What is an AI readiness assessment?
An AI readiness assessment measures how prepared a team is to get real value from AI. Not whether people are excited about it, and not whether the company has bought licences, but whether the work itself is in a shape where AI can help. Here is what one actually measures, what it should produce, and how to tell a real assessment from a sales process wearing its clothes.
The short answer
An AI readiness assessment looks at the work your team does, task by task, and scores each one on how ready it is for AI to take part. A task scores well when it is repetitive, well understood, and produces something a human can check quickly. It scores badly when it lives entirely in one person's head, changes every time, or carries a risk nobody wants to own.
The output is a score you can act on. Not a grade, and not a maturity label like "emerging" or "advanced", but a ranked list of your own workflows with a number against each and a first move attached.
What it measures
- Workflow suitability — which tasks are repetitive and well defined enough that AI can do a first pass a human then edits.
- Process maturity — whether the work is written down anywhere. A workflow that only exists in someone's head cannot be handed to anything, AI included.
- Data access — whether the information the task needs is somewhere reachable, or trapped in an inbox and three spreadsheets.
- People and habits — who is already using AI without being asked, and who will quietly opt out. Both matter more than the tooling.
- Guardrails — whether anyone has said out loud what is safe to paste into an AI tool and what always needs a human review.
What it should produce
A real assessment ends with three things. A score per workflow, so you can rank them. An estimate of hours recoverable, so you can argue for the time or budget to act. And a named first build, so the week after the assessment looks different from the week before.
If what you get back is a slide deck describing your industry and a maturity quadrant with your logo in it, you bought market research. The test is simple: can someone on your team start work on Monday from what the assessment handed you? If not, it did not finish its job. The full method is in the AI readiness assessment guide, and the scoring bands are in the 0-10 framework.
How long it should take
A self-assessment should take minutes. You are answering questions about work you already do; there is nothing to research. A consultant-led assessment for a small team runs one to two weeks, most of which is waiting for people to fill in a survey rather than analysis.
If someone proposes a multi-month readiness study before a single workflow improves, the length is the product. That is a sales process, and the assessment at the end of it will recommend more of the same.
Why the enterprise versions do not fit small teams
Most assessments on the market were built for enterprise IT. They score you on data platforms, model governance, and infrastructure you do not control and will never buy. A ten-person team scoring badly on "MLOps maturity" has learned nothing, because it was never going to build a model.
An assessment that fits a small team scores the work that team actually does. For an event team that means the run of show, the post-event survey, the sponsor prospectus, the budget, the RFP. Score those and the result tells you something you can use. See the event-team version for what that looks like in practice.
Frequently asked questions
- What is an AI readiness assessment?
- An AI readiness assessment measures how prepared a team is to get real value from AI: which of its workflows are suitable, whether the processes are documented enough to hand off, where the data lives, and whether people will actually adopt it. A good one produces a score per workflow tied to specific tasks and hours, not a generic maturity grade.
- What does an AI readiness assessment measure?
- Five things: workflow suitability, process maturity, data access, people and habits, and guardrails. The strongest signal is usually process maturity, because a workflow nobody has written down cannot be handed to anyone, AI included.
- How long does an AI readiness assessment take?
- A self-assessment takes minutes. Bring in a consultant and a small team’s assessment runs one to two weeks, most of it spent waiting on survey responses rather than analysing them. Any multi-month readiness study that improves nothing along the way is a sales process wearing an assessment’s clothes.
- Do I need an AI readiness assessment before using AI?
- No. If you already know which task is costing you the most hours, start there and skip the assessment. Assessments earn their place when a team is spread across many workflows and cannot agree on which one to fix first, or when you need a number to justify the time to someone else.
Want the assessment run on your team?
The AI Readiness Sprint scores your team's actual workflows and ends with a ranked list and a first build, not a maturity grade.
See how it works