Our AI consulting in the UK is for small businesses that want a straight answer from people who also build. Start with a free 30-minute call. If AI looks worth it, the AI Opportunity Review costs £149: an hour with you and the person who does the work, then a written shortlist of your top 3 use cases within 5 working days. Sometimes the answer is “not yet”, and we’ll say so.
Three steps, and you can stop after any of them. Most small businesses need an hour of honest advice and a short written review before they spend anything on a build.
Free
Tell us where your team’s time goes and what you’ve tried. We’ll tell you whether a review is worth paying for, or whether you need an off-the-shelf tool, a tidier spreadsheet or nothing at all.
A short, honest look at where AI could save your business time, and whether it’s worth doing yet.
£149
Quoted per project
Pilots and custom builds are quoted at a fixed scope once we’ve seen your data. Chatbots have fixed prices, from £300 + £29/mo. The review fee comes off any build you commission within 60 days. See what we build on our AI solutions page.
Swipe the table to compare
Consulting decides what to do, in what order, and whether your data can support it. Implementation builds it. Buying one when you needed the other is an expensive mistake, so be clear which you’re after before you ask for quotes.
| What’s included | You need consulting when | You need a build when |
|---|---|---|
| The problem | You don’t yet know which job to fix first | The job is agreed and the scope is clear |
| Your data | You’re not sure your data can support the idea | You’ve tested it, and the data works |
| Build or buy | You’re choosing between a subscription tool and a custom build | That’s decided, and you need it built well |
| Who decides | Several people disagree about priorities | One person can sign off on what gets built |
| History | A previous attempt failed and nobody agrees why | A previous attempt worked and you want more of it |
Already know what you need built? Go straight to our AI solutions or chatbot packages.
Plenty of AI projects never pay back. These are the figures we trust, each linked to the original research.
80%+
of AI projects fail by some estimates, twice the rate of IT projects without AI
RAND, 2024(opens in a new tab)60%
of AI projects without AI-ready data will be abandoned through 2026, Gartner predicted
Gartner, 2025(opens in a new tab)42%
of companies abandoned most of their AI initiatives before production, up from 17% a year earlier
S&P Global, 2025(opens in a new tab)20%
of organisations are already growing revenue through AI, while 74% hope to
Deloitte, 2026(opens in a new tab)28%
of UK businesses with 0–9 employees use at least one AI technology
ONS, 2026(opens in a new tab)Most of this research looks at larger companies. Two of its reasons apply to a ten-person firm just as much: data that isn’t ready, and not knowing where AI would help. The ONS found that difficulty identifying use cases was among the most common barriers for UK businesses, along with cost and a lack of expertise.
Whether your data can support an idea is usually clear in the first week, long before any money goes on a build. These are the six questions we ask in every review. You can answer most of them yourself with our free AI readiness check.
Plenty of ideas depend on information nobody records. To predict which enquiries turn into jobs, you need past enquiries linked to what happened next, and many businesses have never joined the two.
Data locked in a system with no export, held by a supplier who won’t release it, or spread across people’s inboxes isn’t usable yet. Sorting out access can take longer than the build.
The same customer entered four ways, service names that changed last year, free text where a dropdown should be. It’s the most common blocker, and fixing it is real work.
An assistant answering from twenty documents needs very little. Anything that predicts an outcome needs enough past examples to learn from, and that number is worth knowing before you commit.
A one-off export goes stale straight away. Anything that runs every day needs a way to keep its information up to date, and that’s part of the build.
Customer data collected for one purpose can’t always be used for another. Ask whoever handles your data protection early. It’s cheaper than asking after launch.
You don’t need technical knowledge to judge an AI consultant. Ask these five questions and notice whether the answers are about your business or could be about anyone’s.
Slow down if someone recommends tools before asking about your data, delivers only documents with no test of whether anything works, or prices the work by the size of your company rather than the size of the job. For what the job involves day to day, read what an AI consultant does.
In these cases we’d tell you to skip the review and keep your £149.
Skip the advice and ask for a quote for the build.
The consulting would cost more than it saves. A template or a checklist will do.
