AI Readiness Checklist
The 12 questions every leader should answer before greenlighting an AI initiative—from data maturity to change capacity.
How to use this checklist
AI readiness is not a property of the technology. It is the organization’s ability to make a sound decision, execute it safely, and turn it into adopted behaviour that creates measurable value.
This checklist is designed for one proposed initiative—not for rating the organization in the abstract. Choose a real candidate, answer from evidence rather than optimism, and involve the people who own the workflow and will live with the change.
For each question, select:
- Yes — supported by current, specific evidence;
- Partly — plausible, but incomplete or dependent on an untested assumption;
- No — absent, unknown, disputed, or based mainly on hope.
A low score is not a failure. It is an early warning that protects capital before commitment.
1. Is the business problem specific, consequential, and observed?
Can you describe the problem without mentioning AI, a vendor, or a preferred solution? A decision-grade problem identifies who experiences it, where in the workflow it occurs, how often it happens, and what business consequence follows.
Evidence of readiness: workflow observations, customer or employee evidence, incident records, cycle-time data, error rates, lost-margin examples, or another credible baseline.
Warning sign: “We need to use AI” is presented as the problem.
2. Is there a named decision owner with authority to act?
Every initiative needs one person accountable for the decision—not only a project coordinator or technical lead. That owner must be able to approve the next step, resolve trade-offs, allocate resources, and stop the initiative when evidence does not support it.
Evidence of readiness: a named executive or business owner, a clear decision date, an identified decision forum, and explicit authority.
Warning sign: the initiative belongs collectively to a committee, innovation team, or vendor.
3. Have the realistic options been compared—including doing nothing?
AI should compete against other responses to the problem: simplifying the workflow, clarifying ownership, improving existing software, conventional automation, buying a service, partnering, waiting, or doing nothing.
Evidence of readiness: a documented option set with cost, benefit, risk, time, and reversibility considered on comparable terms.
Warning sign: the organization began by selecting a tool and is now constructing the business case around it.
4. Is the current workflow understood end to end?
AI inserted into a poorly understood workflow often accelerates confusion. Leaders need to know the real sequence of work, handoffs, exceptions, decisions, workarounds, failure points, and difference between the documented process and daily practice.
Evidence of readiness: a current-state workflow map validated by the people who perform the work, including exceptions and manual steps.
Warning sign: only managers or system diagrams have been consulted.
5. Is the required data available, usable, and lawfully accessible?
Data readiness is more than having files. The initiative must identify the inputs it needs, their source, quality, completeness, ownership, access conditions, update frequency, retention, representativeness, and permissible use.
Evidence of readiness: representative samples have been inspected; gaps and bias are documented; access is confirmed; and a responsible data owner is named.
Warning sign: the team assumes that years of data automatically mean usable data.
6. Can the initiative fit the existing technical and operational environment?
A useful prototype still fails if it cannot connect safely to the systems and routines around it. Consider integration, identity and access, security, latency, reliability, support, exception handling, and what happens when the AI is unavailable or wrong.
Evidence of readiness: technical dependencies, operational controls, fallback procedures, and support ownership have been mapped at the depth appropriate for the next decision.
Warning sign: integration and operating costs are deferred until after the pilot.
7. Have the material risks and non-negotiable constraints been mapped?
Risk is not limited to cybersecurity. A defensible review covers strategic, financial, data, technical, workflow, governance and compliance, vendor dependency, and human adoption risks.
Evidence of readiness: a risk register names likelihood, impact, mitigation, owner, evidence required, and stop or escalation conditions.
Warning sign: the risk section consists of “legal and IT will review later.”
8. Are accountability, oversight, and escalation clear?
Someone must own system performance, data quality, human review, incidents, vendor management, and decisions about changing or retiring the solution. The level of governance should match the consequence of error.
Evidence of readiness: named owners, review rights, escalation paths, auditability, monitoring expectations, and a defined operating cadence.
