Case 09 · Small business · Illustrative example
AI that earns its keep
Half of small firms use AI, and 12% can point to any revenue from it. We automate one repetitive job, measure the result and keep it only if it works.
What happened
It happened - March 2026
Half of UK small firms now use AI - but only 12% can point to any revenue from it, and 46% say they lack the knowledge to use it properly. Nearly everyone has tried a chatbot; almost nobody has wired AI into a process where it measurably pays.
Your version of it
Someone senior comes back from a conference asking for an AI strategy. Meanwhile the office manager spends every Monday morning retyping supplier invoices into the accounts. What you need is that Monday morning back, and a clear answer on whether the automation is worth paying for.
This is you if
- Someone does the same rule-based task every week
- You've tried AI tools, but none stuck in a real process
- You want to know whether it pays, not a strategy document
How we run it
We find the repetitive work
One session with the people who do the work, listing the rule-based tasks and how long each takes every week.
We automate one task
For example: an invoice arrives, its details are read, and a draft appears in the accounts. A person approves it before anything is paid. Anything involving money keeps a human check.
We run it alongside the person for two weeks
The automation works in parallel and we compare its output line by line. Errors get fixed, and nobody relies on it yet.
We measure it, then keep it or switch it off
Hours before and after, and the error rate. If it doesn't save time, we tell you and switch it off.
Made boring by
A fixed-price build, with the measurement written into the quote. No per-user AI licences to buy from us.
Sources
Real figures: BCC, Mar 2026. The firm above is illustrative.