An AI readiness assessment scores your team across five areas: workflows, skills, data, tools, and culture. It tells you whether you’re ready to get real value from AI, or just ready to waste money on it.
That difference matters more than most people think. Cisco surveyed 7,985 business leaders across 30 countries and found that 98% feel urgency to deploy AI. Only 13% are actually ready. The urgency is rising. The readiness is falling.
If you’re already using AI and want to check what’s working, hire an AI auditor or use the AI audit checklist to run it yourself. This page is for the step before: figuring out if your team is ready to start. And if you haven’t yet mapped where AI fits your work at all, run a broader AI assessment first.
What an AI readiness assessment actually checks
Most people assume readiness means budget and technology. Do we have the money? Do we have the tools? Those matter, but they’re not where teams get stuck.
McKinsey’s State of AI report found that 88% of companies now use AI in some form. But only about 6% are high performers, meaning AI actually moves their bottom line. The other 82% are paying for tools without getting the return.
The gap isn’t money. It’s readiness.
A useful AI readiness assessment checks five things:
- Where AI fits your workflows (not just “can we use AI?” but “where would it actually save time?”)
- Whether your team can use the tools (not just awareness, actual skill)
- Whether your data is usable (clean enough, accessible enough, in one place — see AI data solutions for the full breakdown)
- What tools you already have (most teams are paying for things they don’t use)
- Whether your team will actually change (culture is the quiet killer)
I’ll walk through each one. But first, why this matters right now.
Why AI readiness matters now
The numbers are rough. S&P Global surveyed over 1,000 companies and found that 42% abandoned most of their AI projects in 2025. That’s up from 17% just a year earlier. More than double.
Gartner predicts that through 2026, 60% of AI projects without AI-ready data will simply be abandoned. Not fail slowly. Abandoned.
For a small team, this is even more dangerous. A company with 500 people can absorb a failed AI pilot. A team of 15 can’t. You’ve got one shot to get this right, maybe two. A readiness assessment keeps you from spending three months and real money on something that was never going to work.
The barriers to AI adoption are mostly human, not technical. A readiness assessment catches them before you start.
My take: The companies I see struggling didn’t fail because AI is hard. They failed because they skipped the boring step. They picked a tool before they picked a workflow. A 45-minute assessment would have saved them months.
The five areas a real assessment scores
Enterprise AI readiness frameworks love complexity. Cisco’s index has six pillars. TDWI’s assessment has 75 questions across five dimensions. Microsoft built a seven-pillar model.
That’s built for companies with dedicated data teams and a Chief AI Officer. If you’re a team of 10 to 50 people, you need something you can run in a morning.
Here are the five areas that actually matter, with the diagnostic questions for each.
1. Workflows
This is the one everybody skips, and it’s the most important. MIT’s Center for Information Systems Research studied 721 companies and found that the biggest jump in financial performance comes from one thing: moving past pilots into actually changing how work gets done. Not better AI tools. Not bigger budgets. Just changing the workflow. Most companies are stuck running pilots that never become real.
Ask yourself:
- Which three tasks eat the most time every week?
- Could AI handle 50%+ of any of those tasks?
- Has anyone on the team actually tried?
Score: 1 (no idea where AI fits), 2 (we’ve identified some tasks), 3 (we’ve tested AI on specific workflows)
2. Skills
Deloitte’s 2026 State of AI report surveyed 3,235 leaders and found that the skills gap is the number-one barrier to AI integration. Not budget. Not technology. Skills.
This isn’t about whether your team has heard of ChatGPT. It’s about whether they can prompt well, evaluate the output, and know when to override it.
Ask yourself:
- Can your team write a prompt that gets a useful result on the first or second try?
- Do they know how to check AI output for errors?
- Have they built any repeatable process with AI tools?
Score: 1 (most people haven’t tried), 2 (a few power users, rest are watching), 3 (most of the team uses AI tools in their daily work)
IDC estimates that AI skills shortages could cost the global economy $5.5 trillion by 2026. Only 33% of employees received any AI training last year. The training gap is bigger than the technology gap.
