Test · 2 minutes · nothing is sent
Take the self-assessment
Ten questions, two for each area. Answer with what happens today, not what you'd like. The result appears right here.
10 questions to go
The test above asks you 10 questions, two for each of the five things that decide whether a company can get value from artificial intelligence: processes, data, systems, team and management. Each answer is worth 0 to 3 points. The maximum is 30. With your score it places you at one of three levels: Starting, Under way or Ready.
Here we explain what each question measures, how to read your result and what to do next. If you have not taken the test yet, do it first: this guide is easier to follow with your number in front of you.
Why measure maturity before putting AI in place?
Because AI works with what you already have. If your processes are not clear, it automates the mess. If your data is spread across ten places, it cannot find what it needs. If nobody in the company has time for the project, it ends up half done.
The OECD sums it up in a December 2025 paper on AI in SMEs: a certain level of digital maturity is usually a prerequisite for adopting AI successfully, and at the same time well-implemented AI makes the company more mature. It is a circle, and the assessment tells you where on the circle you are.
It is also the first step the Government foresees. The IA360 Plan, presented by La Moncloa on 21 September 2026, says that rolling out the Artificial Intelligence Voucher (Bono de inteligencia artificial) will include a "prior maturity diagnosis, identification of a specific use case and subsequent measurement of the impact on productivity". First the diagnosis, then a specific case, then measure. In that order.
How many companies already use AI?
Fewer than it seems. According to Spain's National Statistics Institute (INE), 21.1% of Spanish companies with 10 or more employees used artificial intelligence in the first quarter of 2025. Among those with fewer than 10 employees, 13.4%. For the European Union as a whole, Eurostat puts the figure at 20% for 2025.
The IA360 Plan sets a target of going from that 21.1% to 55% of companies with AI in 2030. And it points to a mismatch: 44.2% of the working-age population already uses generative AI. Your team probably uses it more than your company does.
If your result is not high, you are in the majority. What is useful about the test is not the score, but knowing which dimension is holding you back.
What does each dimension measure?
Processes
Whether your company's tasks are defined and repeat. AI performs where there is work done many times and more or less the same way: orders, invoices, appointments, quotes, replies to customers. If each person does the same task their own way, or if only the person doing it knows the process, it has to be written down first.
Data
Whether your information is digitised, organised and accessible. You do not need "big data". You need customers, orders or delivery notes to be in digital format, with a common standard, and for more than one person to know where they are. We go into it in how to prepare your company's data for AI.
Systems
Which programs you use and whether they connect to each other. A company that moves data by hand from email to a spreadsheet and from there to the invoicing software has a systems problem before an AI one. The OECD notes that AI is adopted more in companies that already use cloud services: technologies build on each other.
Team
The current use of AI among your staff and their willingness to learn. Here both count: who already uses AI tools at work and the attitude to change. According to Eurostat, among European companies that considered using AI and did not, the most cited reason was lack of knowledge (70.9%), ahead of legal doubts (52.5%) and privacy (48.8%). The OECD agrees: a shortage of skills is one of the barriers SMEs mention most. More in how to train your team in AI.
Management
Whether there is a clear goal, someone in charge and a time budget. It is the most underestimated dimension. "We want to use AI" is not a goal. "We want to stop typing supplier delivery notes by hand" is. And you need a person, with a name and surname, to push it, with hours reserved in their week to do it.
What does your level mean?
| Dimension | Starting (0-10) | Under way (11-20) | Ready (21-30) |
|---|---|---|---|
| Processes | Everyone works their own way. Nothing is written down. | Key processes are known, but not fully documented. | Repeating processes are written down and followed. |
| Data | On paper, in loose emails or in someone's head. | Digitised, but scattered and with no common standard. | Digital, organised and accessible to whoever needs it. |
| Systems | Spreadsheets and a lot of copy and paste. | Management software, but each one goes its own way. | The main programs share data with each other. |
| Team | Nobody uses AI at work, or it is used in secret. | Some people use it on their own, with no common standard. | Regular use, with clear guidelines and appetite for more. |
| Management | Interest, but no goal or owner. | There is a goal, but no time assigned. | A specific goal, an owner and reserved hours. |
The table is a guide. It is normal to be at different levels depending on the dimension: a company can have very clear processes and its data on paper. Look at your score by dimension, not just the total.
