Franchise review responses: the hard part is who owns the reply

A regional pizza chain with 90 locations, call it Fontana’s, got a one-star review in Dayton last winter. The reviewer was specific: a burnt order, a forty-minute wait, and a shift manager named Rick who told her the coupon on her phone “wasn’t real.” Corporate’s outsourced response team replied inside the hour: “We’re sorry to hear your experience didn’t meet expectations. Your satisfaction is our top priority. Please reach out to our Guest Care line so we can make it right.”

She replied under it: “Do you even know which store this was? There’s no Rick at every Fontana’s.” The response named nothing, fixed nothing, and told every future reader that the brand answers its reviews from a building that has never been to Dayton. Same week, a different Fontana’s franchisee in Ohio got a review alleging a kid had an allergic reaction, and answered it himself with a defensive paragraph about how the allergen was clearly listed. A plaintiff’s attorney screenshotted that one.

Two failure modes, one week: the center that knows the brand but not the store, and the store that knows the customer but not the legal exposure. That is the entire problem of review response at multi-location and franchise scale, and here’s the thesis about it. The hard part is not the words. It’s governance: deciding who owns the reply. Corporate wants a consistent voice and legal safety. The local store has the facts, the names, and the actual relationship with the person who wrote the review. Get the split wrong in either direction and you produce one of those two failures on repeat. You either sound like a call center, or you let a single location say something that follows the whole brand around. Everything worth arguing about here is where you draw that line.

The common advice answers an ownership question with an org chart

There are two camps, and both are selling you a location-level or brand-level answer to a question that lives at the level of the individual reply.

Camp one says centralize for consistency. This is the agency pitch: hand us your logins, we’ll respond to every review across the network in an approved voice, your brand will finally sound like one company. It scales, it’s auditable, legal loves it. It also produces the Dayton reply, because the person writing it has no idea who Rick is and no way to find out before the clock the vendor is being paid to beat runs out.

Camp two says let each location handle its own. Give franchisees the keys, they know their customers, authenticity takes care of itself. And it does, right up until the location that answers an injury allegation with a lawyer-bait paragraph, or the one that gets into a public argument with a reviewer over a refund, or the three locations that never respond to anything at all and drag the brand’s network rating down with them.

Both camps are wrong for the same reason. They’re answering “who responds to reviews” with a name on an org chart, when the honest answer is “it depends on the review.” A five-star note about a great haircut and a review alleging a health-code violation are not the same object, and nothing about your corporate structure should force them down the same pipe. The unit of governance is the reply, sorted by risk. Not the location. Not the brand.

The pre-approval trap

The most common compromise is the worst of the three: let the local manager draft, but route every reply to corporate for sign-off before it posts. On paper it looks like the best of both worlds, local voice plus brand safety. In practice it fails twice.

It fails on voice, because a manager who knows their draft will be edited by someone at headquarters stops writing like a person and starts writing what they think will get approved. Pre-approval launders the local voice right back out of the reply. And it fails on speed, which is not a soft metric anymore. BrightLocal’s Local Consumer Review Survey 2026 found 74% of consumers now look specifically for reviews written in the last three months, and a growing share want to see the business reply within a day. A two-day approval queue at corporate is a direct tax on the one thing that makes responding worth doing. You added a gate to protect the brand and the gate is now the risk.

Pre-approval on everything is the wrong default. It belongs on exactly one category of review, which we’ll get to.

The governance model: a floor from corporate, the room from the store

Here is the split that actually works, and it is not fifty-fifty. Most of the reply belongs to the local store. Corporate owns a small, specific set of things, and the discipline is keeping that set small.

What corporate should own:

• The policy and the escalation map. Which categories of review never get a fast local reply, who they route to, and how fast. This is the single most valuable thing corporate produces, and almost nobody writes it down.
• Brand voice as a floor, not a script. Think dress code, not a line the employee reads off a card. “Warm, plain-spoken, never sarcastic, never argue about the refund in public, always sign with a real name” is a floor. A mandated opening sentence is a script, and scripts are how you get the Dayton reply at 90 locations at once.
• The legal-sensitive lane itself. Any review alleging injury, discrimination, a health or safety incident, a data or privacy problem, or naming active litigation gets written or cleared by corporate with counsel in the loop. Not because the local manager is careless, but because the downside is asymmetric and the local manager is the person most likely to say something true and defensive that becomes evidence.
• Network monitoring and the after-the-fact audit. Corporate watches the whole board, spots the location that stopped responding, and reads a rolling sample of published replies to catch drift. After the fact. Reading a sample is not the same as approving every one before it ships.
• Templates as raw material, never as mandated copy. A shared library the local manager pulls from and edits is a gift. The same library enforced word-for-word is the wallpaper effect with a brand logo on it.