3. Data
If your team’s knowledge lives in random email threads, scattered Google Docs, and someone’s notebook, AI won’t help much. It needs clean, accessible information to work with.
Cloudera and Harvard Business Review found that only 7% of companies say their data is completely ready for AI. The split surprised me: 65% say their structured data (spreadsheets, databases) is ready, but only 39% say the same for unstructured data like emails, PDFs, and chat messages.
For a small team, most of your useful data is unstructured. Client conversations. Meeting notes. Slack threads. That’s the hidden readiness gap. The AI-ready data checklist covers the four things to check before you plug any data into a model.
Ask yourself:
- Is your customer data in one place, or scattered across five tools?
- Could a new hire find last quarter’s campaign results in under 10 minutes?
- Do you have a consistent way to name and organize files?
Score: 1 (data is everywhere, no system), 2 (some things are organized, others are a mess), 3 (most data is accessible, labeled, and in shared systems)
If data is your weakest area, AI data integration covers how to fix it step by step.
4. Tools
Before buying anything new, check what you already have. SAS surveyed 1,600 SMB leaders and found that 67% have data scattered across multiple systems with no connected strategy. Many are paying for tools that overlap or go unused.
Ask yourself:
- What AI features are already built into tools you pay for? (Most CRMs, email platforms, and project tools have AI features now.)
- Is anyone on the team actually using those features?
- How many tools does your team use that do roughly the same thing?
Score: 1 (no idea what we have), 2 (we know our tools but haven’t explored AI features), 3 (we’ve audited what we have and know what’s missing)
Before buying more, see which AI tools for business actually matter at your stage.
5. Culture
This is the one that kills projects after everything else is in place. Accenture found that only 16% of organizations have fully integrated AI into their processes. But those 16% outperform their peers by 2.5x in revenue growth.
The difference isn’t technology. It’s whether people actually change how they work.
Ask yourself:
- When you introduced a new tool last time, did people use it after the first week?
- Does your team see AI as a threat, a toy, or a tool?
- Is there someone on the team who’s already quietly using AI for their work? (Ethan Mollick at Wharton calls these people “secret cyborgs”. Most teams have at least one.)
Score: 1 (resistant or indifferent), 2 (curious but no habits yet), 3 (the team experiments and shares what works)
My take: Culture is the one you can’t shortcut. I’ve seen teams with perfect data and great tools go nowhere because the team lead didn’t buy in. And I’ve seen scrappy teams with messy data do incredible things because everyone was genuinely curious. Start with curiosity. The rest follows.
These five areas are exactly what I walk through with founders and marketing leads when we work together. It takes about 45 minutes, and you walk away with a score and one clear next step.
How to run your own assessment
Most AI assessment tools are built for large organizations with data teams. Here’s how enterprise and small-team versions compare:
| Enterprise assessment | Small-team assessment | |
|---|---|---|
| Time | Weeks to months | A morning |
| Questions | 50-75 (TDWI, Cisco) | 15 diagnostic questions |
| Output | Maturity score + report | One workflow to pilot |
| Who runs it | Consulting team or vendor | You, maybe with an advisor |
| Cost | $10K-50K+ | Free (or a single advisory call) |
For a team under 50, the right-hand column is what you want. Here’s how to do it:
Step 1: List your workflows. Write down every core task your team does regularly. Content creation, research, reporting, email campaigns, competitor analysis, customer support, onboarding. Don’t judge yet. Just list.
Step 2: Score each area. Go through the five areas above (workflows, skills, data, tools, culture). Rate yourself 1-3 on each one. Be honest. A 1 isn’t bad. It’s information.
Step 3: Find the biggest gap. Which area drags the others down? A team with great skills but terrible data will spin its wheels. A team with great data but no skills will stare at it. Your weakest area is your starting point.
Step 4: Pick one workflow to pilot. Choose the task where AI would save the most time AND where you have enough readiness to support it. Don’t pick the hardest problem. Pick the one most likely to work.
Step 5: Set a baseline. Before you change anything, measure. How long does this task take now? What’s the output quality? How many times a week do you do it? You need this to know if AI actually helped.
The whole thing takes a morning. If your assessment takes longer than a day, you’re over-engineering it.
