Blog
Notes on measuring whether training changes behaviour: the science, the stories and what we're building.
-
Beyond completion rates: what an ant mill teaches L&D
Completion rates show that people turned up, not that anything changed at work. A 1921 account of army ants circling a 365-metre loop is a useful picture of activity without progress, and of why L&D needs measures of behaviour, not attendance.
-
The forgetting curve in corporate training: what Ebbinghaus did and didn't show
Ebbinghaus showed in the 1880s that forgetting is fast at first and then slows. He tested one person, himself, on nonsense syllables, measured by relearning time. That shape has been replicated; the popular '70% forgotten in a day' figure isn't what he found.
-
Why training doesn't change behaviour, and what a specific plan adds
People often leave training meaning to act and then don't. Research on 'if-then' plans suggests the gap between intending and doing is partly a planning problem: naming when, where and how you'll act makes action more likely. Design for that moment, then check whether it happened.