24/7 news makes long-term progress harder to see

Imagine three publications covering the same fifty years.

The minute-by-minute feed reports millions of alarms, announcements, rows and reversals. The daily paper selects events that began or ended before bedtime. The fifty-year edition notices child mortality falling, employment changing and diseases becoming treatable.

None owns reality. Each interval admits different evidence.

This creates a nasty incentive in news and organisations alike. Anything important enough to develop slowly struggles to look new today. Gradual improvement becomes background; one setback gets a timestamp and a photograph. Decline works the same way. A system can deteriorate quietly for years while every daily indicator remains within tolerance.

Slow summaries have blind spots too. Averages can hide the day one group suffered badly. Waiting for the long trend during an emergency is statistical elegance applied to somebody else’s fire.

The answer is plural clocks. Fast measures for events requiring action now. Longer windows for direction, distribution and cumulative consequence. The two should be allowed to disagree without one being dismissed as “the real picture”.

A service feed shows complaints rising sharply this afternoon. The annual review shows complaints down overall and concentrated increasingly among one customer group. Celebration and alarm are both inadequate.

The editor of the imaginary fifty-year paper has space for twelve stories. Today’s loudest row receives none. A quiet curve gets half the front page.

Behavioural principles

Behavioural ideas at play in this post

Short, plain-English explanations of the principles behind this post, with links to related books and examples in the archive.