AI-tools uitrollen is makkelijk. Mensen ertoe brengen ze daadwerkelijk te gebruiken — en dat vol te houden — is de echte uitdaging. Lage adoptie is niet primair een trainingsmin; het is een systeemprobleem.
Licence utilisation is the most underreported metric in enterprise AI. Organisations regularly purchase AI tools for hundreds of employees and find, six months later, that 20–30% of licences are actively used — and that a third of those active users are using the tool for basic tasks that represent a fraction of its capability. Getting people to use AI well is harder than getting them to buy it. Here is how to close that gap.
The Utilisation Problem Is Bigger Than You Think
Het gebruiksprobleem is groter dan u denkt
Microsoft's data on Copilot adoption, Salesforce's Einstein utilisation reports, and our own observations across client organisations consistently show the same pattern: licence purchase and genuine adoption are very different things. The gap is not a failure of the technology — it's a failure of implementation strategy.
The most common failure mode is treating AI tool rollout like a software upgrade: announce it, run a training webinar, grant access, and expect adoption. This approach works for tools people have to use (like a new ERP system that replaces the old one). It doesn't work for tools people can choose to use — which is the category most AI tools fall into.
Start With High-Value, Low-Effort Use Cases
Begin met hoogwaardige, laagdrempelige toepassingen
The fastest path to genuine adoption is demonstrating quick wins in the first days of use. This means identifying, for each role type in your organisation, the single most tedious task that AI can meaningfully improve — and making that the first thing every new user does with the tool.
The task should be one that takes 30–60 minutes today, that the person finds genuinely annoying, and that AI can handle in under five minutes. The experience of this first win creates the intrinsic motivation that no amount of top-down mandate can replicate. People who have felt the time saving personally become advocates.
Role-Specific Onboarding Over Generic Training
Rolspecifieke onboarding boven generieke training
Generic AI training — "here is what the tool can do" — creates generic AI users. Role-specific onboarding — "here is how you, as a credit analyst / logistics planner / HR manager, will use this tool for your actual work" — creates genuine capability.
The investment in role-specific onboarding is higher, but the adoption rate is dramatically better. People don't need to understand every capability of a tool to use it well — they need to deeply understand the five or ten use cases most relevant to their work. Build onboarding programmes around those use cases, with live practice on real examples, and you will see adoption rates three to four times higher than generic training produces.
Build Social Proof Through Champions
Bouw sociaal bewijs op via ambassadeurs
In every team, there are people who will explore new tools enthusiastically. These natural early adopters are your most important adoption asset. Identify them, give them early access and dedicated support, and position them as internal champions who can demonstrate the tool to their peers.
The peer-to-peer demonstration is more credible than any vendor presentation or management directive. When a colleague shows you how she saved an hour on the monthly report using AI, the proof is tangible and the source is trusted. Champions also surface the practical questions that formal training doesn't anticipate — making your broader programme better.
Make Non-Adoption Visible
Niet-adoptie zichtbaar maken
Tools that can be ignored are ignored. If AI adoption is tracked and reviewed — in team meetings, in performance conversations, in operational reviews — people take it seriously. This doesn't require punitive enforcement. It requires making the question "are we using our AI tools effectively?" a regular and visible part of management conversations.
Practical tactics: include AI tool usage metrics in team dashboards, recognise and share examples of effective AI use, make AI a standing agenda item in regular team meetings for the first three months, and ask managers to share examples of how they've used AI in their own work.
AI tool adoption is a management challenge as much as a technology challenge. The organisations that achieve high utilisation do so by making AI use the path of least resistance for high-value tasks, providing role-specific capability rather than generic training, building social proof through champions, and maintaining visible management attention on adoption. None of these require large budgets — they require deliberate design and consistent execution.
Visser & Van Zon designs AI adoption programmes that convert licence purchases into genuine organisational capability. If your AI tool rollout isn't achieving the utilisation you expected, we'd be glad to assess the gap and help you close it.