Your Customers Are Asking ChatGPT Which Sofa to Buy - Is Your Brand in the Answer?

Somewhere right now, someone is planning a living room.

A few years ago, that meant a Saturday of showroom visits, or an evening lost in Pinterest tabs and twelve open browser windows comparing corner sofas. Today, a growing number of people are just asking ChatGPT.

"What's a good corner sofa for a small living room, durable enough for a dog?"

"Compare the best sustainable sofa brands in the UK."

"I want something mid-century but not too expensive - what should I look at?"

The AI doesn't send them ten blue links to sort through. It gives them an answer. Usually three or four brands, a short rationale for each, maybe a price range. And whichever brands make that shortlist have just done more persuading in one paragraph than most product pages manage in five scrolls.

The question worth sitting with: if someone asked that question today, would your brand be in the answer?

SEO has spent two decades optimising for a list. Get the keywords right, build the authority, earn the backlinks, and you climb toward position one of ten.

AI answers don't work like that. There's no page two to hide the brands that didn't make the cut. There's just the handful that got mentioned, and everyone else, who as far as that conversation is concerned, don't exist.

For a considered purchase like furniture - where people genuinely research before buying - that's a real shift in where the buying decision actually gets made. Someone who gets a confident AI recommendation for three sofa brands is arguably less likely to go and second-guess it with ten more searches. The AI has done the shortlisting. Ranking well on Google for "best corner sofas" doesn't help you if the AI never reads that page, or reads it and doesn't trust it enough to repeat it.

This is the part most brands haven't got their head around yet, understandably, because it's genuinely different to classic SEO.

Large language models aren't crawling your site in real time and judging your meta descriptions. They're drawing on a mix of what they were trained on and, increasingly, live retrieval from the web - and what they lean on most heavily is where a brand shows up outside its own website.

Reviews. Forums. Comparison articles. Reddit threads where someone asked exactly the question your customer is about to ask. Independent buying guides. The stuff that reads as genuinely useful rather than promotional is what gets pulled into an answer, because that's the stuff the model has learned people actually trust.

Which means the brands showing up in AI answers right now aren't necessarily the ones with the biggest ad budgets. They're the ones that have been talked about, compared, and recommended by people who aren't on the payroll.

A few things genuinely move the needle here, and none of them are a technical SEO fix:

Get into the comparison content that already exists. Sites doing "best sofa brands UK" or "corner sofa buying guide" roundups are exactly the kind of source AI models lean on. If your brand isn't in those pieces, that's worth chasing directly - outreach, product loans for review, whatever gets a fair, honest mention.

Take reviews seriously as a channel, not an afterthought. A steady flow of detailed, specific reviews - the kind that mention fabric wear after two years, or how it looks in a small flat - is exactly the texture AI models pull from. Generic five-star reviews with no detail don't carry the same weight.

Show up where the actual questions get asked. Home renovation subreddits, interior design Facebook groups, forums for new homeowners - these are where "which sofa" conversations happen in public, and where a genuinely helpful, non-salesy answer from someone connected to your brand can end up feeding directly into what an AI later repeats.

Make your own product information easy to lift accurately. Clear specs, honest material descriptions, real dimensions, straightforward care instructions. If an AI does pull from your site directly, it needs the information to be unambiguous enough to quote confidently.

None of this is something you can switch on with a campaign. It's closer to reputation than marketing in the traditional sense - and reputation takes time to build and is hard to fake.

That's frustrating if you're used to performance channels where you can see a result within a week of turning a budget on. This is slower. But it also means once a brand is genuinely earning that kind of trust, it's a lot harder for a competitor to simply outspend their way into the same position.

Start by finding out where you currently stand. Ask ChatGPT, Perplexity, and Google's AI Overviews the kind of questions your customers would ask - not your brand name, the actual buying question. See who gets mentioned. See if it's you.

If it's not, the fix isn't a new landing page. It's asking where the comparison content, reviews, and conversations in your category are already happening, and making sure your brand has a genuine, well-earned presence in them.

The living room example is a small one, but it's playing out across every considered purchase category right now - sofas, mattresses, kitchens, renovations, anything someone researches before they buy.

The brands that show up in those AI answers in twelve months' time are the ones building the reputation for it today, in reviews, forums, and comparison pieces most marketing teams have never thought to treat as a channel.

Worth checking sooner rather than later whether your brand is even part of that conversation - because if it isn't, waiting doesn't make the answer any more likely to include you.