AI Readiness Assessment
Is my company ready for AI?
Your company is ready for AI if your data is trustworthy, your core systems are used properly, your key processes are written down and followed, someone owns the outcome, you have basic rules for AI use, and you can name the business result you want. Most companies past $10M are ready in some of these areas and not in others. The weak spots, not the tools, decide whether AI pays back.
The honest answer to "are we ready?" is usually "partly." That is not a failure. It is a map. This article walks the six pillars of our readiness model as a self-check, so you can see where you are before you spend on anything.
What does AI readiness mean?
AI readiness is whether your company's own conditions will let an AI tool pay back, whichever tool you pick. It is not about the technology you buy. It is about whether your data can be trusted, your systems and processes are in shape to connect to something new, your people have an owner and the time to change how they work, and you have basic rules and a business goal.
We measure it across six pillars: Data, Systems, Process, People, Governance and Strategy. Because one weak pillar stalls the whole effort, readiness is a profile, not a yes or a no. A company can be ready to try one narrow use case and not ready for ten.
Why does readiness matter more than the tool?
Because the tools are not what is failing. MIT's NANDA initiative, in The GenAI Divide (2025), found that about 95% of organizations saw no measurable P&L return from their generative AI pilots. Gartner, in July 2024, predicted that at least 30% of generative AI projects would be abandoned after proof of concept by the end of 2025. Neither says the technology does not work. Both point to what surrounds it: unclear value, poor data, weak ownership.
That is why we start with the business, not the tool. If you want the formal version, see what an AI readiness assessment includes.
What are the six pillars of AI readiness?
For each pillar below, read both columns honestly. If your company looks more like the right column, that is where to start. Score yourself green, yellow or red.
| Pillar | Signs you are ready | Signs you are not |
|---|---|---|
| Data | One agreed answer for "how many customers do we have" and "what did we sell last month." Someone owns each key data set. Reports match across departments. | Two systems disagree about the same customer or order. Leaders keep their own spreadsheets. Nobody trusts the month-end numbers until a person fixes them by hand. |
| Systems | Core systems are current, used the way they were designed and able to connect to others. People rarely work around them. | Key work happens in email and spreadsheets next to the ERP. Systems cannot share data without manual re-entry. Nobody knows which features you already pay for. |
| Process | The five or six workflows that matter are written down, and the real work matches the document. You can say where time and errors pile up. | Every person does a task a little differently. Procedures live in one person's head. Exceptions are the norm. |
| People | Leaders can say who will own an AI initiative. Staff have time to learn. There is a culture of trying and reporting back. | Everyone is already at capacity. Staff fear for their jobs or quietly ignore new tools. Nobody is named to own the result. |
| Governance | A short AI policy exists. People know what data stays out of AI tools. Output is reviewed. Vendor terms get a second look. | No policy. Staff use personal AI accounts with company data. Nobody has read the AI terms in the contracts you have signed. |
| Strategy | You can name two or three business outcomes you want from AI and how you would measure them. Someone owns the budget. | The goal is "do something with AI." Ideas come from vendor demos. No one can say what success looks like. |
What does "ready" look like in practice?
Ready does not mean perfect. It means that when you try an AI use case, the surrounding conditions will not sink it. Here is one example from David's work before he founded Elevate, on the Data and Systems pillars.
A multi-brand auto parts distributor, owned by a private equity sponsor, was running two systems of record that disagreed with each other, with thirteen storefronts on top. An order entered in the new ERP could break in the old one. Before anyone could think about smarter tools, the team had to write down 240 requirements and agree on which system was the truth. That is the unglamorous work readiness depends on. You can read more in the distributor ERP consolidation case study. David left before go-live, so we do not claim how it ended.
Notice that the fix was not an AI tool. It was agreement on data and process. In our experience, that is true of most companies we see.
Ten questions, three minutes, a score out of 100 across the six pillars and the one thing to fix first.
Get your AI Readiness ScoreWhat should you do with your scores?
- Find your weakest pillar. That is the one that will stall you, no matter how strong the others are.
- Fix the cheapest red first. Governance is usually the quickest. A one-page policy can be done this month. Our guide to writing an AI policy for employees includes a template.
- Pick one workflow, not ten. Choose a process you understand, with data you trust, and an owner who wants it fixed.
- Decide how you will know it worked. A number you can measure before and after: hours, error rate, days to collect, quote turnaround.
- Be wary of who is telling you the answer. A vendor will tell you that you are ready for their product. We do not sell software, so an assessment from us can say "not yet."
What score means we are ready?
No single number makes a company ready, and we would be wary of anyone who claims one. The free AI Readiness Score gives you a score out of 100 across the six pillars, and the more useful result is the spread behind it. A company with a high total and one weak pillar still stalls on that pillar. A middling total with an owner, a clean data set for one workflow and a one-page policy is often better placed to start.
Use the score to decide what to do next. Fix the weakest pillar first, start with a single workflow you understand, and rescore after the changes. If the score and your gut disagree, a full assessment, with interviews and a review of your real systems, settles it.
When is a company not ready, and what does that mean?
Not ready does not mean never. It means the next dollar is better spent on the foundation than on a pilot. A company with red in Data and Process that buys an AI tool is likely to join the 95%. The same company that spends a quarter fixing those two usually finds the AI use case gets smaller, cheaper and more obvious.
The reverse also happens. Some owners assume they are behind and find they are in better shape than they thought. A clean ERP, a trusted data set and a clear owner are bigger advantages than most people realize.
When you are ready to go from a self-check to a scored result, with interviews and a review of your real systems, that is what our AI Readiness Assessment does in three to four weeks.
Common questions
What is the most common reason companies are not ready for AI?
Data and process, not technology. In our experience the usual blockers are scattered or untrusted data and work that is done differently by every person. AI tools cannot fix either one.
Can we be ready in some areas and not others?
Yes, and most companies are. The point of scoring each pillar separately is to see which weak one will stall you. A company strong in five pillars still stalls on the sixth.
How long does it take to get ready?
It depends on the gaps. Governance gaps, such as a missing policy, can close in days. Data and process gaps usually take a few months of steady work. An assessment gives you the order and the owners.
Is the free AI Readiness Score enough?
It is a good first read. Ten questions give you a score across the six pillars and the one thing to fix first. It is not a substitute for interviews and a review of your actual systems, which is what a full assessment does.
Not sure where to start?
Take a free three-minute score. If a conversation makes sense, it will offer one.