Once you know your score, the AI adoption framework gives you the phased plan for what comes next. And implementing AI walks through the execution.
For small teams wondering whether outside help is worth it, here’s what AI consulting for small businesses actually looks like.
What most assessments get wrong
There’s a version of AI readiness that looks like progress but isn’t. Companies fill out long questionnaires, score well on every pillar, get a nice report, and still fail. 42% abandoned their AI projects despite having strategies, budgets, and the right tools.
I call this readiness theater. Checking boxes without changing behavior.
Charlene Li, author of Winning with AI (co-written with Dr. Katia Walsh, one of the world’s first Chief AI Officers), put it bluntly. She and Walsh built a full AI readiness assessment: weighted scores, detailed questions, the works. Then they threw the whole thing out.
Their finding: “Readiness assessments don’t eliminate risk. They delay progress and make your uncertainty look more official.”
That’s a strong take, and I think it’s half right. A bad assessment does exactly what she describes. It finds gaps (because that’s what it’s designed to do), produces a report that says you’re not ready, and becomes a delay mechanism. Three months later, you have a deck and no progress.
But a good assessment does the opposite. It takes 45 minutes, ends with one specific action (“try AI on this workflow for two weeks”), and gives you a baseline to measure against. No deck. No maturity score. Just a clear next step.
The difference between a useful assessment and readiness theater is whether it ends in one specific action or a report.
Gartner found that companies getting real results from AI invest four times more in the basics: data quality, people, and change management. The assessment tells you which basic to fix first.
If you want to understand what AI consulting actually covers, that page walks through the whole process.
How I run this assessment
The assessment is the first step of any engagement I do, before we talk about tools, strategy, or budgets.
It works like this: 45 minutes on a call. We walk through your workflows, skills, data, tools, and culture. You score each area honestly. I push back where the score doesn’t match what I’m hearing. We find the biggest gap together.
By the end, you know three things:
- Where you actually stand (not where you hope you stand)
- Which single workflow to try AI on first
- Whether you need outside help or can handle the next step yourself
No pitch. No 47-slide deck. If you can handle it alone, I’ll say so. If you need help, we talk about what that looks like.
If you’re a small team trying to figure out whether AI is worth the investment, this is the starting point. Book a 15-minute spar and I’ll tell you where you stand.
And if you want to see the specific workflows where small teams get the most from AI, AI for small business marketing is the place to start.
FAQ
What is an AI readiness assessment?
A structured way to figure out if your team can get real value from AI. It checks five areas: workflows (where AI fits), skills (can people use it), data (is it usable), tools (what you have vs. what you need), and culture (will the team actually change how they work). A good one takes a morning. A bad one takes a quarter and ends with a deck nobody reads.
How do you assess AI readiness?
Score yourself 1-3 on each of the five areas (workflows, skills, data, tools, culture). Find your weakest area. Pick one workflow where AI could save real time and you have enough readiness to support it. Set a baseline (time, output, quality) before changing anything. The goal isn’t a maturity score. It’s one clear action.
What are the key areas of AI readiness?
Five areas cover it for most teams: workflows (where does the time go?), skills (can people prompt, evaluate, and iterate?), data (clean, accessible, and in one place?), tools (what’s being used vs. sitting idle?), and culture (will the team try and stick with it?). Enterprise frameworks add governance and strategy as separate categories. For a team under 50, those fold into the five.
What questions should an AI readiness assessment ask?
The most useful questions are specific, not abstract. “Which three tasks eat the most time every week?” “Can your team evaluate AI output for errors?” “Is your customer data in one place?” “How many tools do you pay for that do roughly the same thing?” “When you introduced a new tool last time, did people actually use it?” Those tell you more than “rate your AI strategy on a scale of 1-5.”
How long does an AI readiness assessment take?
A practical one for a small team takes half a day. You’re listing workflows, scoring five areas, and picking one pilot. Enterprise versions like Cisco’s index or TDWI’s 75-question model can take weeks. If yours takes longer than a day, you’re measuring too many things and acting on too few. An AI checklist can help you keep the scope tight.