The cost of a small font

A government form was redesigned and the number of errors fell by about a third. No wording changed. The alterations were spacing, a larger type size, and moving the box for the date so it no longer sat immediately beside the box for the reference number.

This category of improvement is poorly named. It is usually filed under design, which makes it sound like decoration, when what it actually does is reduce the number of ways a reasonable person can go wrong.

The date box is a good example. Two boxes of similar size next to each other, both wanting numbers, produce a predictable rate of transposition. Nobody who does this is careless. The form invited the error and then recorded it as a user mistake, which is how most such errors are recorded.

The economics are usually overwhelming once anyone measures them. A third of errors removed means a third fewer forms returned, re-sent, re-opened and re-processed, and the cost of that work dwarfs the cost of the redesign many times over. The difficulty is that the redesign cost appears in one budget and the saving appears in another, several departments away, which is why such projects are hard to approve and easy to justify afterwards.

The general lesson applies well beyond forms. When a system produces the same mistake from many different people, the mistake is a property of the system rather than of the people, and the cheapest fix is almost always at that end.