AI Readiness Template & Scorecard (Copy-Paste)

Copy the block below into a doc or spreadsheet, once per workflow you are considering. Score each factor 1 to 5, where 1 means "not true at all" and 5 means "clearly true today". Be honest about the present state rather than what you intend to fix later — the point of the scorecard is to catch the gap between those two.

Workflow being scored: ______________________
Who owns it today: ______________________

1. Repeatability — this task happens at least weekly, in roughly the same shape each time.  Score: __ / 5
2. Clear inputs and outputs — we can describe what goes in, and what a good and a bad result look like.  Score: __ / 5
3. Accessible data — the material the task needs is findable and reasonably clean, not scattered across inboxes.  Score: __ / 5
4. Review capacity — a named person can check the output before it reaches a customer, and has time to.  Score: __ / 5
5. Measurable upside — we can state the win in hours or money, and we could tell in a month whether it happened.  Score: __ / 5

Total: __ / 25
Decision: start / fix first / wait
If starting — the one workflow we test for two weeks: ______________________

Fill this in for three or four candidate workflows before you decide anything. The ranking is the output that matters, not any individual score. Most teams find the workflow they assumed was obvious scores lower than one they had not considered, usually because it fails on review capacity.

20–25 — start. This workflow is a realistic first AI project. Test it with an off-the-shelf tool for one to two weeks before committing to anything larger or custom-built.

15–19 — fix one thing first. Find the factor that scored lowest and address only that. A 2 on data accessibility usually means a week of consolidation, not a project. Re-score afterwards.

Below 15 — wait, and pick a different workflow. A low total across several factors means the process itself is not stable enough yet. AI will add confusion rather than remove friction. This is a legitimate outcome, not a failure.

One override applies regardless of the total: if review capacity scored 1, do not start, however high everything else scored. An AI workflow with nothing checking its output is not a time saving — it is unreviewed work reaching your customers, and the cost of that shows up later and larger. Our AI readiness checklist covers what a good review step actually looks like.

The five factors hold for any size of company, but larger teams need governance scored alongside them: who owns the data, who approves the workflow, who is accountable when an output is wrong, and what the access controls are. Add those as a sixth factor out of 5 and treat 30 as the new total.

For the full enterprise-grade version of this — governance, ownership, and approval flow scored properly — use the business AI readiness checklist. If your gap is specifically the data rather than the process, the data readiness checklist is the more precise starting point.

If you would rather not score by hand, our 2-minute AI readiness quiz asks the same questions and returns the score and ranking automatically, and the ROI calculator turns factor 5 into an actual number.