Perimattic
Free tool

AI readiness assessment

Answer five questions about your data, team, processes and leadership. Get a readiness score out of 25, a readiness level, and the first gap to close before you invest in AI.

Overhead view of a team working on laptops around a wooden table
AI readiness is how well prepared an organisation is to adopt AI and keep it running. This assessment asks five questions, one each on data infrastructure, data quality, technical talent, process maturity and leadership buy-in, and scores each from 1 to 5. The total out of 25 becomes a percentage: 80% or more is “AI Ready”, 60–79% “Promising”, 40–59% “Developing” and below 40% “Early Stage”. It also names your lowest-scoring dimension and the first thing to do about it.

Assessment

AI Readiness Assessment

1. Data Infrastructure

How centralized and accessible is your data?

2. Data Quality

How clean and consistent is your data?

3. Technical Talent

What AI/ML expertise does your team have?

4. Process Maturity

How standardized are the processes you want to automate?

5. Leadership Buy-in

How committed is leadership to AI investment?

How scoring works

Each dimension is scored 1-5. Your overall readiness is total / 25 × 100.
≥ 80% AI Ready • 60-79% Promising • 40-59% Developing • < 40% Early Stage

Results

Answer all five questions, then click “Calculate Results” to see your AI readiness score.

Method

How the AI readiness score works

The assessment covers five dimensions. For each, pick the answer that describes where you are today, not where you plan to be:

  • Data infrastructure: from siloed and manual (1) to a centralised data lake or warehouse (5).
  • Data quality: from significant issues (1) to high quality with governance (5).
  • Technical talent: from no AI/ML experience (1) to a dedicated AI/ML team (5).
  • Process maturity: from ad-hoc (1) to standardised and measured (5).
  • Leadership buy-in: from sceptical (1) to fully committed with budget (5).

Readiness % = (sum of the five scores ÷ 25) × 100. The level follows fixed bands: 80% and above is AI Ready, 60–79% Promising, 40–59% Developing, and under 40% Early Stage.

Team meeting around a conference table with laptops open

Example

Worked example: a mid-sized operations team

An operations team has a mostly centralised warehouse (4), some inconsistencies in its data (3), basic AI awareness but no data scientists (2), processes that are documented but vary between people (3), and leaders who are supportive with some budget (4).

Scores 4, 3, 2, 3, 4

  1. Total = 4 + 3 + 2 + 3 + 4 = 16 out of 25
  2. Readiness = 16 ÷ 25 × 100 = 64%
  3. Level = Promising (60–79%), some gaps to address
  4. Weakest dimension = Technical talent (2)
  5. Next step = hire or upskill data science / ML engineering talent

How to interpret the result

The percentage tells you roughly where you stand; the weakest dimension tells you what to do next. Two organisations can both score 64%, one with even 3s and one with a 1 in data quality. The second has a harder problem, because poor data undermines every use case. Look at the per-dimension scores, not only the total. Where two dimensions tie for lowest, the tool names the first in the list, so treat both as priorities.

Common mistakes

  • Answering for the best-run team in the company rather than the one that will own the AI system.
  • Scoring planned investments as if they were already in place.
  • Having one person answer alone. Ask data, engineering and business leads separately and compare.
  • Treating a high score as approval for any project. Readiness says you can deliver; it does not say a given use case is worth it.

When to use a related tool

Once you have a use case in mind, estimate its build in the AI agent complexity estimator and its return in the AI ROI calculator. If you work in a regulated sector, the AI compliance hub and the AI observability guide explain what evidence you will need to keep once AI is in production.

FAQ

Frequently asked questions

What makes a company AI ready?

A company is AI ready when its data, people, processes and leadership can support an AI system from pilot to production. In practice that means data that is centralised and clean enough to use, people who can build and run models, processes documented well enough to automate, and leaders who will fund and sponsor the work.

How is the AI readiness score calculated?

Each of the five questions is scored from 1 (the first answer) to 5 (the last). The scores are added to a total out of 25 and converted to a percentage. The level comes from fixed bands: 80% and above, 60–79%, 40–59%, and below 40%. The weakest dimension is the one with the lowest score.

What data requirements must be met before starting an AI project?

You need relevant data you are allowed to use, in a form a system can reach. That means knowing where the data lives, having consistent formats and documented quality, putting access controls and privacy obligations in place, and having a way to deliver fresh data to the model once it is in production.

What team skills are needed for AI adoption?

You need people who can move data, build and evaluate models or prompts, run them in production, and judge whether the output is right. That usually means data engineering, ML or AI engineering, platform or operations skills, a product owner who sets success measures, and domain experts who review results.

What should I do if one dimension scores much lower than the rest?

Fix the lowest dimension before scaling, because it limits everything else. The assessment gives a first step for each: centralise data, set up data governance, hire or upskill ML talent, document and standardise the target process, or build a business case with quick wins to win leadership support.

Where should a company start its AI journey?

Start with one well-understood problem that uses data you already collect and has a result you can measure. Document the current process and its cost first, so you can show the change. Use that first project to build skills and evidence, then use the AI ROI calculator to compare the next candidates.

Perimattic AI Suite

Turn the estimate into measured numbers

A calculator gives you a planning figure. Perimattic AI Suite traces every model call in production with its tokens, cost, latency and quality score, so you can check the estimate against what your AI systems actually do.
  • Free consultation
  • No commitment
  • Tailored to your team