The AI readiness checklist for a UK small business in 2026

Most UK small business owners now accept that AI is worth looking at. The problem is what happens next. They buy a ChatGPT subscription, watch a YouTube video, and spend an afternoon trying things that sort of work but not really. Three months later, the subscription is still live, but nothing has actually changed.

The reason is almost always the same: they skipped the readiness check. They jumped to the tool before they understood the problem.

This post is that readiness check. It is not a technical audit. It is a set of practical questions any business owner can answer in 20 minutes - questions that determine whether your business is in a position for AI to actually help, and if not, what to fix first.

Why readiness matters more than tools

Here is a pattern that shows up in assessment after assessment: a business owner has already spent money on AI tools before we speak to them. Sometimes several. A writing assistant, a meeting transcription tool, an automated scheduling thing they never finished setting up.

The tools are fine. The problem is that the underlying workflows were not ready for them.

AI works best when it is replacing a task that is already well-defined and repeated consistently. It struggles when the process it is meant to help with is different every time, poorly documented, or when the person using the tool has not made a decision about what they actually want the AI to produce.

Readiness is about the process, not the person. You do not need to be technical. You need to have thought carefully about what you want done and how.

The checklist

1. Can you name your three most time-consuming repeated tasks?

Not your busiest tasks - your most repeated ones. The things you do on Monday that you will also do on Tuesday and again next Monday. Writing the same type of email. Pulling together the same type of report. Answering the same enquiry for the fifteenth time.

If you cannot name three specific tasks without thinking for more than a minute, stop here. The first step is a 30-minute time audit: keep a rough log of what you actually do this week, hour by hour. You will be surprised what shows up.

If you can name them, write them down. These three tasks are your starting point.

2. For each task, can you describe it in one paragraph?

This is the test that separates automatable tasks from vague frustrations. "Admin takes forever" is not a task. "Every Friday afternoon I spend 90 minutes creating the weekly summary email for my 12 clients, pulling in updates from my project tracker and formatting them in our standard template" - that is a task.

If you can write one paragraph that explains exactly what the task involves, what the output looks like, and where the inputs come from, AI can almost certainly help with it. If you cannot write that paragraph, you are not ready to automate it yet - you need to define it first.

3. Do you already have the tools your process depends on?

AI tools work on top of existing systems. If you want to automate invoicing reminders, you need to be using invoicing software. If you want AI to help with client onboarding, you need an onboarding process to start from. If you want automated scheduling, you need to be running a calendar that can accept bookings.

A common mistake is buying an AI tool to fix a problem that is actually caused by a missing foundation - no CRM, no consistent intake process, no standard template. The AI cannot build the foundation for you. You have to do that first.

Go through each of your three tasks. What tools does the current process rely on? Do you use them consistently? Are the outputs stored somewhere retrievable?

4. How much would one recovered hour per day actually be worth?

This question matters because it sets your investment threshold. A plumber billing £60 an hour and a management consultant billing £200 an hour have very different equations. For the plumber, recovering one hour a day is worth £300 a week in additional capacity. For the consultant, it is £1,000 a week.

Calculate your number. If you bill by the hour, use your day rate divided by eight. If you are on a fixed salary or draw, use your annual income divided by 1,600 (a standard working-year estimate in hours). This number is what a single recovered hour per day is worth to you in productive capacity.

Once you have it, you can assess any AI tool or service against it properly. A £50/month subscription that saves you 30 minutes a day pays for itself in under a week for most owners.

5. Have you tried at least one AI tool with a specific task?

Not "played around with ChatGPT" - actually used a tool on a real task from your working week. Asked it to draft a client email from bullet points. Asked it to turn a set of notes into a structured report. Asked it to write five social captions for next week.

If the answer is no, this is the quickest thing to fix. The gap between theoretical interest in AI and actual usefulness is almost always closed by a single good first experience. Pick the most straightforward of your three repeated tasks and spend 20 minutes testing a tool on it.

6. Is there one person in the business who owns the implementation?

If you are a solo business owner, this is obviously you. But if you have a team, even a small one, implementation almost always stalls when it is nobody's job. Somebody needs to be the person who sets the tool up, checks it is working, and troubleshoots when it breaks.

This does not need to be a technical person. It needs to be someone who is motivated to make the change, has time to spend an hour or two on setup, and will actually use the tool rather than reverting to the old way.

7. Do you have a way to measure whether it is working?

The question to answer before you implement anything is: how will I know in 30 days whether this made a difference? For time-saving tools, this is usually simple. How long does this task take you now? Write it down. After 30 days using the new tool, how long does it take? If the answer is "roughly the same", the tool is either not right for the task or you have not set it up properly.

Without a before-and-after measure, you will never know whether your AI tools are actually working or just adding noise.

What your score tells you

Count how many of the seven questions you answered yes to.

5-7 yes answers: You are in good shape. You have the process clarity and the motivation to make AI work. Pick the highest-value task from your list and start this week. The free quiz at the bottom of this post will help you prioritise.

3-4 yes answers: You have the intent but some of the foundations are not yet there. Focus on the questions you answered no to before buying any new tools. The most common gaps at this stage are task definition (question 2) and missing underlying systems (question 3).

0-2 yes answers: AI tools will almost certainly disappoint you right now - not because they are not good, but because the conditions for them to work are not yet in place. Spend two weeks on the time audit (question 1), define your three repeated tasks clearly, and then revisit this checklist.

The one thing to do today

If you take nothing else from this post, take this: write down your three most time-consuming repeated tasks before you close this tab. Not the vague ones - the specific ones you can describe in a sentence. That list is the starting point for everything else.

Once you have the list, the free HoursBack AI Readiness Quiz will help you identify which of those tasks is worth tackling first and what a realistic first step looks like. It takes two minutes and gives you a personalised action plan based on your specific workflows.

If you want to go deeper - a proper look at your business, your bottlenecks, and a five-day implementation plan built around your specific situation - that is what the HoursBack Assessment is for. But the quiz is the right place to start.

Ready to find out where you actually stand?

Take the free 2-minute AI Readiness Quiz and get your personalised action plan.