What is an AI readiness assessment for a small business?
Leadership wants an AI initiative, several vendors are circling, and you need an independent read on whether the business is actually ready before committing to anything.
An AI readiness assessment scores whether your organisation can deploy AI credibly, before any tool is bought. It measures six dimensions: data quality and lineage, process determinism, privacy and compliance posture, failure-mode design, measurement baselines, and organisational capacity to absorb change. SMB pricing typically runs $1,900 to $9,900 depending on depth, against $15,000 to $200,000 at enterprise scale. The output should be a scored report and a priced roadmap you can execute with anyone.
That last condition is the one to hold a provider to. An assessment whose recommendations only work if the assessor implements them is a sales document with a score on it. A real one names the specific gaps, what it would cost to close each, and which are prerequisites rather than improvements — and remains useful if you take it to a different firm or do the work in-house. Ask to see a redacted example before commissioning one.
What an assessment actually measures
- Readiness is about process, not enthusiasm. A documented, consistently performed process with a reviewable output is a strong candidate. Judgement that lives in one experienced person's head is not — AI will imitate it confidently rather than reproduce it.
- Data lineage decides everything downstream. If nobody can say where a number came from or who may see it, no model built on it can be governed afterwards.
- Most failures are organisational. Pilots stall because no owner was named, success was never defined numerically, or the workflow around the model never changed — not because the model was wrong.
How to run an AI readiness assessment
-
01
Run the free four-minute readiness check on our site first. If it scores you low, that is the assessment finding and it cost nothing.
-
02
Pick the specific task you want automated, and write down how it is done today. If that cannot be written, that is the finding.
-
03
Ask any assessment vendor what a "not ready" conclusion looks like in their process. The answer is diagnostic.
-
04
Baseline the numbers you expect to improve before anything is built.
-
05
Start where the work is high-volume, low-variance and cheaply reversible — the first project's real job is teaching the organisation how to evaluate and govern this class of system.
The genuinely hard part is that a good assessment often concludes the answer is a data pipeline fix or a reporting definition rather than AI at all — cheaper, faster, and less exciting to present to a board. A vendor whose revenue depends on the next phase has an obvious incentive not to reach that conclusion, which is why the assessment is worth buying separately from the build.
More on AI Transformation
Still not sure this is your problem?
A 20-minute fit check. We will tell you if it is something you can fix without us — that happens often enough that we lead with it.
Book a Strategy Session →Browse every answer.
The full Solution Desk — specific questions, straight answers, no gate.
Open the Solution Desk →