch every location up to date — it's to ask the same handful of useful questions consistently, so a real shift stands out against a stable baseline instead of getting lost in the noise of a system nobody looks at regularly.

The wrong question and the right one

The instinctive question in a monthly meeting is "why did this location's score drop this month?" That question sends everyone chasing a single number that moves for reasons that often have nothing to do with operations — a slow week, a handful of unrelated one-off complaints, normal variance in a small sample.

The more useful question is: "for the top complaint topic at this location, has its share of total feedback gone down since we addressed it?" That question is about a specific, named problem and whether a specific, named response worked — not about a score wobbling within its normal range.

What to look at

• Complaint topics by location, and whether each topic's share is trending up, down, or flat — not just this month's raw count
• Locations with no feedback at all this month (a silence flag, not just a low-score flag)
• Response time and unresolved-item backlog per location
• Anything escalated to HQ this month, and whether it was actually resolved or just acknowledged

What not to over-index on

• A single location's monthly average score in isolation, without the topic breakdown behind it
• Month-over-month comparisons for a location with unusually low feedback volume that month
• Comparing this month's number for one location directly against another without the context from the quarterly baseline

The monthly meeting should end with a short list of specific, named issues to watch — not a scoreboard.

The monthly routine watches for drift. The quarterly review looks backward and asks a harder question: did what we said we'd fix last quarter actually get fixed — and did it work?

This is the step most chains skip, because it requires going back to commitments made three months ago instead of just looking at the current state. But without it, "we'll look into that" becomes a phrase that costs nothing to say, because nobody ever checks whether it happened.

Three questions, in order

For every issue that was flagged and assigned last quarter, ask, in this order:

1. Did it get done at all? Not "was it discussed" — was the specific change (retraining on wait times, a menu fix, a staffing adjustment) actually made, on the ground, at the location.
2. If it got done, did the complaint topic's share actually drop? Go back to the monthly topic tracking from the section above. If the change was real and it worked, the specific topic it targeted should show a decline relative to total feedback for that location.
3. If it got done and the topic didn't improve, was it done wrong, or was it the wrong fix for the wrong root cause? These require different next steps. "Done wrong" means the same fix, executed better, might still work. "Wrong root cause" means the underlying diagnosis needs to be redone — a complaint about slow service might not be a staffing problem at all if the real bottleneck is a kitchen ticket system.

Skipping straight to "let's try something else" without going through step three is how the same complaint keeps reappearing every quarter under a different proposed fix.

Checklist for the quarterly review

• A list of every issue assigned last quarter, with a name attached to who owned it
• For each one: done, not done, or partially done — with evidence, not a status update from memory
• For each "done" item: has the relevant complaint topic's share moved, using the monthly tracking as the source
• For anything that didn't improve despite being done: a decision on whether it was execution or diagnosis that was wrong
• A short list of what gets assigned for next quarter, informed by what did and didn't work this time

A new location isn't a smaller version of an established one — it's a location with no baseline, run by a team still learning the space, often understaffed while hiring catches up. Treating it like every other store from day one, especially in how you compare and judge it, sets it up to look bad for reasons that have nothing to do with whether it's actually being run well.

The first 90 days

The first two weeks repeat week one of this checklist specifically for the new location: confirm Google Business Profile ownership, wire up every feedback channel, and confirm someone is actually watching them — new locations are exactly where a QR code sign gets left in a box in the back room.

From there through day 90, the goal is the same as quarter one for the whole chain, compressed: accumulate enough data to know what normal looks like for this specific location, while expecting more volatility than a mature store would show. A new team is still learning the menu, the till, the layout — feedback in the first month will reflect that learning curve more than it reflects the ceiling of how the location will eventually run.

When to start comparing it to other locations

Not on day 30, and not on a fixed calendar date at all — on a data threshold. A new location is ready to be compared once it has accumulated a volume of feedback in the same range as your other locations' typical monthly volume, not before. A store with three reviews isn't comparable to one with sixty, regardless of how many weeks have passed. Until then, track it against itself — is week 8 trending better than week 2 — rather than against the rest of the chain.

Checklist

• Google Business Profile ownership and access confirmed within the first week, using the same standard as every other location
• Every feedback channel tested end to end before or immediately after opening
• A named person responsible for reviewing this location's feedback, even if it's temporary coverage from HQ while the manager settles in
• A note on the account that this location is "new" so early numbers aren't read the same way as an established store's
• A defined feedback-volume threshold before this location is included in any cross-location comparison
• A check-in at day 30 and day 90 specifically for this location, separate from the regular monthly cycle

Every step above can be run on a spreadsheet and a shared calendar, and for a small chain, that's a completely reasonable way to run it. Three locations, one person checking a shared sheet weekly, a recurring calendar invite for the quarterly review — that setup works, and buying software to replace it wouldn't fix anything that's actually broken.

Where a spreadsheet holds up

For a handful of locations, a spreadsheet handles the monthly and quarterly routines fine: one tab per location for raw notes, one shared view for topic tracking, one row per assigned action item with a name and a due date. The discipline described in this checklist — actually reviewing it, actually asking the right questions, actually following up next quarter — matters far more than the tool it's kept in. A well-run three-location operation on a spreadsheet will outperform a badly-run twenty-location operation with the best software available.

Where it stops holding up

The pain doesn't show up gradually — it shows up sharply once a chain crosses a certain number of locations, and it's rarely the tracking itself that breaks first. It's the merging: pulling reviews, survey responses, and QR feedback from a dozen-plus locations into one place where a topic like "slow service" can be seen as one pattern across the chain instead of forty separate rows across forty separate tabs that nobody has time to read end to end every month. Manually copying data from each location's Google Business Profile, each survey tool, and each inbox into a shared sheet is a task that scales linearly with location count — and at some point the hours it costs each month exceed what it would cost to have it done automatically.

That crossover point is different for every operation depending on how much staff time is already going into this manually. It's worth actually estimating it rather than guessing — the ROI calculator walks through the hours a team currently spends on this kind of manual feedback work and estimates what that time is costing, which is a more honest starting point than assuming a tool is or isn't worth it.

What this looks like in OwnCrew Customer Ops

If you're past the spreadsheet stage, the relevant pieces of this checklist map onto two tiers. Respond covers the week-one and month-one work — a synced review inbox across every location, negative-review alerts so a bad review doesn't sit unseen, and basic analytics with location comparison so the monthly routine has real numbers behind it instead of a manual pull. Operate adds the channels this checklist assumes exist beyond reviews — surveys, QR, and email feedback in one inbox — plus topic, sentiment and severity classification and recurring-issue detection, which is what turns "slow service came up again" from a manual read-through into something visible without reading every entry by hand.

Assigning last quarter's fixes to a named owner and tracking whether they closed, the way the quarterly review section above describes, is still manual work today — worth saying plainly rather than implying otherwise. See the full breakdown on pricing, or read how this connects to the review side specifically in Multi-Location Business Review Management.