What it is
If someone hands you a maturity score, read it as a description of where you stand, and look for results elsewhere. Maturity models place a company on a ladder, often with five levels, and promise that climbing it pays off. Consultancies, vendors and researchers all publish them.
What the evidence shows
The case against relying on the score
- A review of a decade of studies on digital maturity models reports that researchers question the models’ value. The models have no objective rules for scoring, and nobody has data linking a maturity level to performance.
- Many models mainly record what senior managers think, not measurable outcomes.
- A review of 47 studies found that the link between readiness for change and a successful rollout is unclear. Studies measure readiness in many different ways.
- A review of published studies on AI maturity models found that 6 of 13 studies (46%) mainly assess technology (Sadiq and others, in the sources below).
- Going deeper did not go with better hiring plans. The Federal Reserve Bank of New York analysed a 2025 survey of US small firms with employees. Among firms that used AI, those that had built it into their processes did not expect to hire more than those still experimenting. In payroll records analysed by Gusto, which sells payroll software to small firms, firms that used AI for everyday tasks added staff at the same rate as firms that had built it deep into their operations. Both sources are in the list below.
The case for
- A survey of 143 senior executives found that companies with stronger AI capability reported more creativity and better performance. In that survey, the same executives rated both the capability and the performance.
- In an OECD survey of 2,018 small and medium-sized firms in 12 countries, 2% of firms that used AI for isolated tasks reported a transformational effect. Among firms that used it across the business, 23% did. The firms rated the effect themselves. The OECD says the sample is not representative: the firms were reached through digital platforms that work with the OECD, and three-quarters of the answers on AI came from Japan.
Nobody has shown that a maturity score forecasts returns.
How to do it
- Use a maturity assessment to structure a conversation: where are we weak, and do we agree on that?
- Look at the answers to single questions, not the overall level.
- Check who filled it in. If only executives did, get other views. See assess readiness from several sources.
- Do not set “reach level 4” as a goal. Set goals in business terms: time, quality, cost, revenue.
- Do not use a score to rank business units or decide budgets.
- Be wary when a comparison with other companies comes from the firm selling the remedy.
What to measure
- Outcomes of actual AI projects: time saved, quality, cost. See measure against a baseline.
- If you repeat a maturity assessment, how far the views of different groups move closer together.
Where it fails
- Throwing the model away. A structured checklist still helps a leadership team see gaps. The warning is about what the score can prove.
- Regulated settings. Some audits and standards require a documented assessment. Do it, and keep it apart from your business case.
By company size
- Medium: most published models were built for large firms and may not fit. A short checklist is enough.
- Enterprise: maturity scores often end up in board reports. Report project outcomes next to them.
Assessments for this practice
These are existing questionnaires, tests and guides you can use. Each card gives the question it answers, who answers it and whether you can use it freely. All assessments.
- Does the company have what it needs to use AI? AI Capability scale Managers, about their own company · 46 questions Free to use and adapt
- Are staff committed to this change, and do they believe it can be done? Organizational Readiness for Implementing Change (ORIC) Staff, about their organization or team · 12 questions Not established
- Are a public body and its people both ready for AI? Dual-level model of AI readiness in the public sector Managers and staff, in a discussion · 4 groups of criteria. The number of questions is not established Free to use and adapt
Sources
- Thordsen and Bick (2023). A decade of digital maturity models: much ado about nothing?. Information Systems and e-Business Management, 21(4), 947–976. What this study did and found
- Silva, São Mamede and Santos (2024). A new proposed model to assess the digital organizational readiness to maximize the results of the digital transformation in SMEs. Journal of Innovation & Knowledge, 10(1), 100644. What this study did and found
- Caci, Nyantakyi, Blum, Sonpar, Schultes, Albers and Clack (2025). Organizational readiness for change: A systematic review of the healthcare literature. Implementation Research and Practice, 6. What this study did and found
- Mikalef and Gupta (2021). Artificial intelligence capability: Conceptualization, measurement calibration, and empirical study on its impact on organizational creativity and firm performance. Information & Management, 58(3), 103434. Limits or contradicts this practice. What this study did and found
- Sadiq, Safie, Abd Rahman and Goudarzi (2021). Artificial intelligence maturity model: a systematic literature review. PeerJ Computer Science. doi.org
- Aarons and Sarkar (October 2026). AI Adoption and Employment Expectations: Evidence from a Survey of Small Business Owners. Federal Reserve Bank of New York, Liberty Street Economics. libertystreeteconomics.newyorkfed.org
- Tremper and Bowen (2026). Analysis of AI adoption and hiring at small businesses, from payroll records. Gusto, which sells payroll software to small firms. gusto.com
- Bianchini, Lasheras Sancho and others (April 2026). Empowering SMEs in the age of AI: The 2026 OECD D4SME Survey. OECD. oecd.org
Claims on this topic, checked against these studies:
- “BCG’s AI Maturity Matrix shows how mature companies are” Needs correcting
- “Microsoft has a maturity model for responsible AI” Confirmed
- “Only 13% of companies are ready for AI” Needs correcting
Related practices
- Assess readiness from several sources, not one executive’s rating A readiness check filled in by one senior person reflects that person’s view. Ask staff and managers as well, and compare the answers with what you can observe. Do next · Effort: medium (weeks, a few people)
- Measure readiness and resistance separately Staff can be keen on AI and worried about it at once. Ask about the two separately. A single “readiness” score hides the worry. Do next · Effort: medium (weeks, a few people)
- Measure time saved per task against a baseline Record how long a task takes and how good the result is before you add AI. Without that starting figure you cannot tell a real gain from a hopeful one. Do first · Effort: low (days, one person)