The AI Prototype Readiness Checklist for Managers

A good prototype is dangerous in the best possible way. It proves that a workflow can be better. People start using it. Someone says, “Could we give this to the whole team?” and suddenly the fun little experiment is standing near the production door wearing a fake moustache.

Before letting it in, run a readiness check.

1. Is the workflow real and repeated?

A prototype deserves attention when it supports work that happens often enough to matter. One-off tasks rarely need a managed product. Repeated workflows do.

Ask:

• Who uses it today?
• How often?
• What happens if they cannot use it?
• Which manual work does it replace?

If nobody misses it when it is gone, it is probably not ready.

2. Does it touch sensitive data?

The moment customer, employee, financial, operational, or confidential company data enters the picture, the bar goes up.

Ask where the data lives, who can access it, and whether the current tool chain is acceptable for that data. This is where many prototypes quietly fail the enterprise test.

3. Who needs access?

Personal tools can get away with personal access. Team tools cannot.

If the prototype needs company users, roles, approvals, or SSO, it has crossed into real software territory. That connects directly to why enterprise operation is harder than starting the software.

4. What systems must it integrate with?

A prototype may work with pasted CSVs and manual exports. A product usually needs real integrations.

Ask:

• Which systems are sources of truth?
• Which updates must be written back?
• What happens if an integration fails?
• Who owns API changes?

If the answer is “we’ll just copy it manually”, that may be fine for a prototype. It is not a long-term operating model.

5. Who owns it?

Every team tool needs an owner. Not necessarily a full product department, but at least someone responsible for scope, quality, support, and decisions.

If ownership is unclear, the prototype is not ready. It may still be valuable. It just needs structure before more people depend on it.

6. Can it be operated for twelve months?

This is the big one.

Could this tool run for a year with updates, monitoring, support, and security fixes? If not, what is missing?

The answer does not have to be perfect. It just needs to be honest. And if the tool is already becoming important, read why your best prototype may already be business-critical.

The decision

After the checklist, you usually have one of three outcomes:

• Keep it personal.
• Improve it as a team workflow.
• Turn it into a managed product.

The goal is not to make every prototype bigger. The goal is to give the right prototypes the right level of seriousness.