Concepts & methods

Implementation intentions: the when-then move that survives scrutiny

Naming exactly when and where you will act is one of the best-evidenced moves in behavioural science, and one of the easiest to ruin by over-use. A reference entry on the research, the honest effect size, and where the move belongs.

Aleem O'BalogunBy Aleem O'Balogun|Updated |4 min read

The short answer

An implementation intention is a plan that names the cue for an action in advance: when situation X arises, I will do Y. Coined by Peter Gollwitzer in the 1990s, it reliably narrows the gap between intending something and doing it, with meta-analytic support unusual for behavioural science. Its power is also its limit: it works for specific, cue-able actions, and deadens when applied to everything.

What it is, and who established it

The intention-behaviour gap is behavioural science's most relatable finding: people who fully intend to act often do not. Peter Gollwitzer's contribution, developed through the 1990s and set out in a landmark 1999 paper, was a format that closes much of the gap without adding motivation: specify in advance the when and where, not just the what. “I will exercise more” leaves initiation to a future self who must notice an opportunity, decide, and start, all at the worst possible moment. “When I finish Tuesday's last meeting, I will change and run” delegates initiation to the cue; the situation itself triggers the act, and the deciding is already done.

What the evidence actually says

The headline evidence is a 2006 meta-analysis by Gollwitzer and Paschal Sheeran covering 94 studies, finding a medium-to-large effect of forming implementation intentions on goal attainment. Later syntheses correcting for publication bias land lower but still solid, which for a free, one-sentence intervention is remarkable. The most quotable single result is the 2002 exercise study by Sarah Milne, Sheina Orbell and Sheeran: 91 percent of participants who wrote down when and where they would train exercised the following week, against under 40 percent in both control groups. That study carries the exercise guide on this site, which is what the move looks like in practice.

Where it breaks

Three honest limits. First, the move needs a real cue: vague triggers (“when I have some downtime”) trigger nothing, and complex ongoing work does not reduce to one cue-able act. Second, it front-loads a decision, not motivation; a when-then plan for something you do not actually want to do mostly schedules your ambivalence. Third, and most practically: over-application kills it. Formatting every task on a list as a when-then turns a sharp tool into bureaucratic sludge, and the ritual of writing conditions replaces the acting. The move belongs on the few behaviours that are important, fragile, and specific; everywhere else, an ordinary planned block does the same job with less ceremony.

How it maps to the notice loop

A calendar block is an implementation intention wearing work clothes: “Tuesday, 6:30, run” is when-then with the cue set to a time. Which means a well-shaped week is already a page of implementation intentions, formed once, in advance, at the loop's shaping altitude, rather than improvised per task. The research is, quietly, the strongest evidence base under weekly planning itself: deciding when, in advance, is most of why a planned week outperforms a to-do list. Reserve the explicit if-then sentence for the fragile few (the medication, the difficult call, the first run back), and let the calendar carry the rest.

Sources

The primary and meta-analytic sources for the effect, including the honest lower bound after bias correction.

  • Gollwitzer, P. M. (1999), Implementation intentions: strong effects of simple plans, American Psychologist, 54(7).
  • Gollwitzer, P. M. and Sheeran, P. (2006), Implementation intentions and goal achievement: a meta-analysis, Advances in Experimental Social Psychology, 38. Later bias-corrected syntheses land lower but still solid.
  • Milne, S., Orbell, S. and Sheeran, P. (2002), Combining motivational and volitional interventions to promote exercise participation, British Journal of Health Psychology, 7. The 91 percent exercise result.
← All guides