Build a habit for three kinds of day.

Your week does not have one operating condition

A plan such as “read for 20 minutes after dinner” can work beautifully on a normal evening. Then a late train, a guest or a long conversation changes the shape of the day. If the plan has only one valid size, you face a false choice: force the full version into the wrong conditions or record a failure.

The problem is not that life varied. The plan failed to describe what should happen when it did.

Useful flexibility is decided before the difficult moment. It does not mean that any gesture can be counted after the fact. It means that the plan contains several legitimate responses to several predictable conditions.

Use three day types:

1. Ordinary: the conditions you designed the habit around.
2. Constrained: the cue appears, but time, attention or access is meaningfully reduced.
3. Open: the usual action fits and there is genuine room to continue.

These are operating conditions, not identities. You are not “a constrained-day person.” Thursday was constrained. Friday may not be.

Keep one doorway and change the distance

Habits are commonly understood as responses learned through repetition in recurring contexts. In a 12-week study, Lally and colleagues observed that automaticity generally rose as people repeated a selected daily action in the same context, although individual curves varied widely. In a later randomized study of everyday nutrition behaviours, both routine-based and time-based cue plans were associated with growing automaticity when people repeatedly enacted the plan; neither cue type outperformed the other at group level.

The practical implication is modest: if every version starts from a completely different signal, place and procedure, you may be rehearsing three unrelated plans. Preserve the doorway when you can.

For example:

• Cue: after I put my dinner plate in the sink.
• Ordinary: read for 20 minutes in the chair.
• Constrained: read one page before leaving the kitchen area.
• Open: read for 20 minutes, then continue if I still want to.

The size changes. The cue, material and first real action remain recognisable.

This does not guarantee that one cue will operate identically in every setting. Context includes more than a clock time: location, preceding activity, other people and the materials available can all matter. The three-version plan is a way to reduce unnecessary variation, not a method for making context irrelevant.

Write a three-line version map

Choose one habit and complete this worksheet.

Then test each line against four questions.

Is the condition observable?

“When I feel busy” invites a fresh argument each time. “When I have less than ten minutes before the next commitment” is easier to recognise. You do not need laboratory precision, but the condition should be clearer than a mood.

Is the constrained version still the behaviour?

Opening a notebook and writing one honest sentence can be a small writing action. Moving a “write” card to done is record-keeping. Preserve the first real unit of the activity.

Does the ordinary version have a finish?

“Work on the project” can expand without limit. “Sort one folder” or “work until the 20-minute timer ends” tells you when the commitment is complete.

Is the open version optional?

More on an easy day must not increase tomorrow’s minimum. An open version is permission, not a loan that future days must repay.

Example: a language-reading habit

Suppose you want to read fiction in a language you are learning after making evening tea.

• Ordinary day: tea is made before 9:00 and the next half-hour is free. Read two pages, looking up no more than three words.
• Constrained day: tea happens late or another commitment begins within ten minutes. Read one paragraph and underline one unfamiliar phrase for later.
• Open day: finish two pages, then continue until the end of the scene if it remains enjoyable.

All three begin with tea, the same book and the act of reading. The constrained version is not a simulation. The open version does not produce credit that cancels the next ordinary opportunity.

If tea disappears entirely because you are away from home, do not quietly invent a fourth version. Name that as a changed context. Either use a predesigned travel cue or wait for the next real opportunity. A plan becomes less useful when it pretends every circumstance is merely a different-sized day.

Review the distribution, not the morality

At the end of two weeks, count the versions and add one reason beside each constrained day.

• If ordinary versions dominate and the cue is becoming easier to notice, the plan may be well fitted.
• If constrained versions dominate because the ordinary version rarely fits, promote a smaller version to ordinary. Your baseline should describe your actual week.
• If the constrained version appears mostly after avoidable setup friction, improve access before shrinking again.
• If open versions regularly leave you depleted or make the next session harder, give the optional continuation a firmer stop.
• If all versions disappear in one setting, design for that context separately instead of blaming the scale.

Do not ask whether you “cheated” by using the floor. Ask whether the condition matched the rule and the action remained real. A correctly used constrained version is the plan working under constraint.

Flexibility needs boundaries

Three versions can still become a way to preserve a log rather than a meaningful practice. Watch for two failure modes.

The first is permanent downshifting: the tiny version becomes automatic to choose even when ordinary conditions are present. That calls for a review of the baseline, the cue or the value of the habit—not shame.

The second is open-day compensation: a large session is used to erase earlier small ones. Unless the activity has a genuine external deadline, let each opportunity stand on its own.

The goal is not a streak with better excuses. It is a plan that lets you inspect its conditions, adjust without drama, return at the next cue and eventually need less scaffolding.

Evidence notes and sources

• Lally, P., van Jaarsveld, C. H. M., Potts, H. W. W., & Wardle, J. (2010). How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40, 998–1009. Repetition in a consistent context was associated with rising self-reported automaticity, but curves and estimated formation times varied substantially. The paper did not test a three-version planning system.
• Keller, J., Kwasnicka, D., Klaiber, P., Sichert, L., Lally, P., & Fleig, L. (2021). Habit formation following routine-based versus time-based cue planning: A randomized controlled trial. British Journal of Health Psychology, 26, 807–824. Both assigned cue-planning conditions showed increases in automaticity and plan enactment; repeated enactment predicted automaticity, and no group-level advantage emerged for one cue type.
• Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119. Specific if–then plans supported goal attainment across varied studies; this evidence supports advance planning, not the claim that three day types are a validated universal technique.