The other week I was on a train going across the country, and while it was stopped at a station the train manager made an announcement. “I realise this is unusual,” he said, “but we’re running ahead of schedule…” There was some laughter from the other passengers. But I was slightly surprised by what followed, “…so we’re going to sit here for a few more minutes until our slot becomes available.”

This struck me as a very odd situation. The national rail network seems to be system designed to either meet expectations or to fail, but never exceed expectations. You can be on time, you can be delayed, but the system won’t allow you to be ahead of time.

This seems to be asking for trouble. In general, in our work, we shouldn’t be in situations where our very best performance will merely meet expectations, and anything less is failure. In our work, and our projects, we should seek a variable success measure that allows us to do well and yet still allows us to find ways to do better—and, of course, by which we can be fairly accountable for disappointing results.

Contrast the national rail network to London Underground in this regard. Superficially it’s the same—trains there are held back if they’re running ahead of time. But trains are so frequent on the Underground that travellers don’t think about any kind of timetable throughout the day. There are delays and the odd cancellation, but the frequency of trains and the multiple routes round the network allow all those problems to be absorbed. The usual time for any journey on the Underground will be longer than the optimal time for that journey, but that usual time is actually quite acceptable. If everything works exactly as planned then we get to our destination slightly sooner. But if a train is cancelled on the national rail network that’s a significant impact on its passengers’ journeys.

In general we should aim to design systems that allow degrees of success, as well as degrees of failure.

Photo by Adam B