Protein does cost more energy to process than carbohydrate or fat. The measured ranking, from most to least, runs alcohol, protein, carbohydrate, fat, and it holds consistently across studies.
The size is the part worth knowing. Diet-induced thermogenesis across a normal mixed diet comes to about 5 to 15 percent of daily energy expenditure, and protein's advantage within that is real but not large enough to be the reason protein works.
The short answer
A 2004 review in Nutrition and Metabolism by Klaas Westerterp gathered the diet-induced thermogenesis literature and set out both the ranking and the scale.
Daily energy expenditure has three components: basal metabolic rate, diet-induced thermogenesis, and the energy cost of physical activity. Diet-induced thermogenesis, the middle one, is what gets called the thermic effect of food. For a mixed diet eaten at energy balance it comes to 5 to 15 percent of daily energy expenditure, higher when protein or alcohol intake is high and lower when fat intake is high. The macronutrient hierarchy in thermogenesis mirrors the hierarchy in how the body oxidises them after a meal.
Why protein costs more to process
Protein is the only macronutrient the body cannot store as protein in any meaningful reserve. Amino acids arriving from a meal are either incorporated into tissue, converted, or oxidised, and each of those routes has metabolic steps attached to it, including the work of handling the nitrogen that carbohydrate and fat do not carry.
Fat sits at the other end of the ranking for the mirror image reason. Dietary fat can be moved into storage with comparatively little processing, so relatively little of its energy is spent on handling it. This is a difference in overhead costs, not a loophole, and it is why the thermic effect appears reliably in careful measurements and still does not add up to much on its own.
Why the quoted percentages never match
Search this topic and you will find a different number on every page. That is not sloppiness on all of their parts. Westerterp's review is explicit that the measured value depends on the measuring conditions as much as on the food.
The conditions that move it include the nutritional status of the person being measured, how physically active they were during the observation, and how long the observation ran. The diet characteristics that move it are its total energy content and its macronutrient composition. Change any of those and the percentage changes with it, which is exactly why a single authoritative figure for the thermic effect of a given food does not really exist.
The consequence for anyone reading a food list that assigns a precise thermic percentage to each item: treat the ranking as solid and the decimal places as decoration. The order of alcohol, protein, carbohydrate and fat reproduces. The idea that a particular meal burns a specific and knowable share of itself does not survive contact with how these measurements are actually made.
The larger effect hiding behind the small one
Westerterp's review makes a point that gets skipped when this topic is reduced to a percentage: protein-induced thermogenesis has an important effect on satiety, and protein plays a key role in body weight regulation through that route.
A 2004 critical review in the Journal of the American College of Nutrition, from the Harvard School of Public Health, reviewed randomised investigations of high protein diets and reached a carefully graded set of conclusions. There is convincing evidence that higher protein intake increases both thermogenesis and satiety compared with lower protein diets. The weight of evidence also suggests high protein meals lead to reduced subsequent energy intake. On actual weight and fat loss, the authors were more cautious: some evidence suggests more of both, but findings have not been consistent.
Read that ordering carefully, because it is the whole practical answer. The best supported effects are what protein does to appetite and to processing cost. The least consistent is the number on the scale. Anyone citing the thermic effect as a weight loss mechanism has picked the weakest link in a chain whose stronger links were right there.
What that means in a normal week
If you swap some refined carbohydrate for protein, two things happen, and the smaller one gets all the attention. The processing cost of your diet goes up slightly. Your appetite at the next meal goes down measurably. The second effect is doing most of the work, and it is the one you can actually feel.
That is also why the practical advice from this research is not to chase thermogenesis. It is to get enough protein that the meals themselves hold you, which is a target you can hit without arithmetic. The related mechanism on the fibre side is covered in which kinds of fibre change appetite, and the question of where the grams should come from in whey against casein and the staples that make a floor easy to hit.
Why OffRamp has no calorie field at all
OffRamp does not track calories. There is no energy field in the app, no daily budget, and nothing that adds up food against activity. What it has is a single floor: 1.6 g of protein per kilogram of your goal weight, rounded to the nearest 5 g, raised by 10 percent while you are on a reset plan.
The research above is a large part of why the app is built that way. If the thermic effect were the main story, a calorie ledger with a protein adjustment would make sense, and the app would need to be precise about energy in order to be useful. But the better supported effect is satiety, and satiety is served by a floor you clear rather than by a sum you balance. A floor is also a number a person can hit on a bad day, which matters more for a habit that has to survive months than any modelling accuracy would.
So the log tracks grams and reports a hit rate: the days this week you cleared the floor, and your average when you did not. No target is ever set for energy, and that is deliberate rather than an unbuilt feature. Protein needs can differ for people with kidney disease and some other conditions, and that number belongs to your prescriber or a registered dietitian rather than to a default in an app.
