A low outcome on the maturity measurement raises the same question for most executive boards: how bad is this. The answer is that the question itself comes too early. A low number on one of the seven dimensions can mean two scenarios that have little to do with each other, and the instrument alone does not tell you which of the two applies.
The first scenario is the organization that is at the beginning. There is no data management that is sustained, there is no infrastructure that can carry AI applications, and the organization itself has not yet made a habit of working with AI systems. In this scenario, the low score is an accurate description of the situation. There is nothing strange about that; most organizations start here. What does matter is the order in which that is built up. At why the fundamental dimensions come first you can read why organization, IT infrastructure and data management do not stand alongside the other dimensions but beneath them — an intelligence level on decision-making has little value if the data on which those decisions rest is not reliable.
The second scenario is less visible and often more surprising for an executive board. The organization has indeed built up something — pilots, separate applications, a team working with it — but the people who fill in the questionnaire score substantially differently. The CIO sees an infrastructure that is ready; the COO sees processes that still rest on manual control. The CHRO sees an organization that embraces AI; the team leaders see people who leave the tools aside as soon as things get tricky. In this scenario, the low average score is not the problem. The spread is the problem.
This is the reason the measurement works with a plot round: several people score separately, without seeing each other's answers, and only afterward does the spread become visible. An organization with a low average and a small spread knows where it stands and can move forward from there. An organization with an average that looks acceptable but with a large spread has a different problem: there is no shared picture of reality, and that picture has to exist first before a score can be about anything.
The measurement identifies the spread. It does not explain it, and it does not solve it. If the CIO and the COO structurally score differently on IT infrastructure, that is a conversation that needs to be had — about exactly what has been built, who has access to it, and what part of it runs in production versus in a pilot environment. The measurement provides the starting point of that conversation, not the outcome. The same applies to scenario one: knowing that there is nothing yet is not a plan to start building something. At how mature is your organization when it comes to AI and how mature is your IT infrastructure when it comes to AI you will find, per dimension, what the levels from baseline to intelligence look at, so that a low score gets a direction instead of just a judgment.
A low score on the fundamental dimensions does not mean that dependent themes such as decision-making and agent behavior only become relevant a year from now. It means that those themes can already be documented now, even if execution still has to wait. Which decisions are actually made in an organization and by whom is described at what belongs in a decision inventory, and which of those decisions remain in principle beyond the reach of automation — for legal, ethical or governance reasons — you can read at which decisions may never be automated. That work can start while the fundamental dimensions are still growing; it does not need to wait until the infrastructure is in order, although execution will indeed wait for that.
A low score predicts no timeline and no outcome. Two organizations with an identical score can have very different paths ahead of them, depending on exactly where the spread lay, which scenario applied, and how much of the foundation was already informally present without the questionnaire being able to capture it. The measurement is a snapshot with seven dimensions and five levels; it is not a prediction of what happens after that snapshot.
Where the maturity measurement maps the organization's capacity to bear the load — can the organization carry AI, based on what is in place in terms of foundation and shared picture — another question is about something else: what can AI actually take over in this organization. That is a task-level question, not an organization-level question, and it is answered by the work scan of FTE TO AI, which calculates per task which part of it can be taken over by AI. The two questions follow one another: only once it is clear whether, and where, the organization can bear the weight does it make sense to know what can actually be taken over. Anyone who sees a low score on the measurement therefore cannot reverse those two questions — but can already map out both of them in advance.
Vraag maar wat er moet staan voordat AI in uw organisatie kan landen.
Answers come from this site’s knowledge base. Not tailored advice, and not a scan of your company.