Adoption Curve
The path from a few enthusiasts to real usage.
The progression of a tool through a population: early enthusiasts, active majority, holdouts. It usually stalls between the first two unless the tool becomes the default path.
Internal AI tools stall for predictable reasons — first-run failure, unclear permission boundaries, no visible win. Fixing onboarding and making the tool the path of least resistance matters more than adding capability.
The stall point is predictable: enthusiasm gets a tool to early adoption, and the path to the majority runs through making it the default rather than an option. That means integrating it into existing workflows, removing the approval friction, and being honest about where it does not help.
The mandatory rollout that produced resentment
A tool is made compulsory after voluntary adoption plateaus.
People who found it unhelpful for their work now have to route around it, and the reported benefits become unreliable. The durable version is default in the workflows where it demonstrably helps, optional elsewhere, and measured rather than announced.
- reportState of AI-assisted software development 2025 ↗
Frames AI as an amplifier of the organisation it lands in: the same tooling helps strong delivery cultures and compounds problems in weak ones.
- reportWe are changing our developer productivity experiment design ↗
METR explains why self-selection in their follow-up experiment made the productivity estimate unreliable. Read alongside the 2025 result before quoting either as a verdict.
ROI · Return on investment
Did it save more than it cost?
DX · Developer Experience
How it feels to build with a tool or platform.
AX · Agent Experience
How it feels to be an agent using your product.
Shadow AI
Staff using unauthorised AI tools with company data.