Why Deadlines Change Everything

A. In 1955, the British historian Cyril Northcote Parkinson published a short essay containing a sentence that would outlive everything else he wrote: work expands so as to fill the time available for its completion. He offered it half in jest, describing an elderly woman who spends an entire day posting a single postcard — a task a busy person would finish in three minutes. Yet the observation has since been tested repeatedly, and it holds up with uncomfortable regularity.

B. The mechanism is not laziness. Psychologists studying task completion find that when people are given generous deadlines, they do not simply work more slowly; they work differently. They gather more information than necessary, revisit decisions already made, and expand the scope of the task itself. A report that could have been six pages becomes twenty. The extra time is genuinely spent, but much of what it produces adds little.

C. Tight deadlines change behaviour in the opposite direction. Faced with limited time, people become ruthless about what matters. They abandon perfectionism, make decisions on partial information, and — crucially — start immediately rather than later. In a study of software teams at a European university, groups given half the usual time delivered products rated only marginally worse by users, while spending sixty per cent fewer hours.

D. The obvious conclusion — that all deadlines should be short — is wrong, and dangerously so. The same research found that when deadlines fell below a certain threshold, quality collapsed rather than declined gently. Errors multiplied, teams stopped communicating, and the work often had to be redone entirely. The benefit of pressure follows an inverted U: too little produces waste, too much produces failure, and the useful zone between them is narrower than most managers assume.

E. What complicates matters further is that people are poor judges of where they sit on that curve. Asked to estimate how long a task will take, most of us underestimate consistently — a bias so reliable that it has its own name, the planning fallacy. The same person who needs twice the time they predicted will, when planning the next task, predict just as optimistically. Experience, remarkably, does not cure this.