Most AI readiness conversations happen after the project has already stalled, when someone finally asks why the build is taking three times longer than the vendor quoted.
Run earlier, the same conversation is an afternoon well spent. Run after the fact, it is a postmortem.
Short answer: a readiness check scores four areas of your business, people, process, data, and leadership, from one to five. Your lowest score, not your average, tells you what to fix before you spend anything. A company can be excellent in three areas and still fail on the fourth.
Why this belongs in Q1
Most leadership teams will only sit still long enough to assess themselves once a year, and that window is usually January and the first weeks of February, before the operating calendar fills back in.
Doing this now, before any vendor conversation, changes the shape of every conversation that follows. You arrive at a vendor meeting knowing your own constraints instead of discovering them mid-build.
The four areas
People
Do the people whose work would change understand why, and would they support it. Score one if nobody has been consulted, three if the topic has been raised but not discussed, five if the affected team helped identify the candidate workflow themselves.
Process
Could two people in your company describe your target workflow the same way, including its exceptions. Score one if it varies by who is doing it, five if it is documented, current, and the edge cases are known.
Data
Is the information the system would need accurate, findable, and consistently structured. Score one if it is scattered across inboxes and spreadsheets, five if it is clean and accessible with a single definition per field.
Leadership
Is there a named owner with allocated time, and has leadership agreed on what stays under human control. Score one if the interest is enthusiasm without hours behind it, five if ownership and governance are both settled.
Scoring it honestly
| Lowest score | What it means | What to do first |
|---|---|---|
| 1 | A structural gap that will sink any build | Fix that single area before scoping anything |
| 2 to 3 | Workable with a contained, low-risk first project | Choose a workflow that does not depend on the weak area |
| 4 to 5 | No structural blocker in this dimension | Proceed to workflow selection and sequencing |
Notice the logic. A company with excellent data and process but no named owner will still fail, because ownership is the dimension that catches problems after launch. Averaging the four scores hides that risk. Taking the lowest one exposes it.
What the check usually finds
Three patterns show up more than any others. Strong systems paired with undocumented process, which is a whiteboard problem rather than a technology one. Strong intent paired with no allocated ownership, the single most reliable predictor of a stalled project. And clean data paired with no governance conversation, which is cheap to fix now and expensive to retrofit onto a live system later.
What to do with a low score
A low score is not a reason to abandon AI for the year. It is a reason to choose your first project carefully. A company weak on data can still succeed with an intake workflow that generates its own clean input going forward, rather than one that depends on reading a messy history.
FAQ
How long does this actually take?
Ninety minutes with the right three or four people in the room, longer only if you need to go find out how a process actually works.
Who should be in the room?
Whoever owns the candidate workflow, whoever owns the data, and whoever would be accountable for the result.
Should we do this every year?
Yes, and it is worth repeating after the first project completes, since the categories move once a real build forces clarity.
What if we score low in every category?
Then the first project is not AI. It is documenting one process and cleaning one data source, which has value on its own.
Does a strong score guarantee success?
No, but it removes the predictable failure modes. The unpredictable ones remain, as they do in any project.
Where to go next: Run the four categories with your leadership team this month. Bring the lowest score into the next conversation about what to build first.
