# If-then plans: the small planning trick that helps eating habits stick

October 14, 2026 · Weight Management · 7 min read · https://burnweek.fit/blog/if-then-plans-for-eating-habits/

> 'I'll eat better' rarely survives a Tuesday evening. One sentence written in advance does better, as long as you avoid the phrasing that backfires.

**Key takeaways**

- Across 94 tests, forming an implementation intention ('if situation X, then I do Y') improved goal attainment with a medium-to-large average effect (d = 0.65).
- For eating, a 23-study meta-analysis found d = 0.51 for adding healthy foods but only d = 0.29 for cutting unhealthy ones.
- A 12-study meta-analysis of plans to reduce fat intake found a moderate effect (d = 0.49), larger in studies with more men and without professional monitoring.
- In a 55-woman trial within a commercial program, adding if-then planning roughly doubled 2-month weight loss (4.2 vs 2.1 kg).
- Plans phrased as 'if X, then NOT Y' produced ironic rebound effects in experiments; replace the habit instead of negating it.

## Writing "if X happens, I'll do Y" works better than good intentions, with modest and uneven effects on eating

An if-then plan, which researchers call an *implementation intention*, is one sentence written in advance: **"If [situation], then I will [action]."** "If I get home hungry at 6 pm, then I'll eat the yogurt before I start cooking." It sounds too simple to matter. Across dozens of experiments, it has on average done better than simply intending to eat well. The effect on eating is modest, and it depends on what you are trying to change.

The best summary of the evidence: if-then plans help more when they **add** a healthy behavior than when they **remove** an unhealthy one. For eating, a meta-analysis found a medium effect for adding healthy foods and a small one for cutting unhealthy ones ([Adriaanse et al., 2011](https://pubmed.ncbi.nlm.nih.gov/21056605/)). And in experiments, one common way of phrasing them, "if X, then I won't Y," sometimes strengthened the habit people were fighting.

This is a companion to our [guide to calorie deficits](/blog/calorie-deficit-weight-loss) and to our broader guide on [sustainable weight-loss habits](/blog/sustainable-weight-loss-habits). Here, the focus is on this one tool: how well it works, where it fails, and how to write one.

## The effect sizes, laid out

Effect sizes here are reported as Cohen's *d*, a standardized measure where about 0.2 is small, 0.5 medium, and 0.8 large.

| Evidence | Scope | Effect |
|---|---|---|
| Gollwitzer & Sheeran 2006 | 94 independent tests, all goal types | d = 0.65 on goal attainment |
| Adriaanse et al. 2011 | 23 studies, eating | d = 0.51 for eating more healthy foods; d = 0.29 for eating less unhealthy food |
| Vilà et al. 2017 | 12 studies (N = 3,323), reducing fat intake | d = 0.49 overall |
| Luszczynska et al. 2007 | RCT, 55 women in a commercial weight-loss program | 4.2 kg vs 2.1 kg lost over 2 months |

**The broad picture.** The founding meta-analysis pooled 94 independent tests covering health, study, and other goals. Forming an implementation intention produced an average effect of **d = 0.65** on reaching the goal, "medium-to-large" in the authors' words. It helped people start, stay on track despite distractions, and drop strategies that weren't working ([Gollwitzer & Sheeran, 2006](https://doi.org/10.1016/S0065-2601%2806%2938002-1)). The eating-specific syntheses below find smaller effects.

**Eating specifically.** A 2011 meta-analysis of 23 studies found **d = 0.51** for adding healthy items, such as eating more fruit, and **d = 0.29** for reducing unhealthy eating, such as eating fewer snacks. The authors added a warning: in the add-healthy-foods studies, weaker control conditions tended to produce bigger effects, so some of those effect sizes "may have been inflated" ([Adriaanse et al., 2011](https://pubmed.ncbi.nlm.nih.gov/21056605/)).

**Cutting fat.** A 2017 meta-analysis of 12 studies with 3,323 participants, varying in design, looked only at plans to reduce fat intake. It found a moderate effect, **d = 0.49**, larger than the authors expected for a "stop doing" goal. Two study-level moderators stood out: effects were larger in studies with more men, and in studies without professional monitoring (d = 0.67 vs 0.23) ([Vilà et al., 2017](https://pubmed.ncbi.nlm.nih.gov/28244173/)). One possible reading, not tested in the review, is that if-then planning adds the most when nothing else is providing structure.

## The weight-loss trial: planning doubled the loss in two months

Most of this research measures eating behavior, not body weight. One randomized trial measured weight directly. In it, 55 women with overweight or obesity who were already enrolled in a commercial weight-loss program were randomly assigned either to an implementation-intention prompt or to a control condition. Two months later, the planning group had lost an average of **4.2 kg**, against **2.1 kg** in the control group. The frequency of planning partly explained that difference ([Luszczynska et al., 2007](https://pubmed.ncbi.nlm.nih.gov/17605571/)).

The caveats: it was a small trial, lasted two months, included only women, and was an add-on to a program already running. It shows that planning can add to an existing effort. It doesn't show that planning alone produces weight loss, or that the extra loss lasts.

> The plan doesn't need to be clever. It needs to be decided before the moment arrives, so there is less to decide when you're tired and hungry.

## Why it works, and why the "then not" version can backfire

The proposed mechanism is that the plan links a specific situation to a specific response in advance. When the situation comes up, the response is already chosen. Gollwitzer and Sheeran found evidence that implementation intentions make the planned-for situation easier to notice and make the response more automatic ([Gollwitzer & Sheeran, 2006](https://doi.org/10.1016/S0065-2601%2806%2938002-1)). That's the same automaticity that everyday habits build through repetition. How long that takes on its own is the subject of our piece on [how long it takes to form an eating habit](/blog/how-long-to-form-an-eating-habit).

