Studio openStart a project →

#Brand strategy

Why brand systems need to scale for the people, the press, and the platforms

A brand used to need one audience and one channel. Now it needs three: customers who recognise it, press who can quote it, and AI platforms that can cite it. Here's what each one actually wants.

Why brand systems need to scale for the people, the press, and the platforms

For most of my career, brand strategy was a one-audience problem.

You picked a customer. You built a system that spoke to them. You ran it through paid and earned media. If the system worked, the customer recognised you in market and you sold things. If it didn't, you tweaked the look or the tagline and tried again.

That model has been quietly collapsing for fifteen years. Social broke the channel side of it. AI is now breaking the audience side of it.

The brands I see winning right now aren't winning because they have prettier identities. They're winning because their brand system is doing work for three audiences at once. People. Press. Platforms. Each one wants something different. Most studios are still designing for the first one and hoping the other two figure it out.

They don't.

What changed under the hood

The old map was simple. One audience, two channels. Customers, reached through paid media you bought and earned media you pitched.

Now there are at least three distinct audiences making decisions about your brand, and the channels they use barely overlap. A customer scrolling Instagram sees one slice of you. A journalist researching a feature pulls a different slice from your About page and a couple of quotes you gave on a podcast. An AI model answering "who's the best in this category" reads a third slice entirely, drawn from your structured data, your Wikipedia presence, the way Reddit talks about you, and what your llms.txt file says.

If those three slices contradict each other, you have a brand integrity problem before you have a marketing problem. If two of the three are empty, you're invisible in places you don't even know to look.

A modern brand system has to work simultaneously across all three. Not in a sequence. Not "customers first and we'll handle press and AI later." All three, from day one.

For people

This is the audience studios already know how to design for, so I'll be brief.

People want emotional clarity and recognisability. They want to know what you are within about a second of landing on you. They want a colour, a shape, a tone of voice they can pattern-match to a feeling. They want the thing on Instagram to feel like the thing on the website to feel like the email signature.

What they need from a brand system, specifically:

A visual identity that's recognisable at thumbnail scale. If your logo only works at full size, your brand only works in its own home.

A tone of voice that sounds like a person, not a press release. Customers can smell committee copy in two sentences.

A clear "what we do" statement that any twelve-year-old could repeat back. If your homepage hero needs three sentences to explain what you sell, the brand isn't doing its job.

Consistency across the messy in-between surfaces. Receipts. Onboarding emails. The error page. Customers form opinions in those places more than they do on the homepage.

This part of the work hasn't changed. Studios who do it well will keep doing it well. The problem is that this is now the floor, not the ceiling.

For the press

Journalists have always been a real audience, but most brand systems treat them like a footnote. Logos, colours, a press kit page nobody updates. That's not enough anymore, because press now isn't only press. It's editorial publishers, podcasters, newsletter writers, Substack authors, and the long tail of trade publications that AI platforms cite heavily.

What the press needs from a brand system, specifically:

Quotable substance. Not "we're a leading provider of solutions." A real position with a real argument behind it. A journalist isn't going to invent a quote for you. If your About page reads like beige, you don't get quoted.

A founder or studio voice that can speak in public. The brand has to extend to how the people behind it talk on podcasts, in panels, in long-form interviews. Most brand books stop at the logo lockup. They should keep going.

Verifiable specifics. Press won't print numbers they can't check. Named clients, named outcomes, dates, places, awards. No vague lines about caring deeply about craft.

A press page that actually works. Hi-res logos in usable formats, founder bios with hi-res photos, a one-paragraph studio summary written in third person, and a contact route that gets answered within a working day.

I've watched brands lose feature placements because the journalist couldn't find a usable logo or a quotable line in under ten minutes. That's a brand system failure, not a PR failure.

For the platforms

This is the audience most brand strategy work hasn't caught up to yet, and it's the one that's growing fastest.

By "platforms" I mean AI models, search engines, and the recommendation systems they feed. ChatGPT, Claude, Gemini, Perplexity. Google's AI Overviews. Whatever's coming next that I haven't seen yet. These systems are increasingly the first surface a buyer touches when they research a category. If you're not in the answer, you're not in the consideration set.

What platforms need from a brand system, specifically:

Structured, machine-readable claims. Schema.org markup on every page. Organisation schema, Service schema, FAQ schema. A clean llms.txt at the root of the site. AI doesn't guess. It extracts.

Consistent claims across surfaces. If your website says you're a Belfast brand studio and your LinkedIn says you're a UK design consultancy and your podcast bio says you're a freelance designer, the model has three signals to choose from and no clear winner. It picks one or hedges.

Citable presence on the sources AI models trust. Wikipedia. Reddit. Quora. YouTube. Major editorial publishers. The trusted-twenty domains across the main AI platforms account for roughly half of all citations. That's where the leverage lives.

A plain-language description of what you do, in two paragraphs, somewhere obvious on the site. AI summarises. Make the summary easy.

Unique entity signals. A consistent name spelling. A canonical website. A clear founding date and location. The boring metadata that tells a model "this is one specific entity, not a fragment of another one."

Most brand books I see have nothing to say about any of this. The platform layer of the brand sits in a developer's task list, treated as an SEO afterthought. That's a positioning mistake. The structured claims are the brand at this layer. They are the version of you the model sees.

The fragile brand and the durable one

A brand system that wins for one audience but fails for the other two is fragile. I'd argue it's already losing, even if the customer-facing numbers look fine.

Beautiful identity, no quotable substance, no structured claims. Customers love it. Press won't write about it. AI won't cite it. The brand looks healthy until a competitor with a less polished identity but tighter substance and cleaner data eats their lunch in the answer engines.

Strong press positioning, weak visual system, no platform layer. Gets featured. Looks impressive in pitch decks. Doesn't convert customers and slowly disappears from AI recommendations.

Tight structured data, no emotional pull, no human voice. Cited by ChatGPT. Ignored by humans.

The durable brand is the one that does all three. Recognisable to customers. Quotable to press. Citable to platforms. Each layer reinforces the others. The press writes about you because customers love you. The AI cites you because the press wrote about you. The customers find you because the AI cited you. It compounds.

That loop is what "designed to be seen" actually means. Not seen by one audience. Seen by the three audiences that decide whether your brand exists in the rooms you care about. People. Press. Platforms.

A brand that's only seen by one of them is a brand on borrowed time.