AI Strategy for Small Business: Start With the Problem, Not the Tool

Most small businesses don't need more AI. They need fewer hours lost to admin, fewer jobs falling through the cracks, and less information copied by hand from one system into another. AI can help with all of that, but only once you know exactly which problem you're solving.

That's the principle behind how I work as an AI consultant and strategist for SMEs. I start with your business, not the technology: how the work actually flows, where the time goes, and what's quietly holding your team back. Only then do we decide what to build. Sometimes the answer is a tool you already pay for, set up properly. Sometimes it's an automation that joins your systems together. And sometimes it's a completely custom solution, designed around the way you work.

In short: an AI strategy for a small business is a prioritised plan that ties AI to specific problems, not to a list of tools. Build it in four steps: discover the real pain points, strategise by deciding which are worth solving, build the simplest fix that works, then drive adoption so your team actually uses it.

In this article I'll explain that approach, what the latest UK research says about where small businesses are getting stuck, and how to tell whether you need an off-the-shelf tool or something bespoke.

Why are so many small businesses stuck with AI?

The most common barrier isn't the technology. It's knowing where to point it. In the Department for Science, Innovation and Technology's AI Adoption Research (published February 2026; 3,500 UK businesses with five or more staff, surveyed in 2025), 71% of businesses not using AI said they hadn't identified a use for it, the most common reason given, and 60% said they had limited AI skills or knowledge. Only 23% said the cost was too high.

Adoption is rising fast, but it's shallow. According to Office for National Statistics figures published in July 2026, around 35% of UK businesses with 10 or more employees were using at least one AI technology in June 2026, up from around 12% in late 2023. For businesses with fewer than 10 employees, the figure was 28%. And among the businesses with 10 or more staff that do use AI, only 10% say they use it extensively.

In other words, most businesses aren't short of AI tools. They're short of a clear answer to a simple question: where would this actually help us? That's a strategy problem, and it's where I always start.

What does an AI consultant actually do?

An AI consultant works out where AI will genuinely improve your business, then helps you put it in place safely and get your team using it. A good one is tool-agnostic: the recommendation should come from your problems, not from whichever product they happen to sell.

In practice, my work with an SME covers four stages: understanding the business, agreeing a prioritised plan, delivering the solutions, and making sure your team adopts them. Here's what each one involves.

How to build an AI strategy for a small business

1. Discover: find the real pain points

Every engagement starts with listening. I sit down with the owner and the people doing the work, and we map how things really happen day to day: how enquiries arrive, how jobs are quoted and scheduled, where information gets typed in twice, which reports eat an afternoon, and what your customers end up waiting for.

The questions are simple. What takes too long? What goes wrong most often? What would you do with an extra day a week? The answers are rarely what people expect, and they're almost never "we need a chatbot".

2. Strategise: decide what's worth doing

Not every pain point needs AI. Some are better solved with a simpler process or a tool you already have. I weigh each opportunity on impact, effort, cost, risk and how ready your data is, then turn the shortlist into a plain-English roadmap: quick wins that can be running within weeks, and bigger pieces that deserve proper investment.

This is also where data protection gets settled. For every tool, we agree what information goes into it, who can access it and how it fits your UK GDPR obligations, before anything is switched on.

3. Build: from off-the-shelf to fully custom

Then we deliver, choosing the right approach for each problem. That might mean configuring tools you already pay for, connecting your existing systems so information moves between them automatically, or designing and building something bespoke. There's more on how to choose below.

4. Adopt: make it stick

A tool nobody uses is just another subscription. I train your team on the new tools in the context of their actual jobs, write clear guidelines for using them, and agree simple measures so we can see what's working and adjust what isn't. Training people is familiar ground for me: as a Production Director, I created training and e-learning content for global brands.

Off-the-shelf or custom AI: which does your business need?

Start with what already exists, and build custom only when you need to. Most SMEs get their first real gains from off-the-shelf tools that are set up properly, then from automations that connect the systems they already use. A custom build earns its place when the problem is specific to how your business works and nothing on the market solves it well.

DSIT's AI Adoption Research points the same way. Of businesses using or planning to use AI, 65% planned to invest in off-the-shelf AI applications over the following 12 months, 59% in embedding AI into their current applications and processes, and 22% in building bespoke AI solutions.

