AI readiness starts with a clear piece of work and the conditions needed to do it responsibly. These six questions help you decide what to investigate before buying another tool or asking employees to change how they work.
Use them as a discussion guide. Record what you know, what is uncertain and who will find the missing information. A numerical score is less useful than an honest explanation.
1. What work are you trying to improve?
Describe one task, the person who does it and the result they need. “Prepare a first draft of our weekly operations summary” gives you something to evaluate. “Use more AI” does not.
Identify how you will judge the result: accuracy, completeness, time spent reviewing or another measure relevant to that task. Capture how the work happens today so you have a fair comparison later.
2. Is the source information ready?
Where does the information live, who owns it and which version should be used? Start with a small, approved set. List missing files and inconsistent formats before treating them as implementation details.
If people cannot reliably find the current document, an AI tool needs a clearer starting point too.
3. Which systems would the workflow depend on?
List the accounts, applications and connections it would need. Identify the person responsible for each and ask what access is available for the proposed task.
Review the Microsoft 365 and cloud environment supporting the work. A new workflow still depends on ordinary account administration, permissions and reliable access.
4. Who will use and review the output?
Choose a business owner and an experienced reviewer. Ask what employees would need to understand before using the workflow, including when to stop and ask for help.
Give the reviewer the source information needed to check the result. Include their review effort when evaluating whether the workflow is useful.
5. What rules apply to the work?
Write down which information may be used, which tool and account are approved, and whether output must be checked before it reaches a customer or business system. Keep the rules close to the actual task so employees can apply them. AI Governance explains how these decisions fit together.
6. Who will look after it once it is working?
Decide who handles a failed run, changes the source material, reviews results and approves a change in purpose.
Also define a manual way to complete the work. Before expanding the workflow, review a representative set of outputs with the people who know the task. A useful first demonstration is a reason to investigate further, not proof that every variation is ready.
Frequently asked questions
Do we need perfect data before starting?
No. You need suitable information for the proposed task and a clear understanding of its limits. Narrowing the first use case may be more practical than reorganizing everything at once.
Does this checklist replace an AI Adoption Review?
No. Use the checklist to prepare for a conversation about your environment. ProIncident's AI Adoption Review is a step toward or within an ongoing managed relationship, with deliverables agreed for the engagement.
Should AI come before basic IT improvements?
Address dependencies that would prevent the work from being used responsibly. ProIncident leads with Fully Managed IT, with AI Adoption available when the business is ready to go further.
Choose a useful starting point
Bring one business task and the questions you could not answer. We can discuss a practical starting point within an ongoing managed relationship.