An AI readiness assessment is a structured one-to-two-week engagement that evaluates a company's processes, data quality and digital maturity to identify precisely where artificial intelligence will deliver the most business value — and where it is not yet worth the investment. The output is not a deck of ideas: it is a prioritised opportunity map and a concrete roadmap with timelines, estimated costs and named owners.
It is the first engagement César García Cabeza, senior freelance AI consultant based in Andorra, recommends to any SME that wants to adopt AI without getting the first move wrong.
Why start with an assessment rather than a tool?
Because most AI projects fail due to the order in which decisions are made, not because the technology does not work.
According to S&P Global Market Intelligence, 42 % of companies abandoned their main AI initiatives in 2025, up from 17 % in 2024 — a 147 % increase in one year. Gartner warns that 60 % of AI projects will be abandoned before 2026 if organisations do not have AI-ready data. The root causes: poor data quality (43 % of cases), insufficient technical maturity (43 %) and a shortage of skilled people (35 %).
The pattern is always the same: a company spots an attractive tool or model, runs a pilot, the pilot does not connect to a real business process and the project is dead within three months. That is not a budget problem or an ambition problem — it is a sequencing problem.
A prior assessment breaks that cycle. Industry data shows that a maturity evaluation before the first project reduces implementation costs by 30–40 %, by eliminating false starts before they happen.
The right question is not "which AI tool do I need?" — it is "which business process costs me the most time or money, and do I have the data to address it?" The assessment answers that before you open your wallet.
What does an AI readiness assessment actually cover?
The assessment evaluates the company across several dimensions: business strategy alignment, data quality and accessibility, technology infrastructure, team capabilities, governance and regulatory compliance — including alignment with the EU AI Act, whose AI literacy obligations have applied across the EU since February 2025.
The process has three phases:
- Interviews and context gathering. Working sessions with the people who own key processes to understand how the business actually runs, what data exists and where the real bottlenecks are.
- Analysis and prioritisation. Candidate use cases are identified and scored by impact, confidence and ease of implementation — a variant of the ICE model (Impact, Confidence, Ease) used by organisations including Google, PwC and Capgemini to rank AI initiatives.
- Report delivery. An actionable document with findings and next steps — not an 80-page PDF that sits unread.
What does the assessment deliver?
César García's assessment produces four concrete outputs:
- Opportunity map. All viable AI use cases for your business, plotted on an impact-versus-effort matrix. High impact, low effort cases are quick wins to tackle first. High impact, high effort cases are strategic bets for the next planning cycle. Low impact, high effort cases are eliminated — and knowing what not to build is equally valuable.
- Technology stack audit. A clear picture of your current systems, how they connect and what needs to be addressed before AI can be integrated reliably.
- 90-day roadmap. First steps ordered by priority, with realistic timelines, cost estimates and clear ownership for each action.
- Live review session. A working meeting to walk through the findings, answer questions and align the team before deciding on the next move.
What the assessment does not deliver: installed technology, code in production or commitments beyond the agreed scope. That is deliberate — the point is to understand clearly before building anything.
How much does it cost and how long does it take?
César García's AI assessment is priced between 1,000 € and 2,000 €, depending on company size and complexity. It takes one to two weeks from kickoff to final report delivery.
If you proceed to a build project after the assessment — whether that is back-office automation, a private AI assistant with Enclave, or another system — the assessment fee is deducted from the project cost. You are not paying for the map and then starting again from scratch.
For context, market benchmarks put SME AI maturity evaluations at between 2,000 and 8,000 USD internationally. César's pricing sits deliberately at the lower end of the European range, because the goal is to lower the barrier to entry — not to turn the assessment itself into a large engagement.
How does César García run this with an Andorra-based SME?
Andorra has a particular business landscape: agile small businesses with data spread across several systems, small teams without dedicated technical staff and, in many cases, compliance obligations on both the Andorran and European sides of the border.
César García Cabeza has worked with this type of company for years — from professional services firms and law offices to real estate businesses and fintechs — and the assessment is designed around that reality, not around a large corporation's needs. It is not a generic questionnaire: it is a collaborative process where César sits down with the business owners, understands the real workflows and prioritises based on what is actually achievable with the resources available.
The Andorran context brings one more relevant factor: the Govern d'Andorra and Microsoft signed a five-year AI and digital transformation partnership in January 2025. The PDE programme (Programa de Digitalització d'Empreses) offers grants of up to 6,000 € for microenterprises. Despite this, the AI and data analytics lines within PDE remain the least applied-for — there is unmet demand. The assessment can also be the starting point for identifying which projects are eligible for this kind of support.
For businesses handling sensitive data — client files, contracts, case documents — the assessment evaluates whether deploying models on private infrastructure makes sense, including options like Enclave, the private enterprise AI assistant that keeps your data off third-party servers.
For a broader view of the adoption journey, see AI for SMEs in Andorra: where to start or the guide on how to choose the right AI consultant in Andorra.
In summary
An AI readiness assessment is not a bureaucratic step or a nice-to-have: it is the only reliable way to know what to build, in what order and with what budget before committing resources. It costs between 1,000 € and 2,000 €, takes one to two weeks, and the fee is deducted from the project if you move forward.
Forty-two per cent of companies that started by buying tools have already abandoned their AI initiatives. The ones that started by understanding the problem have systems running in production.
Ready to figure out where your business stands?