The habit loop is the model most popular habit books are built on, and for good reason: it matches how the brain’s habit system, centered on the basal ganglia, appears to work. Charles Duhigg described it as cue, routine, reward. James Clear expanded it to cue, craving, response, reward. The details differ; the practical use is the same. A habit is not a single behavior but a loop, and each part of the loop is a place you can intervene.
The three parts
1. Cue
The trigger. Cues come in a handful of types: a time (7am), a place (the kitchen), a preceding action (finishing coffee), an emotional state (boredom, stress), or other people (a colleague heading out for a smoke). Cues are what make habits automatic: the brain learns to start the routine the moment the cue appears, before you have consciously decided anything.
2. Routine
The behavior itself, which can be physical (a walk), mental (a worry spiral), or emotional (reaching for reassurance). This is the part most people try to change directly, which is why most people fail: the routine is the middle of the loop, and the loop is held together at both ends.
3. Reward
What the routine delivers. The reward teaches the brain whether the loop is worth remembering. Immediate rewards wire loops in fast; delayed rewards barely register, which is why exercise (delayed reward) is hard to make automatic and scrolling (instant reward) is not.
The craving is the glue. Over time the cue starts to trigger anticipation of the reward, and that anticipation is what drives the routine. You are not craving the cigarette; you are craving the relief it predicts.
Using the loop to build a habit
| Stage | Make it… | How |
|---|---|---|
| Cue | Obvious and reliable | Attach the habit to something that already happens daily. Habit stacking: “after I pour coffee, I…” The habit stacking planner writes the sentence. |
| Routine | Easy | Shrink it to two minutes. Remove steps between the cue and the action. See the two-minute rule. |
| Reward | Immediate | Add something now: the punch on a card, a podcast only during the walk, coffee only after the writing. Rewards that support the identity last longer than rewards that undercut it. |
Using the loop to break a habit
You rarely eliminate a loop; you edit it. Two strategies work:
- Keep the cue and the reward, swap the routine. The 3pm slump (cue) leads to a walk to the vending machine (routine) for a break and a chat (the real reward). Keep the cue, keep the break and the chat, replace the vending machine with a walk to a colleague’s desk. This is the core move in The Power of Habit and it works because the craving is satisfied.
- Remove or hide the cue. Phone in another room, app deleted, snacks not bought. If the cue never fires, the loop never starts. This is why environment design beats willpower; see building self-discipline.
The first step in either case is diagnosis. For a week, when the bad habit fires, note the cue type (time, place, preceding action, feeling, people) and what the routine actually delivers. The real reward is often not the obvious one.
Why tracking fits the loop
Most good habits have a delayed reward and an immediate cost, which is the worst possible shape for the loop. A visible tracker fixes the reward end: the punch is immediate, satisfying, and tied to the routine. It also strengthens the cue end, because a card on the fridge or a widget on the home screen is itself a cue. That double role is why tracking is one of the best-supported techniques in behavior change, and why the medium matters: it has to be visible without effort.
The loop over time
Early on, the loop needs all three parts deliberately built. After enough repetitions, the cue alone triggers the routine and the reward fades into the background; the habit has become automatic. That takes about two months on average, with a wide range depending on the behavior. The habit formation calculator maps the research landmarks onto your calendar. Missing a day does not break the loop; long gaps weaken it, which is the reasoning behind the never-miss-twice rule.
Limits of the model
The habit loop is a simplification. Not all behavior is habitual; goals, values, and mood matter, and many bad “habits” are better understood as coping strategies for something else. The loop is most useful for behaviors that fire automatically in a stable context. For those, it is the most practical tool available, because it turns a vague resolution into three specific questions: what is the cue, what is the routine, and what is the reward?
Track it with a punch card
Punchcard turns any habit or countable goal into a card of punches on your iPhone home screen, with reminders, rewards, and shared cards for up to four people.
Get Punchcard freeFrequently asked questions
What is the habit loop?
A three-part model of how habits work: a cue triggers a routine, and a reward reinforces it, so the next time the cue appears the routine fires more automatically. It was popularized by Charles Duhigg in The Power of Habit.
What are the four stages of the habit loop?
James Clear’s version in Atomic Habits adds craving: cue, craving, response, reward. The craving is the anticipation of the reward that the cue triggers and that motivates the response.
How do I use the habit loop to build a habit?
Make the cue obvious by attaching the habit to an existing routine, make the routine easy by shrinking it, and make the reward immediate, for example by marking each repetition on a visible card.
How do I use the habit loop to break a bad habit?
Identify the cue and the real reward, then keep both and replace the routine with something that delivers the same reward. Alternatively, remove the cue from your environment so the loop never starts.
What types of cues trigger habits?
Five common types: a time of day, a location, a preceding action, an emotional state, and other people. Identifying which one triggers a habit is the first step to changing it.