You keep repeating a behavior and can’t say why. The habit loop explains it: a cue sets off a routine, and a reward makes you likelier to repeat it. Name those parts and you can start changing the pattern.
This guide is part of our wider look at habit building, and it covers the pattern behind most repeated behaviors.
| Item | Value |
|---|---|
| Elements in the popular habit loop | 3 (cue, routine, reward) |
| Median time to reach 95% of maximum automaticity (Lally and colleagues, 2010) | 66 days |
| Spread across individuals in that study | 18 to 254 days |
| Volunteers in that study | 96 |
| Median in a 2024 meta-analysis of habit-formation studies | 59 to 66 days |
| Effect size of if-then plans across 94 studies (Gollwitzer and Sheeran, 2006) | d = 0.65 |
These figures come from Goals & Progress and The Behavioral Scientist. They can change as new research appears, so check the source before you rely on them.
What the habit loop is
The habit loop is a popular description of repeated behavior organized around three elements: a cue, a routine and a reward. Charles Duhigg popularized this three-part version in The Power of Habit, as The Behavioral Scientist glossary notes. It helps you describe a pattern, but it is not a complete theory of how every habit forms or changes.
Habit research gives the idea a firmer footing. Wood and Rünger (2016) describe habits as responses learned through repetition in recurring contexts, according to the same glossary. Rewards and goals can support learning, yet an established habit does not require a consciously anticipated reward on every occasion.
The three parts of the habit loop
Each part of the loop has a plain job. The routine is also called the response, because it is what you do when the cue appears.
- Cue: a signal or context associated with the response, such as arriving home.
- Routine: the response itself, such as opening a snack cupboard.
- Reward: an outcome that may encourage repetition, such as eating something enjoyable.
Other writers use different labels. Goals & Progress calls the same mechanism Trigger, Action and Reward, and says its words are plainer but its mechanism is identical.
Cues come in many forms. The same source lists five recurring categories: a time of day, a location, the end of a preceding action, a mood and a person. In a 2024 interview study, researchers asked 43 office workers what prompted them to move during the workday, and the cues were concrete and environmental, such as a drink, food or a standing desk, according to that write-up.
A preceding action that already runs on its own makes a dependable cue. That is the basis of habit stacking, where you use a finished habit as the cue for the next one, as Goals & Progress describes.
A worked example: the afternoon cookie
Suppose you buy a cookie from a shared kitchen every afternoon. The time and place are possible cues, buying the cookie is the routine, and the outcome may be food, a break, or conversation with a colleague, as The Behavioral Scientist sets out.
The useful question is what the behavior does for you. Moving the colleague’s desk will not solve hunger, and replacing the cookie with tea will not solve a need for social contact. The diagram helps you name questions, but observation and testing tell you which answer matters.
You can try a simple version yourself:
- Note when and where the behavior happens.
- Name the routine in one plain sentence.
- Ask what you get afterward: hunger relief, a break, company or something else.
- Change one thing and watch whether the behavior shifts.
How long does a habit loop take to form?
Lally and colleagues (2010) measured the formation curve directly in 96 volunteers across twelve weeks. They reported a median of 66 days to reach 95% of maximum automaticity, with individual variation from 18 to 254 days, and a single missed day did not disrupt the underlying curve, according to Goals & Progress.
The same page reports that a 2024 systematic review and meta-analysis of twenty habit-formation studies put the median in the same neighborhood, 59 to 66 days. It also says the popular 21-days-to-a-habit claim has no empirical backing.
Keep the limits in mind. The 2010 study looked at changes in self-reported automaticity during repeated everyday behavior, and The Behavioral Scientist stresses that it did not test every part of the popular loop or show that replacing a routine always changes a habit.
Using the habit loop to change a behavior
Identifying a cue, response and outcome can help you decide what to try changing. Whether a replacement works depends on what keeps the original behavior going and whether the alternative fits, according to The Behavioral Scientist. If you want to go further with an unwanted pattern, read our guide on how to break a bad habit.
Naming the cue in advance also helps. Gollwitzer and Sheeran’s 2006 meta-analysis across 94 studies found a medium-to-large effect (d = 0.65) for if-then plans that link a specific cue to a specific response, per Goals & Progress. A plan such as “When the kids leave for school, I will sit at the desk and write” names both halves.
Be careful with the popular advice to keep the cue and reward while swapping the routine. Treat it as an idea to test, not a universal law. The relevant outcome may be uncertain, several outcomes may matter, and the context itself may need to change.
Where the habit loop falls short
The loop is a starting point for investigation, not a recipe that guarantees lasting change. It describes simple recurring responses well, but broader, sustained behavior change needs more than a tidy diagram.
It is also not the same thing as a product-design model. The glossary says frameworks such as the Hook Model are not interchangeable with the cue-routine-reward description, because that model was designed to analyze repeated product engagement. A behavioral habit can exist without a digital product, a variable reward or the full product loop, per The Behavioral Scientist.
Frequently asked questions
What is the habit loop in simple terms?
Show answer
It is a three-part description of a repeated behavior: a cue signals the moment, a routine is what you do, and a reward is the outcome that may encourage you to repeat it. The Behavioral Scientist glossary calls it a useful way to describe a pattern, not a full theory of habits (source).
Is the habit loop the same as the Hook Model?
Show answer
No. The glossary says product frameworks such as the Hook Model are not interchangeable with the cue-routine-reward description, since the Hook Model analyzes repeated product engagement (source).
Do habits really take 21 days to form?
Show answer
The evidence does not support that figure. Goals & Progress reports a median of 66 days in the 2010 Lally study and 59 to 66 days in a 2024 meta-analysis, and says the 21-day claim has no empirical backing (source).
Does a single missed day ruin a habit?
Show answer
In the Lally study, a single missed day did not disrupt the underlying formation curve, according to Goals & Progress.
Can I swap the routine and keep the same cue and reward?
Show answer
You can test it, but it is not a guaranteed method. The right swap depends on what the behavior does for you and whether the new routine fits your context, as The Behavioral Scientist explains.