I think about it as three levels:

• Configure: tools like Microsoft Copilot, ChatGPT and Google Gemini, along with the AI features already built into software you use, set up with the right settings, permissions and prompts for your team. It's the fastest route to results and usually the cheapest.
• Connect: automations that move information between your existing systems without anyone copying and pasting, such as enquiries routed into your CRM, documents summarised and filed, or reports assembled overnight. Even the enquiry forms on this website run through an automated workflow.
• Build: custom software for the problems that are unique to your business. I design and build bespoke platforms end to end, like the H Prestige Estates platform, which syncs every listing straight from the agency's CRM with no manual uploads, and Screenli, a digital signage platform that pulls property listings from estate agency CRMs automatically.

My rule of thumb: if a well-configured off-the-shelf tool gets you most of the way there, use it. If the part it can't do is where your competitive advantage lives, that's when custom is worth the investment.

Where can AI actually help a small business?

Usually in the unglamorous places: the repetitive, time-consuming work that sits between your team and the work they're actually good at. These are the areas I look at first:

• Admin and inboxes: drafting replies, summarising long email threads and sorting enquiries by urgency.
• Double data entry: forms, CRMs, spreadsheets and accounts packages that don't talk to each other.
• Response times: answering common customer questions and capturing enquiries out of hours, with a clean hand-off to a real person.
• Reporting: pulling numbers from several systems into one view, without someone losing an afternoon to it.
• Knowledge in people's heads: turning documents, price lists and processes into something the whole team can search and ask questions of.
• Content and marketing: first drafts, repurposing and visual production.

AI is part of my own production work too. For Bramble Rise in Cobham, we used AI editing to visualise a full renovation, a rear extension and a landscaped garden inside the property film, and for Pinehurst in Sunninghill, AI editing "teleported" the agent from the house out into the village around it.

Notice what isn't on that list: replacing people. In research published by the British Chambers of Commerce in March 2026, 95% of SMEs using AI said it had made no difference to the size of their workforce over the past year, and most were using it to support employees rather than replace them. That's the right way to think about it. The goal is to give good people their time back.

Is it safe to use AI with your business data?

Yes, as long as you set the rules before you start rather than after something goes wrong. The risks for an SME are real but manageable: confidential information pasted into the wrong tool, answers that sound confident but are wrong, and uncertainty over who owns what the tools produce.

That's why every AI strategy I put together covers which tools are approved for which kinds of information, how access is controlled, how your obligations under UK GDPR are met, and where a person must check the output before it goes anywhere important. None of it is complicated, but it needs to be written down, agreed and shared with the whole team.

How long does it take to see results?

A well-chosen quick win, such as setting up an existing tool properly or automating one painful process, can be up and running within a few weeks. Larger custom builds take longer, and they should be phased so that each stage pays its way before the next one starts.

It's worth being realistic, too. In DSIT's AI Adoption Research, three quarters of businesses already using AI reported improved workforce productivity, yet the ONS figures show that only 10% of AI-using businesses with 10 or more staff use it extensively. The way to get real results is to point AI at a clearly defined problem and measure what changes. That's exactly why discovery comes first.

About the author: Ryan Cook

I'm Ryan Cook, founder of One Stop Creative Agency in Ascot, Berkshire. I've spent more than 17 years in web design, development and video production, and I've been building websites since I was 13.

My background is as operational as it is creative. I've worked close to the top of nationwide marketing firms, led teams of more than 300 people, and served as Production Director creating training and e-learning content for brands including Jacobs Douwe Egberts, Britvic, General Mills, Henkel and Moët Hennessy. I also designed and led the rebuild of a nationwide field marketing application, from the first concept through to deployment.

Today I design and build custom platforms, automations and web applications, alongside the agency's video and creative work. That combination matters for AI consultancy. I can look at a business problem from the operational side, recommend the simplest thing that will solve it, and then build it when it needs building.

If you run an SME and you're not sure where AI fits, that's exactly the right place to start. Take a look at my AI consultancy service for SMEs or get in touch, and we'll begin with a conversation about your business, not a sales pitch for software.

Ryan Cook
AI Consultant & Strategist, One Stop Creative Agency
Ryan Cook on LinkedIn