AI readiness vs AI strategy: why most SMBs get this the wrong way round

Most small and medium businesses we speak to want to start with an AI strategy. They've read the headlines, sat through the webinars, and feel a quiet pressure to 'have a position' on AI. The board has asked. The team has asked. Their biggest customer has just mentioned it in a quarterly review. So they sit down to write a strategy.

It's the wrong place to start.

Strategy answers the question 'where should we go?'. Readiness answers the much more useful question 'what's actually going to happen when we try to go there?'. Skip readiness and you'll learn the answer to the second question the hard way - usually about three months and £40,000 into a pilot that quietly stalls.

What readiness actually is

Readiness is the unglamorous work that makes any AI investment pay off. It isn't a 60-page maturity model. It's six or seven things that, in our experience, decide whether AI sticks in an SMB or not:

• A short, plain-English acceptable-use policy that the team has actually read.
• Someone senior who can describe, without jargon, what AI is and what it is not.
• The two or three core data sources tidied up enough that the AI has something useful to chew on.
• A small group of curious people who are allowed to experiment without sign-off.
• A budget line that's small enough not to scare the board and big enough to learn something.
• A clear understanding of which client or customer data must never be pasted into a public tool.
• One named person who owns 'AI' as a portfolio, not a job title.

None of that needs a consultant. Most of it can be done in a fortnight by people you already employ.

Strategy without readiness fails predictably

We see the same pattern again and again. A senior leader champions a flashy pilot. Tools get bought before workflows are mapped. The team is asked to 'try it' on top of their day jobs. Adoption peaks at week three and quietly collapses by week eight. By month four, somebody on the leadership team says 'we tried AI, it didn't really work for us', and the next twelve months get harder, not easier.

The frustrating part is that the pilot might have been the right idea. The strategy might have been sound. But the conditions to make it work weren't there - and nobody had spent fifteen minutes asking whether they were.

Readiness without strategy is also a trap

There's an opposite failure mode: businesses that get hooked on the audit. They do a readiness assessment, then another, then a third with a different framework. They produce beautiful slides about data maturity, change capacity, and AI literacy. Six months in, they still haven't done anything that affects a customer.

Readiness without strategy is busywork. The point of getting ready is to do something useful, sooner. If your readiness work isn't pointing at a real, named opportunity within thirty days, it has drifted into theatre.

What to do instead

Spend two weeks on readiness before you spend two months on strategy. Audit your data, your skills, your governance, and your appetite for risk. Get a clear picture. Then - and only then - decide where AI should go to work first.

When you do pick a starting point, keep it small. One workflow. One owner. One success metric. One thirty-day review. The point of the first project is not to transform the business. The point is to learn what your team is actually capable of when given the right conditions and a sharp objective.

Get that loop right once, and the second project is much easier. Get it right twice, and you've built something most of your competitors don't have: a business that knows how to absorb new technology without falling over. That's the real competitive advantage AI offers SMBs - and it starts with readiness, not with strategy.