What the local store must own:

• The specifics. What actually happened, checked with whoever was working, and the name of the person who is going to fix it. This is the part corporate structurally cannot supply, and it is the part readers can smell the absence of.
• The relationship and the offline follow-up. The public reply is a few sentences; the phone call that actually resolves it belongs to the person the customer might see again.
• Speed inside the guardrails. For everything outside the legal-sensitive lane, the local manager publishes without asking permission. That is the whole point of a floor: it lets you move without a gate.
• The judgment to escalate. The one thing you do train hard is recognizing a red-flag review and stopping. Not writing the reply, recognizing that this is not theirs to write.

Google’s own tooling assumes roughly this shape. Business groups let a brand share management of a set of locations and assign owners and managers per profile, which Google describes as a shared folder for your locations. The platform gives you a corporate owner over the group and a local manager on each profile. It’s handing you the governance structure and leaving the policy blank. The policy is the part you have to write.

Notice the tradeoff this resolves, because everyone names it as if it were unresolvable: consistency versus authenticity. The whole multi-location literature treats these as a dial you set once for the brand. They’re not. You don’t trade them off across every reply. You assign each reply to the layer that has the thing it needs. Consistency is a floor everything sits on. Authenticity is the default voice on top of it. The trade only becomes real in the legal-sensitive lane, where you knowingly accept a stiffer, slower, corporate-voiced reply because the alternative is a fast, authentic sentence that shows up in a deposition.

A worked example: the review that isn’t yours to answer

Copperline Plumbing franchises 60 locations. One of them, Meridian Copperline, is run by an owner-operator named Dana who is good at her job and answers her own Google reviews, usually well. On a Tuesday she gets this, one star:

“Copperline installed our water heater in March. Six weeks later it failed overnight and flooded our finished basement. Eight thousand dollars of damage and their tech Marcus told us on the phone it ‘wasn’t their unit’s fault.’ We have retained an attorney. Do not use this company.”

Dana’s instinct, and she is a good operator, is to defend her shop. The draft she starts writing goes something like: “We’re surprised by this review. Our records show the homeowner signed off on the installation, and Marcus is one of our most experienced technicians. The unit carries a manufacturer warranty and the failure appears to be a manufacturing defect, not an installation error. We’d be happy to discuss.”

Every sentence of that is defensible and every sentence is a mistake. It names an employee next to a fault dispute. It makes a factual claim about cause (“manufacturing defect, not installation error”) before anyone has inspected anything, in writing, on a page a lawyer is already reading. It puts the words “signed off” in public, which reads to the customer as “we’re going to blame you.” Dana is trying to protect Marcus and her shop, and she is drafting the opposition’s first exhibit.

Under the governance model, this review never reaches Dana’s reply box as hers to answer. It trips two red flags: a stated legal action with a property-damage claim, and a named employee. It auto-holds and routes to the brand’s reputation lead, who loops in the counsel liaison. The reply that goes out is short and says almost nothing:

“We take reports like this seriously and we want to understand exactly what happened. Our regional service manager will contact you directly today to work through it with you. Thank you for letting us know.”

No name. No cause. No “signed off.” No warranty argument. It admits nothing and it stonewalls nothing, which is the narrow target a legal-sensitive public reply has to hit. Then the ownership hands back: Dana still owns the fix. She makes the call that afternoon, the claim goes to the franchise’s insurer, Marcus gets a conversation in private rather than a public defense he never asked for. Six weeks later, after the insurer settles the water-damage claim, the customer quietly edits the review up to three stars and adds a line that the company “handled it.”

The difference between the two replies is not skill. Dana can write. The difference is that one of them was written by the person with the facts and the relationship, and it needed to be written by the person with the exposure map. That is what a governance model is for: matching the reply to the layer that has what the reply needs, and in this one lane, that layer is not the store.

Everything outside that lane runs the other way. When Dana’s next review is a genuine complaint about a plumber who showed up in the wrong two-hour window, she answers it herself, that afternoon, names the dispatcher who is fixing the scheduling, and never asks anyone’s permission. Corporate reads it in next month’s audit sample and moves on. The whole system exists to make that second case fast and the first case safe, and to never confuse the two.

