Generative AI loves big brands. Here’s how to compete when you’re small
Generative AI keeps recommending big brands because they've been everywhere for ten years. But they fall short on hyper-specific queries. As a small brand, your edge is specialisation and speed. Answer your customers' real questions better than anyone else.
- Author
- Axel Corton
- Subject
- eCommerce
- Date
- Read time
- 4 min

Key takeaways
- AI cites the most visible brands: press, forums, Reddit, comparisons built over ten years of history.
- You can't win broad queries; target 50 hyper-specific ones where you're the best answer that exists.
- A supplement client doing 40K a month ranks on AI by answering its niche formula best, not by being famous.
- Start from real customer questions: support tickets, reviews, Instagram DMs, service emails.
- A small brand rewrites its product pages in a week; a giant needs six sign-offs and three months.
A cosmetics brand founder messaged me last week. Basically he says, this is all cool, but when I ask ChatGPT to recommend a moisturiser, it spits out L'Oréal, Clarins, the names everyone already knows. Never mine. So what do I do as a small brand?
It's a totally fair question. And yes, the big brands have an advantage. A big one. Sometimes not even earned.
But that doesn't mean you're done for. I'm going to explain why they win, and more importantly what we actually do on the agency side to get visibility when you're starting from scratch.
Why AI always cites the same names
When you ask ChatGPT, Perplexity, or Google's AI search a question, the model answers with what it's seen the most. And what it's seen the most is the brands everyone's been talking about for ten years. Magazines, forums, comparison articles, Reddit, press pieces. The more a brand gets cited everywhere, the more the AI treats it as a safe answer.
Same thing as a Sephora salesperson you ask for a moisturiser. They'll name the three brands people ask for all day long. Yours they don't know, so they don't mention it.
So big brands have a double advantage. They've got the budget to be everywhere, and they've got the history. Ten years of articles, reviews, mentions. You've got six months of existence and three Trustpilot reviews.
Except it's not a lost cause. Because AI has one huge weakness, and that's exactly where we play.
The big brands' flaw is generality
A big brand gets cited on broad queries. "Best moisturiser." "Best running brand." There you've got zero chance, forget it.
But nobody actually asks the question that way in real life. People ask hyper-specific stuff. "Fragrance-free moisturiser for reactive atopic skin." "Trail shoe for wide feet on muddy terrain." "Magnesium bisglycinate supplement with no additives for athletes."
And there, L'Oréal loses a lot of ground. A brand that targets everyone ends up answering off the mark on a hyper-specific request. You, if you're ultra-specialised, you answer the question better.
I've got a client in supplements, around 40K a month, who started showing up on ultra-targeted AI queries around his specific formula even though he's nothing compared to the giants in his space. And he showed up just because he's the only one who answers that question well, not because he's famous.
So drop the "how do I beat L'Oréal." Ask yourself instead which 50 ultra-specific queries you can be the best answer in existence for.
What we actually do when a brand is starting out
Concretely, here's how we move.
The first step is listing all the real questions customers ask. Not generic keywords, the actual phrases. We go find them where they already live, in customer service, in reviews, in Instagram DMs, in support emails. People tell you exactly what's blocking them before they buy. That's gold.
Once that's laid out, we write the content that answers it better than anyone. Not 2,000-word articles that beat around the bush, but a clear, data-backed answer, with the detail the big brand will never give because it has too many products to manage. If you sell a supplement, you put the exact dosage, the chemical form, who it's for, who it's not for. On that point, how you reassure buyers with scientific proof on a product page changes everything, because AI loves precise, sourced answers.
Then we go get mentions off your own site. A guest article, a comparison on an industry blog, an interview in a niche newsletter, detailed customer reviews. AI doesn't just trust you. It cross-references what's said about you all over the place. The more you're cited across your corner of the market, the more you become a credible answer.
And the site itself has to be technically clean. Complete product pages, an FAQ on every page, properly marked-up data. The AI has to be able to read your site without a struggle. Something dumb but huge, a lot of sites mostly serve noise to the bots. If you've already noticed that a big chunk of your sessions came from bots, you know crawling isn't a detail.
The advantage you have and they don't
One thing people forget. A small brand moves fast. You can rewrite all your product pages in a week. You can ship ten niche pieces of content this month. At a big brand, the same change goes through six approvals and three months of meetings.
So yeah, they have the resources. But you move fast and you're specialised, and that matters a lot more than people think.
The mistake I see all the time is the small brand trying to play the big brand's game. It tries to rank on broad queries, talk to everyone, look like the leader. That's the fastest way to vanish. You'll never be the low-cost version of L'Oréal in an AI's mind.
Stay on your turf. Become the reference in your niche instead of chasing a big brand. The big brand wants everyone, you want the 2,000 people your product is exactly the right answer for. And those are the people asking AI the precise questions.
Where to start this week
If you want to move fast, here's what I'd do in your shoes.
Take your last 20 customer service conversations. Note the real questions that keep coming up. For each one, ask yourself whether your site answers it clearly in one page. If not, write the page. Just that keeps you busy for the next few weeks.
And stop watching what your market's leader is doing. They're playing a game you can't afford to play. Your thing is precision. Start with those 20 conversations, the rest follows.
Frequently asked questions
Does generative AI really only recommend big brands?
On broad queries like “best moisturiser”, yes, almost every time. They cite what they've seen the most. But on precise, niche questions, a well-positioned small brand can come out ahead, because it answers the exact question better.
How long before my brand gets cited by an AI?
It's not instant, models take time to absorb what's said about you. Count on several months of precise content and external mentions before you see results. The brands I see succeed have worked the niche for 6 to 12 months.
Should I write for the AI or for my customers?
For your customers. A clear, data-backed, honest answer to a real customer question is exactly what AI loves to cite. If you write to please the bots, it sounds fake and it doesn't convert anyway.
Do mentions on other sites really count?
Yes, massively. AI cross-references what's said about you all over, not just on your own site. A niche comparison, an interview in a newsletter, detailed customer reviews, that makes you credible in the model's eyes.
What do I do first with a small budget?
List your customers' real questions, find them in your customer service and reviews, and write the pages that answer them better than anyone. Zero media budget, just time and precision.
About the author

Axel CortonFounderLinkedIn
Keep reading
Subscribe to our newsletter
Our methods on paid media, eCommerce and CRM, in your inbox.


