I have been in tech for 27 years. I have seen the same pattern repeat with every new wave of technology. The companies that get the most value are never the ones running the big visible programmes. They are the ones quietly getting on with it.
AI is no different.
The AI Programme Trap
Here is what happens. A board reads about AI in the FT. The CEO tells the CTO to “do something with AI.” A task force is assembled. A strategy document is produced. A pilot is scoped. Six months later, the pilot finishes, a deck is presented, everyone nods, and nothing changes.
I have watched this play out at companies of every size. The bigger the company, the more elaborate the theatre. Steering committees. Innovation labs. AI centres of excellence. Beautiful PowerPoint slides about “AI maturity models.”
Meanwhile, the accounts team at a 40-person logistics firm in Swindon has been using ChatGPT to draft their month-end commentary for the last eight months. The ops manager at a manufacturing company in Leeds has automated their purchase order chasing with a simple AI workflow. A marketing director at a professional services firm runs their first-draft content through an AI tool and saves six hours a week.
None of those people are doing an “AI project.” They are just doing their jobs, slightly differently.
Why Programmes Fail
The problem with AI programmes is that they treat AI as a destination. You set a goal, you run a project, you deliver an outcome, you tick a box. That works for ERP rollouts and office moves. It does not work for AI.
AI is not a system you implement. It is a capability you develop. The tools change every few weeks. The models get better, the prices shift, new capabilities appear. If you treat AI as a fixed project with a start and end date, you are building something that will be outdated before the steering committee meets again.
I saw this at close range while leading enterprise technology teams. The organisations that got value from technology were the ones that built a habit of experimentation. Not a one-off programme, but an ongoing practice of “could we do this bit better?” The ones that struggled were the ones that turned every new technology into a two-year initiative with a project board and a risk register.
That approach is even worse for AI because the window between “interesting new capability” and “everyone is doing this” has collapsed to weeks, not years.
What the Winners Actually Do
The businesses I see getting real value from AI share a few characteristics, and none of them involve a strategy document.
They pick one thing. Not five things, not a roadmap, one specific task that takes too long and annoys someone. They fix it with whatever AI tool is appropriate. Then they pick the next thing.
They do not wait for permission from a committee. A team lead spots an opportunity, tries it, measures whether it worked, and keeps it or drops it. The cost of trying is low enough that you do not need a business case for every experiment.
They share what works. When someone in the finance team figures out a useful AI workflow, they show the person sitting next to them. No formal knowledge management system required. Just people talking to each other.
They are realistic. They do not expect AI to replace a person. They expect it to take the boring 20 percent off someone’s day so that person can do the interesting 80 percent. That is not a transformation. It is just a sensible improvement.
The Danger of the Strategy Mindset
There is a whole industry built around selling you an AI strategy. Consultants will charge you five figures to produce a document telling you where AI could add value in your business. Here is the thing: if you run a business and you do not already know which tasks eat your team’s time, a strategy document will not save you.
You know where the friction is. You know which reports take three days to produce. You know which customer queries follow the same pattern every week. You know which spreadsheet someone manually updates every Friday afternoon because the systems do not talk to each other.
You do not need a strategy to identify those. You need to pick one and fix it.
The strategy document gives you the comfortable feeling of progress without the risk of actually doing anything. It is a form of organisational procrastination dressed up as diligence. I have seen companies spend more on the AI strategy than they would have spent just trying five things and seeing what stuck.
When You Do Need Help
I am not saying you should never get outside help with AI. There is a point where DIY stops working and you need someone who has done this before. Integration with your existing systems, data security, compliance, anything that touches customer data, anything that needs to be reliable enough to run without human supervision. Those are the bits where experience matters.
But that help should look like a builder, not a strategist. Someone who rolls their sleeves up and helps you ship something specific, not someone who writes you a report about possibilities.
The best AI implementation UK businesses can get is one that starts with a real workflow and ends with a working tool, not one that starts with a workshop and ends with a slide deck.
The Uncomfortable Truth
Here is the contrarian bit. If your business has not started using AI in any practical way by now, it is not because you lack a strategy. It is because you have been spending your time talking about AI instead of using it.
The barrier is not knowledge. There are dozens of free tools, hundreds of tutorials, and more how-to content than any human could consume. The barrier is the belief that you need to understand AI before you can use it. You do not. You need to understand your business, and you need to be willing to try something small.
The person who uses AI badly for six months learns more than the person who studies AI perfectly for six months and never tries it. I have seen this play out repeatedly. The doers pull ahead while the planners are still planning.
AI is not going to take your job. But the person down the road who has been quietly using AI tools for a year, who has built up a feel for what works and what does not, who has shipped a dozen small improvements, that person is going to be genuinely difficult to compete with. Not because they have an AI strategy. Because they have a habit.
If you want to talk through what AI could actually do for your business, rather than what a strategy document says it could do, I am easy to reach.
Recommended Reading
- Practical AI for Business UK: Cutting Through the Hype - Which AI tools actually deliver value
- AI Strategy for SMEs UK: From Audit to Implementation - Building a practical AI strategy for your business
- AI Adoption UK: Overcoming the Top 5 Barriers - Understanding why UK SMEs struggle with AI adoption
We work across every UK sector. Browse our industry guides to see AI use cases for your business type.
Book a free discovery call with our team to discuss how to start with practical AI in your business.