AI Readiness Assessment for Small Businesses: A 20-Minute Checklist
Most AI readiness assessments are built for large companies: maturity models, survey tools and a long report. A small firm needs something shorter. This checklist asks 20 yes/no questions about one process you'd like AI to help with, scores your answers as you go, and tells you what to fix before you spend money on a tool.
What "AI readiness" means for a small business
For a firm of five to fifty people, AI readiness isn't about data science teams or a company-wide strategy. It comes down to one question: is there a specific, repeated piece of work that you understand well enough to hand part of to AI, and can you tell whether it worked?
That's why this assessment starts with the process, not the technology. If you can't describe how a task gets done today, who does it, how long it takes and what "done" looks like, no tool will fix it. If you can, you're usually closer to ready than you think. You don't need a perfect data warehouse or an AI policy binder to start. You need one well-understood task and a way to measure it.
Pick one process before you start, ideally something that happens every week and frustrates people: client intake, quoting, invoicing, scheduling, report drafting or answering the same email questions. Answer every question about that process, not the business as a whole.
The checklist: 20 questions, five areas
Tick each statement that's true today. Be strict: "sort of" counts as no. Your score updates as you go.
How to read your score
- 0–7: Not ready yet, and that's fine. The gap is almost always the process, not the technology. Map the task first. Until you can describe it and time it, any tool you buy is a guess.
- 8–13: Ready to prepare. You understand the work but have gaps in ownership, cost or risk. Fix the lowest-scoring area before you trial anything.
- 14–17: Ready for a small test. Pick the one step in the process where AI could save the most time, run it on real examples with a person checking every output, and measure the cost per task.
- 18–20: Ready to implement. You have the basics. The remaining work is choosing the right tool for your stack and putting the change into daily use.
Look at the section scores too. A total of 14 with 0 out of 4 on risk is not the same as 14 spread evenly. Any section at 0 or 1 is the thing to fix first, whatever the total says.
Why the process comes first
Most small businesses that try AI and give up didn't pick the wrong tool. They automated a task nobody had written down, so nobody could say what the AI was supposed to do or whether it did it better. Mapping the process, step by step and person by person, shows where the time actually goes. Often the biggest win is removing a step or fixing a handoff, with no AI at all.
If the first section was your weakest, start with our guide to process mapping or see some business process mapping examples.
Price one finished task
The cost section is the one most firms skip, and it's the one that decides whether AI is worth it. Work out what one finished task costs today: the minutes, including checking and rework, multiplied by the loaded hourly cost of the person doing it, plus any other cost. Then estimate the same figure with AI, counting the human review time and the tool costs. If the second number isn't clearly lower, don't buy the tool. Our AI ROI calculator does this math for you.
A simple AI implementation roadmap
Once you have a score, this is the order we'd follow. It's a roadmap for one process, not a company-wide program. Repeat it for the next process once the first one is working.
- Map (week 1). Write down the steps, owners, tools and time for the chosen task. Mark where work waits.
- Price (week 1). Calculate the cost of one finished task today and set a target cost.
- Fix the basics (weeks 2–3). Close the gaps the checklist found: name an owner, gather test examples, write the data rule, decide what a person always checks.
- Test (weeks 3–6). Try AI on one step using tools you already pay for where possible. Check every output and record the time.
- Measure (week 6–8). Compare the new cost per task with the old. Keep it, adjust it or drop it.
- Roll out and repeat. Write the new steps into your procedure, train the team, then move to the next process.
How to use AI for a small business without a big project
For most small business owners, the first step isn't new software. It's using a general assistant like ChatGPT, Claude or Microsoft Copilot well on a task you've already mapped: drafting replies from your own templates, summarizing long documents, turning notes into a first-draft report, or cleaning up a spreadsheet. ChatGPT for small business owners works best with clear instructions, real examples of good output, and a person checking the result before it reaches a client.
Dedicated tools and automations (connecting your inbox, CRM or accounting system) come later, once you know which step is worth automating and what it costs today. Moving in that order keeps your spend tied to a task you can measure.
AI maturity assessment: the five stages
A readiness check looks at one process. An AI maturity assessment looks at the whole firm. Large-company maturity models and digital transformation assessment tools can run to dozens of pages; for a small business, five stages are enough to place yourself:
- Ad hoc. A few people use AI on their own, with no rules or shared approach.