The part nobody tells you: the franchise agreement is where this lives or dies

You can design the cleanest governance model on the whiteboard and watch it evaporate the first time a franchisee decides they don’t feel like following it. The thing that makes it real is not the diagram. It’s whether the franchise agreement, and the brand standards manual it incorporates by reference, actually says any of this.

Most agreements don’t. They have a boilerplate clause requiring the franchisee to “comply with brand standards” and a social media paragraph written in 2014 about not posting offensive content. Nothing about review response, nothing naming an escalation category, nothing granting corporate the authority to take over a reply during a legal event. That vacuum breaks the model in both directions.

In one direction, corporate has no contractual right to intervene, so when a location is publicly arguing with a customer who is alleging an injury, the brand can ask nicely and nothing more. By the time the franchisee stops, the thread is already screenshotted. In the other direction, corporate over-reaches, claims the logins to every profile, and responds to everything from the center to be safe. The franchisee, now locked out of their own reviews, stops paying attention, stops feeding corporate the facts, and the whole network drifts to the Dayton voice. A right to take over becomes a habit of taking over.

The fix is to write the split into the agreement, or the standards manual it points to, in language that does three specific things. It names the escalation categories and defines them tightly enough that a manager can recognize one. It grants corporate a defined, narrow right to respond in that lane, and only that lane. And it affirms the franchisee’s right and duty to own everything else, on their own voice, on their own clock. The point of writing down what corporate can take over is equally to write down what it can’t.

There’s a reason to protect local ownership beyond principle. The International Franchise Association’s State of Franchise Marketing report found that 61% of franchisees with full control over their local messaging said they were very satisfied, against just 20% of those with limited or no control. Autonomy over the local voice is not a courtesy you extend to keep franchisees happy. It is what keeps the person who has the facts engaged enough to supply them. The moment a franchisee feels like a spectator on their own Google profile, you’ve lost your only source of the specifics that make any reply worth reading. Consistency you enforced by taking the keyboard away is consistency built on a store that has stopped caring.

Brand standards do need teeth, and the agreement is where the teeth go: defined consequences for the location that repeatedly ignores the floor or lets its reviews rot unanswered. But the bigger lever is almost never enforcement. It’s making the compliant path the easy one. A franchisee who has a good template library, a clear one-line rule for what to escalate, and the freedom to just answer the ordinary reviews will follow the model, because the model is less work than going rogue. Most rogue-location problems are really unsupported-location problems wearing a costume.

Where this connects to the rest of the machine

Governance is the layer above two others this cluster covers in depth, and the split above only works if those are in place under it. The escalation map you hand each location is only as good as the process behind it, the staffing, the rotation, the tiers, which we treat separately in building a response process that doesn’t burn out your team. And the question of how much of the ordinary, positive-review volume you draft with software rather than by hand is its own decision with its own threshold, worked through in when to automate review responses and when not to. Governance decides who owns a reply. Those two decide how the owner gets it done without drowning.

The tooling question comes last, not first, which is the opposite of how it usually gets bought. Once you know your escalation categories and your per-location ownership, you can evaluate a multi-location platform against them: does it route red-flag reviews to a different queue, can it hold a category for corporate while letting locations self-publish the rest, does it show you response rate by location so the audit is real work and not a vibe. Most of them optimize for bulk publishing, which is the one thing you specifically do not want to make frictionless. If you’re weighing the category, our look at whether GatherUp is worth it walks through the same fit-first way of judging a tool against your actual workflow rather than its feature list.

Some verticals live almost entirely in the legal-sensitive lane, and multi-unit operators in those spaces should default the split further toward corporate. Property management is the sharp example: fair-housing exposure means a huge share of tenant reviews are one careless sentence away from a complaint, and we walk through that specific balance in the property management reputation playbook. The framework is the same. The dial just sits in a different place when every reply is closer to the red-flag lane than the ordinary one.

If you’re writing the brand standards from scratch and want a starting point for the ordinary-review floor rather than a blank page, a tone-matched template library is a reasonable place to lift raw material, as long as you hand it to locations as clay and not as law.

The line to hold

The brand you’re protecting was never the logo at the bottom of the reply. It’s whether the person reading it believes someone who actually works at that store, in that town, wrote the words. Almost every reply should pass that test, and governance is just the discipline of making sure it can, without a gate slowing it down. The handful of replies that can’t pass it are the ones you take out of the store’s hands on purpose, and knowing which handful those are is the entire job.