Is Your Business Actually Ready for AI? A 10-Minute Self-Assessment

By Domenic DiNatale

Most of the advice about AI readiness is written to make you feel behind so you'll buy something. This isn't that. This is a short, honest self-assessment you can run on your own business in about ten minutes, with nobody selling you anything at the end of it.

The goal is simple: find out whether an AI project would actually land in your company right now, or whether there's cheaper, more basic work to do first. Because here's the thing nobody says in the sales meeting: plenty of businesses aren't ready, and for them the smartest AI move is to not spend on AI yet, and fix the thing underneath instead.

Four areas decide it. Score yourself honestly on each.

1. Data: Can You Find the Truth?

AI runs on your data, so the first question is whether your data is in any state to be run on.

Ask yourself:

• If I needed every record about one customer (orders, emails, invoices, notes), could we assemble it, and would the pieces agree with each other?
• Is there one place that's the source of truth for the things that matter, or do three systems each have a different version?
• How much of what actually runs this business is written down anywhere, versus living in people's heads?

Green if there's a findable, mostly-trustworthy source of truth. Yellow if you could assemble it with effort. Red if your systems openly disagree and nobody fully trusts any of them. If you're red here, that's not an AI problem: it's a data problem that AI will only amplify, and it's where your money should go first. (See you're paying to type it twice: disagreeing systems are usually a symptom of this.)

2. Process: Do You Know How the Work Actually Happens?

AI applied to a process you understand is leverage. AI applied to a process you don't understand is an expensive way to scale confusion.

Ask yourself:

• Could I describe, step by step, how one important thing gets done here, or does it just "happen"?
• Is the process roughly the same each time, or does everyone do it their own way?
• Are there steps we do only because we've always done them?

Green if your key processes are understood and reasonably consistent. Yellow if they're consistent but undocumented. Red if nobody could actually draw the steps. A red here means the honest next move is to understand and tighten the process first, because AI won't fix a broken process, it'll just run it faster.

3. Decision Boundaries: Do You Know Where the AI Stops?

This is the one businesses skip, and it's the one that quietly kills projects. You need to know, in advance, where the AI's judgment ends and a human's begins.

Ask yourself:

• For the work I'd hand to AI, what happens if it's wrong and nobody catches it?
• Which of those outcomes are "fix a draft in two minutes," and which are "money moved or a promise went out"?
• Am I prepared to keep a human on the button for the second kind?

Green if you can clearly separate the draft-it work from the decide-it work. Yellow if you're not sure. Red if your instinct is "just let the AI handle it end to end." That instinct is exactly what blows projects up. The whole rule fits on a sticky note: let it draft, never let it decide.

4. The First Task: Do You Have One Real, Painful, Repetitive Job?

AI readiness isn't really about readiness in the abstract. It's about having a first task worth doing: one concrete, painful, repetitive, well-understood job where a win would be obvious and trust could be built.

Ask yourself:

• What does my team do over and over that everyone finds tedious?
• Where do we spend hours producing something from information we already have: a summary, a first draft, a lookup, a report?
• Is there one task where, if it got 70% faster, everyone would immediately feel it?

Green if an obvious candidate jumped to mind. Yellow if you have a few maybes. Red if nothing came to mind, which usually means you haven't yet found where the pain actually is, and that's the thing to figure out before anything else.

Reading Your Score

Mostly green: you're ready. Pick the task from #4, draw the line from #3, and start on your real data, with a named owner.

A couple of yellows: you're close. Do the small prep work the yellows point to, then start. It's usually weeks, not months.

Any reds: don't start with AI yet. The red is telling you where the real money should go first: the data, the process, the pain you haven't located. Fixing that isn't a detour from your AI project. It's the part of the AI project that determines whether it works.

The Takeaway

Readiness for AI isn't about how advanced you are. It's about four plain things: whether your data is trustworthy, whether you understand your own processes, whether you know where the AI stops, and whether you've got one real painful task to point it at first.

Spend the ten minutes. If you come out green, you'll start from a position that actually ships. If you come out red, you'll have just saved yourself the cost of an AI project that was never going to work, and found the more valuable thing to fix instead.

A short, paid assessment is often the cleanest way to get an outside read on all four. If you want one, that's exactly what we do first. More on the Intellitech blog.