Writing

Short pieces on behavioural science in practice. Mostly about diagnosis, and about what happens to a finding once it leaves the researcher.

We have spent fifteen years teaching people that consent is an obstacle

Across 1,280 recorded answers in one study, 17 were specific consent. The alarm-fatigue literature explains why, and points at a spillover nobody has measured.

August 2026

The default that nobody was bound by

England switched to opt-out organ donation in 2020. Deemed consent now runs at 48 percent against 87 for expressed consent, and the gap is the argument.

August 2026

The intervention was chosen before the diagnosis

Training, campaigns and dashboards have vendors and budget codes. Removing an approval step has neither. That asymmetry shapes what gets done.

August 2026

Heatwaves may not just be an environmental problem, but also a behavioural one

Heat lowers the threshold for aggression rather than creating it. Cities that were never built for heat have nothing absorbing the impact.

August 2026

When warning everyone is the same as warning no one

A COM-B diagnosis of warning fatigue in a UK Met Office brief, and why simplification and personalisation solve different problems.

July 2026

How to use behavioural science to scale businesses via enhanced AI adoption

Transformations fail on the human system, not the technology. What a COM-B diagnosis of low AI uptake actually looks like.

July 2026

LinkedIn|[email protected]