- Ruled. There's an AI policy, approved tools and a named owner.
- Tested. At least one mapped process has had AI tested on a step, with a before-and-after cost per task.
- Embedded. What worked is written into procedures and staff are trained on it.
- Managed. AI use is reviewed quarterly against measures, and new processes go through the same map-price-test cycle.
Like a process maturity model, each stage depends on the one before; skipping from stage 1 to 4 is where most AI adoption frameworks go wrong in practice. Most small firms are at stage 1 or 2, which is a fine place to start. The checklist above tells you whether your chosen process is ready to take you to stage 3. For the year-long view, see the digital transformation roadmap.
Want a second opinion on your answers?
We'll take one process through this assessment with you: map it, price one finished task, and tell you plainly whether AI belongs in it yet. If it doesn't, we'll say so.
AI adoption in business: maturity model, metrics and where firms stand
Readiness is about whether you can start. Adoption is about whether AI is actually used once you have. They're different, and most firms stall between the two.
A simple AI adoption maturity model
- Informal: some staff use free tools on their own; nobody tracks it.
- Sanctioned: approved tools, basic rules, a few named uses.
- Embedded: AI is part of written workflows, with an owner and a review step for each use.
- Measured: every use has a cost-per-task number, and weak uses get switched off.
AI adoption metrics worth tracking
- Active use: share of people with access who used the tool in the last 30 days. Seats nobody opens are pure cost.
- Uses per person: how many different tasks each person uses AI for. Gallup's 2026 data found employees using AI for more types of work were more likely to report productivity gains.
- Cost per finished task: before and after, including review time. This is the one that reaches the P&L.
- Error and rework rate: how often AI output had to be redone.
AI adoption in marketing and other functions
Marketing is usually where AI adoption starts in a small business, because drafting posts, emails and ad copy is low-risk and easy to review. Use it as the pilot, then copy what worked to admin, finance and operations, where the bigger savings tend to sit.
What the IBM Global AI Adoption Index found
IBM's Global AI Adoption Index 2023, a survey of IT professionals at companies with more than 1,000 employees published in January 2024, found about 42% had actively deployed AI and another 40% were exploring or experimenting. The top barriers were limited AI skills and expertise (33%), data complexity (25%) and ethical concerns (23%). Those are large firms, but the barriers look familiar at any size: skills, messy data and trust. For how adoption turns into results, or doesn't, see why workplace AI productivity is stalling.
Why we insist on one loop, measured
MIT NANDA's GenAI Divide research found about 95% of organisations with no measurable profit impact within about six months. The efforts that paid were narrower: one real workflow, a baseline, and a decision date. Read our MIT hub breakdown, price the task with the AI ROI calculator, and run any pilot as the 90-day test instead of leaving it in pilot purgatory.
Frequently asked questions
What is an AI readiness assessment?
It's a structured check of whether a business, or one process within it, is prepared to use AI. It looks at whether the work is understood, whether the data is available, whether people will own and test it, whether the cost case adds up, and whether the risks are handled.
What questions should an AI readiness assessment ask?
For a small business: can you describe the process, is the information digital and reachable, who owns it, what does one finished task cost today, and which steps must a person always check. The 20 questions above cover those five areas.
Is there a free AI readiness assessment template?
Yes. The checklist on this page works as a template: pick one process, answer the 20 questions, and use the score guide. You can print the page or copy the questions into a spreadsheet.
Do small businesses need an AI readiness framework?
Not a heavy one. Large-company frameworks and maturity indexes are built for enterprise IT. A small firm gets more from checking one process at a time against a short list like this.
How long does an AI readiness assessment take?
The self-assessment takes about 20 minutes for one process. A fuller review that includes mapping the process and pricing the task usually takes a single working session.
What should I do if I score low?
Start with the lowest section. In most cases that means mapping the process and working out what a finished task costs today. Both are useful even if you never add AI.
What happens after an AI readiness assessment?
Fix the lowest-scoring area first, then test AI on one step of one process with a person checking every output. If you want that done with you, see AI consulting services or the AI implementation consultant page.