You've seen the demo. You've got a budget line. You're ready to "do AI." Stop for a week. The companies that get real value from AI almost never start with a tool — they start with an honest look at whether the business can absorb it. Here's the readiness check we run before anyone signs anything.

The five questions before you spend a dollar

One: What specific decision or task gets faster or better? If you can't name it in a sentence, you're buying a hammer and hoping to find a nail. Two: Who owns the outcome — not the tool, the outcome? Three: What does "good output" look like, and who decides? Four: Where does the input data live, and is it clean enough to trust? Five: What happens when the AI is wrong? If you have crisp answers to four of these five, you're ready. If you're guessing on three or more, you're not.

Your data is the real bottleneck

The single biggest predictor of an AI project stalling isn't the model — it's the state of the data feeding it. If your customer history is spread across three systems and a spreadsheet someone maintains by hand, the AI inherits that mess. You don't need a data warehouse to start. You do need to know, for the one workflow you're targeting, where the source of truth lives and how often it's wrong. Fix the obvious gaps before automating on top of them.

Someone has to own this

AI initiatives that succeed have a named human accountable for the result — usually a department lead, not IT. The owner sets the quality bar, reviews early output, and has authority to say "not good enough yet." When ownership is fuzzy, the tool gets bought, used twice, and quietly abandoned. That's the most common failure mode we see, and it has nothing to do with the technology.

The honest caveat

Readiness isn't a gate you pass once. The first workflow teaches you things that change your answers for the next one. Treat this as a recurring check, not a one-time approval.

Your next step

Pick one workflow this week. Answer the five questions on a single page. If you can't, that's not a failure — it's the most useful thing AI has done for you yet, because it told you where to look first.