Kaksio Labs
AI Maturity Index
Nine questions about whether your organisation is ready to run AI on real customer data. Not about how many people have tried ChatGPT. Two minutes, no contact details.
The result is calculated in your browser. We do not store your answers unless you ask for a comment.
Request a comment
I will review your result and recommend where to start
- I reply personally, you will not get an automated message
- I will tell you what is most likely the right first step in your case
- I will say plainly if this is not worth touching yet
Free of charge. We use your address only to reply, we will not add you to a newsletter list. We process your data in accordance with our privacy policy.
Thank you. I have your result now.
Antti Nevalainen
Head of AI, Kaksio Labs
What are the benchmark lines based on?
The vertical lines on the bars show the typical readiness level reported in international studies. They are not Kaksio Labs' own data, nor an industry-specific peer group.
- Deloitte, State of AI in the Enterprise 2026. 3 235 business and IT leaders from 24 countries. Readiness by dimension: strategy 42 %, data management 40 %, technical infrastructure 43 %, governance 30 %, skills 20 %. Only 21 % report a mature governance model for autonomous agents.
- McKinsey, The State of AI 2025. 88 % of organisations use AI in at least one function, but almost two thirds are still piloting. Thorough redesign of workflows is rare outside the top performers.
- MIT NANDA, The GenAI Divide 2025. 95 % of generative AI pilots produced no measurable effect on results.
The figures are international and weighted towards large organisations. In a small or mid-sized company the starting point is typically lower, so crossing the line does not mean you are ready.
background taustaa
AI Maturity and What the Test Measures
What is AI maturity?
AI maturity indicates how ready an organization is to use artificial intelligence in real-world business operations. It does not measure how many people have tried ChatGPT.
Maturity determines whether a pilot project moves into production or remains just a demonstration. The most common reason for failure isn’t the technology. It’s data that machines can’t read and processes that no one has documented.
Six Areas
The test scores these six areas. Strategy and data carry the most weight.
Strategy. Has a person been designated to be responsible for AI, and are there measurable goals? Without ownership, experiments do not build on each other. Everyone starts from scratch.
Data and systems. Is customer data centralized, organized, and accessible via APIs? AI is only as good as the data it has access to.
Processes. Are recurring workflows documented in a way that allows for automation? An undocumented process leads to misunderstandings in an AI project.
Expertise. To what extent do employees use AI in their work? Are best practices shared across teams, or do they remain the sole knowledge of a single person?
Production and Impact. Is AI used in production at the customer interface? Is its impact measured in hours or euros?
Governance. Do employees know what data they’re allowed to feed into the tools? The EU’s AI Regulation will make this a requirement in the coming years.
Five maturity levels
The result falls into one of these five levels.
1. Observer. Experiments are conducted, but no one takes ownership of the big picture.
2. Experimenter. Pilots are launched quickly, but they don’t make it into everyday operations.
3. Builder. The first production deployment exists. The next bottleneck is reproducibility.
4. Scaler. Multiple use cases and measurable impact. The issue is no longer technology but management.
5. Agent Organization. Agents manage processes independently. Areas for improvement lie at the periphery: monitoring and exception handling.
Frequently Asked Questions
How long does the test take?
About two minutes. Nine questions, no contact information.
What are the benchmark figures based on?
Deloitte’s “State of AI in the Enterprise 2026” survey, which includes 3,235 executives from 24 countries, and McKinsey’s “The State of AI 2025” report. The figures are international and focus on large organizations. In a Finnish small or medium-sized enterprise, the starting point is typically lower, so exceeding the benchmark does not mean you are ready.
Will my answers be saved?
The result is calculated in the browser and isn’t saved anywhere. Your email address is only requested at the point where you can ask for feedback on your results.
Where should I start if my score is low?
With your weakest area. Most often, that’s data or ownership—not the tool itself. A new tool built on top of weak data will quickly produce impressive demos but eventually lead to disappointment.
What happens after the test?
You’ll see your results on the screen right away. If you leave your email address, I’ll review your results and write down my thoughts. I’ll respond personally, so you won’t receive a reply on the same day.