An AI readiness assessment is a structured check of whether a business can adopt AI successfully, scored across strategy, data, systems, people and governance. For a business under 50 staff, it works best when it scores readiness for one named problem, such as quoting or invoice chasing. It is shorter than an AI audit: it tells you whether you can start, while the audit tells you where AI will help most.
What does an AI readiness assessment actually measure?
An AI readiness assessment measures the gap between what AI needs to work and what your business currently has. The large frameworks agree on the broad areas. Cisco’s AI Readiness Index scores six pillars: strategy, infrastructure, data, governance, talent and culture.
Microsoft’s version uses seven pillars, adding model management and splitting business strategy from AI strategy. Both were built for large organisations with IT departments, GPU budgets and data teams.
A 12-person firm has none of those. For a small business, the same questions shrink to five plain ones. Do you know what you want AI to fix? Is the information it needs written down somewhere? Do your systems let it in? Will your team use it? Do you know what staff must never paste into it?
Why do most readiness assessments give small businesses the wrong answer?
Most readiness assessments give small businesses the wrong answer because they score the whole company in the abstract. A general score of “3 out of 5 on data” tells an owner nothing about what to do on Monday.
Readiness is always readiness for something. A plumbing firm can be fully ready to automate job booking, because every enquiry already lands in one inbox. The same firm can be nowhere near ready to automate quoting, because prices live in the owner’s head.
The gap is also between using AI and using it well. The Office for National Statistics found that UK business AI use rose from around 12% to around 35% of firms with 10 or more staff since late 2023. The same release found only 10% of those firms use AI extensively.
Most firms are “ready” enough to buy a ChatGPT subscription. Far fewer are ready to change how a job gets done. Bedrock AI’s view is that the second kind of readiness is the only one that changes anything.
How is an AI readiness assessment different from an AI audit?
An AI readiness assessment asks whether the business can support AI, while an AI audit asks where AI will create value and in what order. The readiness assessment is a gate. The audit is a map.
In practice the readiness check is often the first week of an audit. Our guide to how an AI audit works covers the full process, from stakeholder interviews to the opportunity register.
| Question | AI readiness assessment | AI audit |
|---|---|---|
| What an AI readiness assessment or audit answers | Can this business adopt AI safely now? | Where will AI help most, and in what order? |
| Typical length | A day to a week | Two to four weeks |
| Who can run it | The owner, with a checklist | An outside specialist |
| Main output | A score and a list of gaps | A prioritised opportunity register and roadmap |
| Scope | Foundations: data, systems, people, rules | Specific workflows and use cases |
| Best moment to use it | Before buying any AI tool | Before committing budget to a build |
An owner can run a readiness assessment alone. An audit usually needs someone who has seen the same patterns across many businesses, because the hardest part is spotting which bottleneck matters most.
What should an AI readiness checklist include?
An AI readiness checklist for a small business should score six areas against one named problem, each out of three. Pick the job first. Write it as a sentence: “We lose two hours a day chasing unpaid invoices.” Then score each area for that job only.
- Problem clarity. Can you name the task, who does it, how often, and what it costs in hours? Score 0 if the answer is “we should be using AI”, 3 if you can state the hours.
- Process consistency. Is the job done the same way each time? If three staff do it three ways, AI will copy the chaos. Score 3 only if a new starter could follow written steps.
- Information access. Does the information the task needs live in one place, in a format software can read? Emails and a shared drive score 2. The owner’s memory scores 0.
- System openness. Do your core tools (accounts, CRM, job management) connect to other software? Most modern cloud tools do. Desktop software from 2012 often does not.
- People and time. Is there one person who will own the change and has two hours a week to spend on it? ONS data shows that a lack of AI expertise is one of the barriers firms cite most often.
- Data rules. Do staff know which client, financial and personal data must never go into a public AI tool? Score 0 if nobody has said anything. Score 3 if the rule is written down.
A score of 14 to 18 means start a small pilot on that job now. A score of 8 to 13 means fix the lowest one or two areas first, usually process or information. Below 8 means that job is not the right first project, so score a different one.
How do you run an AI readiness assessment in a small business?
A small business can run a useful AI readiness assessment in a single afternoon, using the checklist above and three candidate problems. The work is in the choosing and the honest scoring.
Start by listing the jobs that eat the most hours and earn nothing directly. Invoice chasing, quoting, booking, onboarding and answering the same customer questions are the usual suspects. Pick the top three.
Score each one separately. The job with the highest total is your first project, even if it is the least exciting. A boring job that scores 16 beats an exciting one that scores 9.
Ask the person who actually does each job to score it with you. Owners tend to overrate process consistency, because they know how the job should run rather than how it does.
Keep the result on one page: the problem, the six scores, the lowest area and the fix. That page becomes the brief for any tool you buy or anyone you hire.
What happens after an AI readiness assessment?
After an AI readiness assessment, a small business either pilots AI on its highest-scoring job or spends four to eight weeks fixing the foundation that blocked it. Both are progress.
If the job scored well, run a pilot of four weeks with one clear measure, such as hours saved per week. If it scored poorly on data rules, write a short AI use policy before anyone touches client data. If it scored poorly on process, document the job first. That work is worth doing even if AI never arrives.
The mistake to avoid is buying a platform before the assessment. Tool-first buying is how firms end up with subscriptions nobody opens. Our analysis of why AI tools get abandoned traces most of those failures back to a workflow that was never mapped.
This is why Bedrock AI treats readiness as the first step of diagnosis. The point is to decide what gets built and in what order, and that starts with knowing what the business can support today.
FAQ
How long does an AI readiness assessment take? A self-run AI readiness assessment takes an afternoon for a business under 50 staff, if you score three named problems. A facilitated assessment usually takes one to five days, including staff interviews and a look at your systems and data.
Is an AI readiness assessment the same as an AI maturity model? No. An AI readiness assessment scores whether you can start one piece of AI work now. An AI maturity model describes stages an organisation passes through over years. Small businesses get more practical value from the readiness assessment, because it points to a specific next step.
Can a small business run an AI readiness assessment without a consultant? Yes. An owner can run a basic AI readiness assessment with a checklist and the staff who do the work. Outside help earns its fee when choosing between problems, or when the scores are low and you need to know which gap to fix first.
What is a good AI readiness score? On the six-area checklist above, 14 to 18 out of 18 means a job is ready for a pilot. Scores of 8 to 13 mean fix one or two foundations first. A score below 8 means pick a different job to start with.
Bedrock AI finds where AI will actually save your business time and money, before you spend anything on building. Book a free 15-minute call