Starting level (0-10 points): what do I do first?
Your company works, but it works on people's memory. Before thinking about tools, there is tidying up to do. The good news is that it costs no money and you can start this week.
- Write down the three processes that repeat most. Step by step, as they are done today, not as they should be done. Ask whoever does them to review it.
- Gather the basic data in one place. Customers, suppliers, orders. A shared spreadsheet with fixed columns is already real progress.
- Name an owner. A person who reserves a few hours a month for this. Without that name, the other two steps stay as intentions.
When you have that, repeat the test. If you want to understand what a project involves afterwards, read how to set up your first AI project and, to see the whole path, the 5 phases for putting AI into your business.
Under way level (11-20 points): what do I do first?
You have the foundations, but they are loose. It is the most common level and the most rewarding: with a little connecting work, a first project already has something to stand on.
- Choose a single process to start with. The one that repeats most and takes the most time. Not a general plan with ten fronts: one, well chosen.
- Look at what data that process needs and where it comes from. If it passes through three programs by hand, connecting them may be the first improvement, even before bringing in AI.
- Set guidelines for the use your team already makes. Which tools can be used, with which data and with which not. That way what is free-for-all use today becomes learning for the company.
Before you launch, work out what you expect to gain. In how to measure the profitability of an AI project we explain how to do it without inventing figures.
Ready level (21-30 points): what do I do first?
Your company has clear processes, data in order and someone in charge. You are in a position to put AI into a real process, not just try it out.
- Define the use case with a metric. "Reduce the response time for quotes" and how long it takes today. Without the starting figure you will not know whether it worked.
- Run a limited pilot. One process, one team, one deadline. Measure before and after, and decide with data whether to extend it.
- Review the grants available. A prepared company is the one that makes the best use of a call when it opens. You have the overview in grants for putting AI into businesses in 2026.
A high score does not guarantee the project will go well. It guarantees that, if it goes badly, it will not be for lack of a foundation.
What does this have to do with the AI Voucher?
The artificial intelligence voucher of the IA360 Plan cannot be applied for yet: the plan foresees a pilot in the first half of 2027 and the general call before the end of 2027. But the document already makes the order of work clear: maturity diagnosis, specific use case and impact measurement.
That has two consequences for you:
- The diagnosis does not depend on any call. You can do it today, and what you discover is useful with or without the voucher.
- The test on this page is not that diagnosis. It is not known what format the rules will require or who will be able to carry it out. Take it as a guide to know where you stand.
Everything that is known about the voucher, and what is not, is in our guide to the Artificial Intelligence Voucher.
How does it differ from a diagnosis done by a third party?
The test captures what you believe is happening in your company. And whoever runs a business usually has a different idea of the processes from whoever does them every day. It is not a flaw: they are seen from different places.
A diagnosis done by someone from outside goes further:
- Interviews with the people who do the work, not just with management.
- Review of the processes at your workstation, seeing how they are really done, with their shortcuts and patches.
- A look at the programs and the data: what there is, where it is, what is connected and what is copied by hand.
- An actionable result: the processes where AI can save most, in priority order, and a first specific use case.
At Xora we do this diagnosis. We go to the company, talk to the team and look at the processes where they happen. We explain how we approach it in our business audit. If you prefer to talk it over first, write to us with your test result and we will go through it together.
Sources
Xora IntelligenceAI consultancy and custom software in Los Barrios (Cádiz). We write about what we build for businesses in the province. If you have a question about your case, write to us.




