Even seemingly inconsequential bits of friction in changing behaviour

The extraordinary reductions in suicide resulting from changes in levels of carbon monoxide might have happened by accident, but the insight can be used to make deliberate changes that have reduced suicides. For example, a number of countries have introduced legal restrictions on the number of paracetamol tablets and similar everyday medications that can be bought in one go. There is not much to stop the determined buyer from going into several stores in a row and buying more pills, but it has been shown that in the UK alone such measures were associated with around 70 fewer suicides a year as a result of paracetamol ingestion (a 42 percent reduction), and an even bigger reduction of 61 per cent of patients needing a liver transplant as a result of damage from paracetamol. Similarly, there is evidence that where such pills are required to be sold in pop-out packs, rather than loose in a bottle, this also reduces suicide rates since the pills have to be taken out one at a time. A little friction, it turns out, is not always a bad thing.

Friction has been slandered by people selling one-click buttons.

Sometimes slowing an action is the product. Medication packs that make tablets emerge one at a time place tiny moments between impulse and irreversible harm. A bank checking a new payee interrupts a transfer just long enough for doubt to catch up. Typing a resource name before deletion makes the finger consult the brain.

Good friction is a speed bump outside a school.

Bad friction is the hotel hairdryer: attached to a wall, permanently underpowered and apparently designed by someone angry at damp people. Re-entering an address, phoning to cancel, creating an account to buy a train ticket—these do not protect customers. They protect a metric, a decaying system or a retention target. “For your security” has concealed a heroic amount of corporate convenience.

Four questions expose the difference: Who gets slowed? At what moment? Which harm is prevented? How easily can that harm be reversed?

Remove effort when intent is stable and error cheap. Add a pause when emotion is hot, consequence large or recovery impossible. Tell people why it exists. Useful protection can withstand an explanation; bureaucratic vandalism generally changes the subject.

A humane journey does not worship speed. It puts effort where effort improves the decision. Every extra second is not a failure. Sometimes it is the second in which somebody avoids one.