Unit of change
The thing an AI program sets out to change: a tool, a task, a workflow, a decision or the whole operating model. Choosing too small a unit is why adoption rises without value.
The unit of change is the level at which an organization decides to apply AI. It is usually chosen implicitly, and it determines what results are possible.
Common units form a ladder: the tool (give people a licence), the task (automate a step), the workflow (redesign a sequence of steps), the decision (change who or what decides, and when) and the operating model (change roles, governance and handoffs). Most programs sit on the first two rungs because those are easy to count and launch.
Why it matters
Adoption is the wrong unit of change. The mistake is not automation; it is choosing the tool or task when the economically meaningful unit is usually larger. A faster task inside an unchanged workflow mostly makes yesterday's process run faster, and the gain does not reach the P&L.
Moving up the ladder is the substance of AI transformation, and the reason AI adoption metrics alone mislead. See also operating model.
Read more in Stop adopting AI.
Related terms
AI adoption
How widely an organization's people use AI tools — licences, active users, use cases. It is rising faster than enterprise value, because adoption is not the same as changing how the work gets done.
AI transformation
Changing how a company works — its workflows, decisions and operating model — around what AI makes possible, rather than adding AI tools to existing processes.
Workflow redesign
Rethinking the steps, roles and decisions of a process around what AI makes cheap, instead of inserting AI into the existing steps. Stronger AI performers do it more often.
Operating model
How an organization actually runs: its workflows, decisions, roles, governance and handoffs. AI creates value when it changes the operating model, not when it is bolted onto the old one.