The hope, unspoken but very common, is that there is a date. Get far enough in and the wanting stops, and eating becomes an administrative task. People who have maintained a large loss for years will tell you it does not work like that, and the research agrees with them, but with a more useful shape than simply saying no.
The short answer
Cravings decline substantially and they do not reach zero. A 2011 analysis in Obesity followed 270 adults randomly assigned to either a low-carbohydrate or a low-fat pattern for two years, measuring cravings and food preferences throughout. Cravings dropped in both groups. What they did not do was disappear, and what they dropped for turned out to be the interesting part.
What a two-year trial found
Two years is unusually long for this kind of measurement, which is what makes the study worth citing over the many six-week ones. 134 people were assigned to the low-carbohydrate arm and 136 to the low-fat arm. Cravings were tracked for several categories separately: sweets, high-fat foods, fast-food fats, and carbohydrates or starches.
The low-carbohydrate group showed significantly larger decreases in cravings for carbohydrates and starches, and in preference for high-carbohydrate and high-sugar foods. The low-fat group showed significantly larger decreases in cravings for high-fat foods. Each group's wanting followed each group's eating.
The decline is specific, not general
That is the finding worth carrying around. Cravings are not a fixed quantity of desire that you deplete over time. They attach to particular foods, and they loosen from the foods that stop showing up. The authors put it plainly: prescribing diets that restricted specific types of food resulted in decreased cravings and preferences for the foods that were targeted.
Which means the answer to "will I always want this" depends on what "this" is, and on what your eating actually looks like over a long stretch. Not on how much you want it today.
It also means the process is slower than a reset and more like a drift. Nobody wakes up on a particular Tuesday no longer interested in the thing.
Why the first weeks feel worse
People often report the opposite of a decline early on, and that is real. The 2012 Appetite study that recorded 393 craving incidents across a week found that women actively dieting to lose weight had more cravings than non-dieters, rated them stronger, and rated them harder to resist. Women who described themselves as watching their weight rather than dieting sat in between.
So the early phase and the long phase point in different directions, and both are documented. Restriction raises craving intensity in the short run. Sustained change in what you actually eat lowers it over years. The gap between those two timescales is where most people quit, having concluded from week three that it is only going to get worse.
What going away actually means
Ask a long-term maintainer and the description is rarely "I stopped wanting it." It is closer to: the thought arrives, and it does not come with the same pull, and it does not automatically become a plan. The craving still happens. Its authority has dropped.
That is a real change and it is worth naming properly, because measured against "gone" it looks like failure and measured against week three it looks like a transformation. If your food thoughts feel intrusive rather than merely frequent, obsessive food thoughts are a different problem with different answers.
Who ends up with the quietest version
The two-year data gives a hint that gets missed. The decline attached itself to what each group actually ate, week after week, for a long time. It did not attach itself to how hard anyone was trying, because trying was not the variable being manipulated. The people whose cravings fell furthest for starch were the people who had spent two years eating less starch.
That is worth stating bluntly because it removes a common source of despair. If cravings tracked effort, then a bad week would undo progress, and a person who found it hard would be permanently stuck. They track exposure instead. A rough fortnight does not reset the two years behind it, and a person who finds it hard gets the same adaptation as a person who finds it easy, provided the eating is similar.
It also sets a realistic floor. Anything you keep eating regularly, you will keep wanting regularly. That is not a warning, it is just the same mechanism running in the other direction.
Why they get louder again in maintenance
Cravings commonly pick back up when weight loss stops, and particularly for people coming off a medication that had been quieting appetite. The pattern people describe is not a slow return but a fairly sudden one, and it lands at the exact moment the external structure of active loss has been removed.
There is a structural reason this happens at exactly that point rather than randomly. Active weight loss supplies a lot of free scaffolding: the number moves, the feedback arrives weekly, the goal is legible. Maintenance removes all of it at once and asks for the same behaviour with none of the reinforcement. The cravings may not have genuinely increased so much as become more audible against a quieter background.
Nothing about that is a personal failing and none of it is a reason to change anything about a medication on your own. That decision belongs with your prescriber. What is squarely yours is the ordinary maintenance machinery: what actually turns food noise down, protein, sleep, and a structure that does not depend on wanting things less. If appetite signals themselves feel scrambled, hunger cues can be rebuilt.
How OffRamp tells a loud week from a real drift
A week of loud cravings feels like evidence, and it usually is not. This is the specific problem Regain Radar exists to solve. It reads your weigh-ins as a 7-day rolling average against a baseline, and it will not say anything at all until fewer than seven days of history have been filled in. Below that it reports stable, because it does not have enough to say otherwise.
Past that, the thresholds are fixed and dull on purpose. The rolling average has to sit at or above baseline times 1.02 for seven consecutive days before it calls drift, and baseline times 1.04 for seven consecutive days before it escalates. Seven consecutive days, not one bad Thursday. The design choice is that cravings are a terrible predictor of what your weight is doing, and the trend is a good one, so the app watches the thing that predicts and stays quiet about the thing that does not.