If no one can pick it up in the next six months, the shortlist will be out of date before anyone uses it.
Write the process down and tidy it first. That often gets most of the benefit before any AI is involved. Start with where service businesses waste time.
If you just need an off-the-shelf tool, our guide to AI tools for service businesses covers the main options, with prices.
The first and last are anonymised.
A professional services firm came to us wanting an AI assistant for prospective clients. The assessment found their most expensive repeated work wasn’t answering prospects. It was senior staff re-explaining process details to existing clients mid-project. Same capability, different audience, a bigger saving, and the process documents already existed. The assistant they built was internal first and client-facing second.
On the Evietek build, we started with a content audit and a new site structure. Only then did we add the conversion layer: Calendly booking, WhatsApp chat, contact forms and a chatbot. Bookings and chats are up 25% since launch. An assistant added to a site whose pages can’t answer a question inherits the same gap.
A distributor wanted to predict which customers were about to stop ordering. A good idea, and the data question stopped it. Order history lived in one system, the reasons customers left lived in nobody’s, and nothing linked the two. We recommended six months of recording three fields whenever a customer left, then a second look. They did, the second conversation was much shorter, and the project worked. Turning it down cost us a build. The client has since referred two others.
This is a summary, not legal advice. For your own situation, read the official guidance linked here or ask a data protection adviser.
If an AI tool handles customer or staff data, UK GDPR applies as it does to any other processing. The ICO’s guidance on AI and data protection says you must be open about how AI uses people’s data, and that most uses of AI will need a data protection impact assessment. The ICO is reviewing that guidance after the Data (Use and Access) Act 2025.
Since 5 February 2026, the Data (Use and Access) Act 2025 has changed the rules on significant decisions made about people by automated means. More lawful bases are available, safeguards still apply, and special category data such as health information stays more protected.
It can apply to a UK business when its AI system’s output is used in the EU (Article 2). Its transparency rules, which include telling people when they’re dealing with an AI system, apply from 2 August 2026. Rules for high-risk uses such as hiring and credit decisions apply from 2 December 2027 (Article 113).
With us, the first 30-minute call is free and the AI Opportunity Review is £149, one-off. You get a written shortlist of your top 3 use cases within 5 working days, and the fee comes off any build you commission within 60 days. Pilots and builds are quoted per project at a fixed scope.
A consultant decides what to do, in what order, and whether your data can support it. A developer builds it. We do both, which keeps our advice honest, because we have to be able to build what we recommend. If a consultant can’t build, ask who will, and bring them in early.
Usually less of it than larger firms. A ten-person business needs an honest answer on which one or two jobs are worth automating, and a check that the data supports them. That’s an hour and a short written review. If someone proposes a long programme for a business your size, ask what the extra weeks buy.
A 1-hour remote session with you and the person who does the work, a quick look at your data and systems, and a written shortlist of your top 3 use cases. Each one gets a feasibility verdict, a rough cost and a likely saving. Our recommendation can be “not yet”. It arrives within 5 working days.
The review takes about a week: an hour together, then the written shortlist within 5 working days. A pilot or build is a separate project with its own timeline, agreed before work starts. For custom AI builds we aim for a working prototype on your real data in about 2 weeks.
Only if your AI system’s output is used in the EU, for example by EU customers. Its transparency rules apply from 2 August 2026, and the rules for high-risk uses such as hiring apply from 2 December 2027. For most UK small businesses, UK GDPR and the ICO’s guidance matter more. This isn’t legal advice.
That’s a common place to start. First we work out why, because the fix depends on it: the wrong problem, data that couldn’t support it, a tool nobody fitted into the working day, or success that was never defined. It’s a short piece of work, and it usually shows what to try next.
Tell us where your team’s time goes and what you’ve already tried. We’ll tell you which ideas are worth a review, which depend on data you may not have, and what we’d do first.
Tell us about your business, the job that eats your week and what data you keep. We’ll reply within one business day to arrange a time.
A real person reviews your notes and comes back within one business day.
We go through what you need, your budget and your deadline, then tell you honestly which option fits.
Scope, milestones and pricing in writing within 48 hours of the call.
Review £149