Warning sign: responsibility becomes unclear where the vendor’s platform ends and the client’s workflow begins.
9. Is the next step proportional, reversible, and gated by evidence?
Readiness does not mean committing to full implementation. It means choosing the smallest credible next step that can resolve the most important uncertainty without exposing the organization to disproportionate cost or risk.
Evidence of readiness: a bounded validation plan, budget and time box, decision gates, success and stop criteria, and an explicit path to pause, revise, or reverse.
Warning sign: a pilot is approved with no decision it must inform.
10. Have the affected people shaped the decision?
The people who perform, supervise, receive, or depend on the work often see risks invisible to the project team. Their involvement should influence the solution and may legitimately simplify, delay, or veto the preferred option.
Evidence of readiness: representative frontline users and adjacent stakeholders have been heard; their concerns are documented; and feedback changed at least part of the design or plan.
Warning sign: users will first see the initiative during training or launch communications.
11. Can the organization absorb this change now?
Even a strong initiative can fail in an overloaded organization. Readiness depends on available attention, capability, leadership bandwidth, process stability, competing changes, training needs, and the credibility created—or damaged—by prior initiatives.
Evidence of readiness: affected teams have protected time, named support, realistic capability plans, and a change load they can absorb.
Warning sign: adoption work is added to existing responsibilities without removing or rescheduling anything.
12. Are success, adoption, and value measurable from a baseline?
The organization should know what observable signals would prove that the initiative works and that people use it correctly. Technical performance alone is insufficient; measurement must connect the system to workflow behaviour and business consequence.
Evidence of readiness: a baseline, target, measurement source, owner, review date, adoption signal, and financial or operational outcome are defined before execution.
Warning sign: success is described as “increased efficiency,” “better experience,” or launch completion without a measurable definition.
Interpret your result
This checklist is a decision aid, not a certification. A high total cannot cancel a serious “No” on safety, legality, data access, decision ownership, or the ability to measure value.
| Score | Readiness posture | Recommended response |
|---|---|---|
| 0–8 | Not ready to commit | Pause procurement or build activity. Clarify the problem, owner, workflow, and evidence first. |
| 9–16 | Promising but under-evidenced | Run focused discovery or an AI Clarity Workshop. Resolve the most consequential unknowns before piloting. |
| 17–20 | Ready for controlled validation | Design a small, reversible validation with explicit gates, owners, risk controls, and outcome measures. |
| 21–24 | Strong decision readiness | Proceed to controlled execution—without treating the score as permission to skip diligence, governance, or adoption design. |
Five answers that should stop automatic progression
Regardless of total score, do not approve material commitment while any of these remain a clear “No”:
- a specific and consequential business problem;
- an accountable decision owner;
- lawful and credible access to required data;
- material risks and constraints mapped; or
- measurable success and adoption signals.
What to do next
Mostly “No”
Do not start with vendors. Name the problem and owner, observe the workflow, establish a baseline, and list the evidence required to make the next decision.
Mostly “Partly”
Convert assumptions into a short evidence plan. Assign each unknown an owner and deadline. Use the findings to decide whether to stop, defer, investigate, or validate.
Mostly “Yes”
Challenge the weakest answers and the most expensive assumption. Then authorize only the next proportional step, with a decision gate before further capital is released.
Readiness is not confidence that AI will work. It is confidence that your organization can learn quickly, decide responsibly, and stop when the evidence says stop.
About morpho360
morpho360 helps leadership teams make and execute complex technology-enabled transformation decisions with greater clarity, commercial discipline, and adoption confidence.
The morpho360 Method™ connects decision quality with human adoption through six steps: clarify the real problem, translate complexity, isolate decisions and trade-offs, build a proportional roadmap, enable ground-level adoption, and establish continuous measurement and feedback loops.
Suggested call to action: If your answers reveal competing opportunities or important unknowns, start with an AI Clarity Workshop or commission an AI Opportunity Audit.