6.10.5 Hazard warning measures

What this heading covers

A warning that is displayed and ignored has not worked. The measure therefore has two parts: delivery and effect. Delivery is countable from the product; effect needs a user study or an incident review. The warning should be tested in the conditions where it matters, including a distracted user and a noisy environment. The measure should also record false warnings, because alarm fatigue degrades every later warning.

Record position: family 39 of 40 in clause 6, under 6.10 Safety, which carries 5 families. It is one of the 40 families a programme has to map its measures to, and its elements are listed below. The heading is 6.10.5 in the public contents; the text behind it is not public.

How a measure for this family is built

Count the hazard conditions in which the user noticed and acted on the warning as the numerator over the conditions tested. Report delivery time, notice rate and action rate separately. Record false warnings in the same period.

What to put in place before publication

1. Test the warning with users in realistic conditions.
2. Record false warnings alongside the true ones.

Evidence to keep

Two questions to ask

• Was the warning tested with users in realistic conditions?
• Were false warnings counted alongside the true ones?

Where teams usually fall short

• A warning display test with the product team watching for it.

What the number cannot tell you

It cannot tell you whether the warning would work for a different user group. The notice rate belongs to the people in the study, and a different group may not see the warning at all.

What happens next

Safe integration is the last family: the product inside a larger system, where the boundaries carry risk.