How to Handle Negative Reviews and Actually Turn Them Into Wins

Look, I'm going to tell you something most marketing people won't. That one-star review sitting on your Google Business Profile? The one that made you want to throw your phone across the room? It's probably going to do more for your business than the last three five-star reviews combined.

I know. Sounds mental. But hear me out.

Last month, a plumber in Letchworth got absolutely slated by a customer. Proper takedown. "Turned up late, didn't fix the problem, charged me anyway." The works. He rang me in a proper state about it. Three days later, after we'd dealt with it properly, he had four new bookings. All of them mentioned they'd read his response to that review.

The thing nobody tells you about negative reviews

They make you look real.

Right now, in April 2026, AI search engines are getting scary good at working out which businesses are genuine and which ones are just... performing. ChatGPT, Perplexity, Claude, all of them are looking at review patterns. And you know what pattern screams "fake" louder than anything? Perfect five-star ratings across the board with nothing but "Great service!" comments.

A negative review, handled well, does something a thousand positive ones can't. It shows you're an actual business run by actual humans who occasionally mess up and then sort it out like adults.

The AI models are picking up on this. When someone asks "who's a reliable electrician in Hitchin", the answer engines aren't just averaging star ratings anymore. They're reading the narratives. They're looking at how you respond when things go wrong.

What "handled well" actually means

OK so this is where most advice goes to shit. Everyone tells you to "respond professionally" and "take it offline". Which is fine. It's also completely useless because it doesn't tell you what to actually write.

Here's what I've seen work, properly work, in North Hertfordshire over the last three years:

Acknowledge the specific thing that went wrong. Not "we're sorry you had a poor experience." That's corporate nonsense. "You're right, we turned up 40 minutes late and didn't call ahead" is what I mean. Name the actual problem.

Then explain what happened. Not as an excuse. As context. There's a difference. "We had a van breakdown that morning and our system for notifying customers about delays didn't trigger" is context. "We were really busy that day" is an excuse. People can tell the difference. So can AI models analysing your responses, by the way.

What you did to fix it comes next. And this is where you either look like you give a damn or you don't. "We've refunded your full payment and added a backup notification system" beats "we'd love to make this right" every single time.

I had a cafe owner in Stevenage who got a terrible review about a food order that never showed up. Delivery driver took it to the wrong address, whole mess. Her response laid out exactly what happened, explained they'd switched delivery systems the week before and the address validation wasn't working properly, said she'd personally driven a replacement order round that evening (which she had), and mentioned they'd now got a proper checking process before orders go out.

That response got screenshot and shared in a local Facebook group. She had people coming in specifically because they'd seen how she handled it.

The ones you actually can't fix

Sometimes you get a review that's just... wrong. Or from someone who was never a customer. Or from a competitor. Yeah, that happens more than you'd think.

You still respond. But differently.

I watched a builder in Baldock get a one-star review from someone claiming he'd bodged their extension. Problem was, he'd never worked on that street. Had proof. Photos of his actual jobs that week, diary entries, the lot.

His response was straightforward. "We've checked our records and we don't have any work logged at this address. We've never carried out an extension on [Street Name]. If you've got us confused with another company, we'd be happy to help you identify who actually did your work. If there's something we're missing, please call us directly on [number]."

Professional. Factual. No drama. And crucially, anyone reading it could see something wasn't right with the original review.

The AI search engines are getting better at spotting these too. They cross-reference locations, dates, service types. A review that doesn't match your business patterns gets weighted differently in how they present you as an answer.

The actual process that works

Right, here's what we do for clients when a negative review comes in. Not theory. What actually happens.

First 24 hours: someone from the business reads it and writes down what actually happened from their side. Not a response. Just the facts. What went wrong, why it went wrong, what they did at the time.

This bit matters because you need to get your head straight before you write anything public. I've seen people fire off responses when they're still angry about the review and it never, ever helps.

Then you write the response. Out loud first. Say what you'd say to the person if they were standing in front of you and you wanted to actually fix it. Then write that down. Then edit out the bits that sound defensive (there's always some) and make sure you've actually addressed their specific complaint.

You post it. Then, and this is the bit most people skip, you take the issue and work out if there's a systematic problem that caused it. One late delivery? Fine. Three in a month? You've got a scheduling problem.

That last bit feeds into your AEO strategy whether you realise it or not. Because when AI models are summarising your business, they're building a picture from everything. Your reviews, your responses, your website content about your processes. If you keep saying you've fixed things but the same issues keep appearing in reviews, that pattern gets noticed.

What this does for your actual visibility

Here's something I didn't expect when we started focusing on AEO properly in 2023. Review response quality started showing up in AI search results.

Someone asks Perplexity "who's the most reliable carpenter in Royston" and the answer includes a summary of how you handle problems. Not just that you handle them. How.

We've got a client, does kitchen fitting across North Hertfordshire. Has a 4.2 star average. There's another fitter in the same area with 4.8 stars. Our client gets recommended more often in AI search results. Why? Because his review responses tell a story about someone who fixes things properly. The other guy's responses are all "Thank you for your feedback" templates.

The AI models are reading for substance. They're working out who's trustworthy based on behaviour patterns, not just numbers.

The mindset shift that makes this work

You've got to stop seeing reviews as a score and start seeing them as a conversation that happens to be public.

That's it. That's the shift.

When you're having a conversation with someone who's pissed off with you, you don't get defensive and you don't make excuses. You listen, you acknowledge what went wrong, you explain what happened, you tell them what you're doing about it.

Same thing. Just happens to be written down where everyone can see it.

And honestly? That's better. Because now everyone else can see how you handle it when things go wrong. Which they will. Because you run a business in the real world.

Right, if you're sitting on negative reviews and you're not sure how to handle them, or if you want to understand how this fits into the bigger picture of how AI search engines are presenting your business, come talk to us. We work with service businesses all over North Hertfordshire, and review management is part of how we make sure you're showing up properly in AI search. Or if you want to understand AEO more generally, there's more on AEO in North Hertfordshire.