A new pattern needs repeated ordinary days before a person can safely rely on it.

Early success with a change proves less than it feels like it does. The first good week runs on novelty and attention, neither of which lasts, and the real test is whether the thing survives a dull Tuesday, a cold, a busy patch and a run of low motivation. Reliability is built out of unremarkable repetitions rather than strong starts.

Give any new habit a longer trial than seems necessary before you count on it. Keep it small enough to do on a bad day, since that is the day that decides. And treat a missed occasion as information about the design rather than as evidence about you.