That mechanism also explains the most useful warning in this research. In a set of experiments, researchers compared plans phrased as a negation, "if I'm watching TV, then I won't eat chips," with plans that named a replacement behavior. The negation plans **backfired** in these experiments. They produced "ironic rebound effects" in both thinking and behavior, strengthening the link between the situation and the habit people wanted to break. The effect was largest when the habit was already strong. Replacement plans didn't show this problem ([Adriaanse et al., 2011](https://pubmed.ncbi.nlm.nih.gov/21177875/)).

One plausible explanation: the sentence "if TV, then not chips" still links TV to chips in your head. A plan that names what you *will* do links the situation to something else.

## How to write one for the moments that actually break a plan

The steps below are practical guidance drawn from the studies above, not a tested protocol.

**1. Find the situation, not the food.** Look back at a recent week and find where plans break. The usual culprits are predictable: late evening, arriving home hungry, the office snack table, weekends, meals with other people. The "if" part should name a moment you can recognize: a time, a place, a feeling, or an event.

**2. Name a replacement rather than a negation.** Write what you *will* do, not what you won't.

| Weak or risky | Stronger |
|---|---|
| "I'll snack less in the evening." | "If it's after 9 pm and I want something, then I'll make a cup of tea and sit with it for ten minutes." |
| "If I'm at a party, then I won't overeat." | "If I arrive at a party, then I'll get a drink and a plate of vegetables before I go near the buffet." |
| "If I'm stressed, then I won't eat chocolate." | "If I notice I'm stressed at my desk, then I'll take a five-minute walk around the building." |
| "I'll eat more protein." | "If I'm making breakfast, then I'll start with eggs or yogurt." |

**3. Start with adding.** The evidence is strongest for adding a healthy behavior. Your first plan should put something in: a protein source at breakfast, a vegetable with dinner, water before a meal. Cutting things out can come later.

**4. Keep it to a few plans.** A handful of plans aimed at your real trouble spots beats a long list you can't remember.

**5. Write the social ones in advance.** Parties and restaurant meals are where plans made in the abstract fall apart. Our guide to [tracking calories at parties](/blog/how-to-track-calories-at-parties) has concrete options you can turn into if-then sentences.

**6. Plan for the plan failing.** "If I eat more than I meant to at dinner, then I'll log it and decide only what breakfast looks like." That sentence keeps one miss from becoming a week off. Our article on [diet fatigue and adherence](/blog/diet-fatigue-and-adherence) covers why the recovery matters more than any single meal.

## What to expect

An effect size near 0.3–0.5 is useful but not dramatic. On average it improves follow-through; it doesn't make the difficult moments easy. If-then plans are a small tool that fits alongside the bigger things, like an eating pattern you can sustain, enough protein, and a deficit you can live with. They reduce how often a hard moment turns into a lapse. They don't remove hard moments.

## FAQ

### What is an implementation intention?

It's a plan in the form "If [situation], then I will [action]." Across 94 tests, forming one improved goal attainment with an average effect of d = 0.65 ([Gollwitzer & Sheeran, 2006](https://doi.org/10.1016/S0065-2601%2806%2938002-1)).

### Do if-then plans work for cutting out junk food?

Less well than for adding healthy foods. A meta-analysis of eating studies found d = 0.29 for reducing unhealthy eating, against d = 0.51 for eating more healthy foods ([Adriaanse et al., 2011](https://pubmed.ncbi.nlm.nih.gov/21056605/)). Plans to reduce fat intake did better, around d = 0.49 ([Vilà et al., 2017](https://pubmed.ncbi.nlm.nih.gov/28244173/)).

### Why shouldn't I write "if X, then I won't eat Y"?

In experiments, plans phrased as negations produced ironic rebound effects and could strengthen the habit they were meant to break, especially strong habits ([Adriaanse et al., 2011](https://pubmed.ncbi.nlm.nih.gov/21177875/)). Name a replacement action instead.

## Sources

- [Gollwitzer PM, Sheeran P. Implementation intentions and goal achievement: a meta-analysis of effects and processes. Adv Exp Soc Psychol. 2006;38:69-119.](https://doi.org/10.1016/S0065-2601%2806%2938002-1)
- [Adriaanse MA, Vinkers CD, De Ridder DT, Hox JJ, et al. Do implementation intentions help to eat a healthy diet? A systematic review and meta-analysis of the empirical evidence. Appetite. 2011. doi:10.1016/j.appet.2010.10.012](https://pubmed.ncbi.nlm.nih.gov/21056605/)
- [Vilà I, Carrero I, Redondo R. Reducing fat intake using implementation intentions: a meta-analytic review. Br J Health Psychol. 2017. doi:10.1111/bjhp.12230](https://pubmed.ncbi.nlm.nih.gov/28244173/)
- [Luszczynska A, Sobczyk A, Abraham C. Planning to lose weight: randomized controlled trial of an implementation intention prompt to enhance weight reduction among overweight and obese women. Health Psychol. 2007. doi:10.1037/0278-6133.26.4.507](https://pubmed.ncbi.nlm.nih.gov/17605571/)
- [Adriaanse MA, van Oosten JM, de Ridder DT, de Wit JB, et al. Planning what not to eat: ironic effects of implementation intentions negating unhealthy habits. Pers Soc Psychol Bull. 2011. doi:10.1177/0146167210390523](https://pubmed.ncbi.nlm.nih.gov/21177875/)

Source: BurnWeek — "If-then plans: the small planning trick that helps eating habits stick", https://burnweek.fit/blog/if-then-plans-for-eating-habits/. Licensed CC BY 4.0: free to quote or reuse with a link to this page.