What do you check to judge AI readiness?
Six things, roughly in order. They move from the plumbing to the part almost nobody has yet.
One, the technical back end of the site. A big chunk of getting found in AI search comes down to whether your site is structured so machines can read it. If the foundations are a mess, nothing else compounds, which is why we treat it as the base of all our AEO work.
Two, how enrichment is used. Not whether it exists, but whether it's shared. Is enrichment happening company-wide, or is everyone sitting on their own Apollo and Clay account, enriching in private?
Three, process automation between idea and output. Can someone go from "I want a one-page test asset" to the thing itself without a full design cycle? Most early tests don't need a design team. They need an agent or a skill you can hand the brief to.
Four, a shared AI tool, so the team works in one place and the tool learns from itself instead of everyone prompting in isolation.
Five and six sit together, and the sixth is the one I wish everyone had.
What is a marketing brain, and why does it matter so much?
A marketing brain is one shared store of context that every AI tool in the business reads from and writes back to. It's the readiness gap that separates a team playing with AI from one running on it.
Picture a repository of files: your company and its history, product information, brand guidelines, past awards and events, proposals you've sent, sales processes, what's working and what isn't. Anyone can point their own Claude or ChatGPT at it and pull the same current context. Connect it to your analytics and website and it ingests weekly performance snapshots, so whatever it suggests is based on your real numbers, not a guess.
The moment your tools share one brain, everyone stops building in private and starts building on top of each other.
The second half is what makes it compound. When you work against the brain, your work saves back into it, so the person who logs on an hour later is working from everything you just did. A shared Notion board is a step towards it. The reason most of these fail isn't the tool, it's that nothing keeps them current, because no AI is actually reading from or writing to them. Building that shared context is the core of our AI transformation work.
What's the most common readiness gap you find?
Shared enrichment on the sales side. Almost every time. Everyone has their own Apollo and Clay accounts, enriching their own lists and their own tier-one targets, none of it connected. So people target similar or identical accounts, can't see what's already been tried on a prospect, and make worse calls as a result.
The "we cover different regions so why connect them" logic falls apart the moment you remember companies operate UK-wide. Splitting the accounts splits the intelligence, and you lose the context a teammate already gathered on the exact person you're about to contact. Getting that shared and clean is really a CRM and automation job.
If you want an outside read on where your team sits against these six, book a free Growth Audit and we'll tell you honestly.
Frequently asked questions
Is heavy ChatGPT use a sign of readiness?
Not on its own. Individual tool use without shared context, clean data and a readable site is just busy. Readiness is about the foundations that let AI compound across the team, not how many prompts get typed.
Do I need a marketing brain to start?
No, but it is the highest-impact thing to build towards. Start with a shared store of context, even a well-kept Notion board, and make it the single source your tools read from. The value comes from it being current and shared, not fancy.
Why does shared enrichment matter more than tools?
Because siloed enrichment causes double-targeting, wasted credits and worse outreach, and it hides context a teammate already found. Shared enrichment turns individual research into team intelligence, which is where the real gains are.
How do I keep a shared context store up to date?
Connect it to the tools people already work in, so it is read from and written back to as part of the job, and where possible pipe in analytics snapshots automatically. Stores go stale when updating them is a separate chore nobody owns.
