The targets · Decision 4 of 12

Why protein comes first

Protein is a target to reach, not a limit to stay under. The strongest reason is what your body keeps and loses while you are in a calorie deficit. The effect is smaller than most apps suggest.

Written and maintained by Mattias Geniar. Last updated .

What the app does
Shows protein as a target to reach, not a ceiling to stay under, and puts it above fat and carbohydrate on every screen.
How sure we are Firm , Pooled trials that agree with each other.
Higher-protein diets modestly preserved lean mass in short trials. −441 kcal/day came from just 19 people in a sequential feeding study, not a promised appetite effect. A two-year trial found no statistically significant protein-group weight difference.

The evidence for it

Three findings with different designs and populations. These are study averages, not expected results from using Snapkin.

−441 kcal/d average reduction in energy intake in a sequential study changing protein from 15% to 30%; all food provided Weigle et al. (2005) · 19 people, 12 weeks · ±63 kcal/d standard error; no parallel control
+0.43 kg lean mass kept rather than lost, on diets with the same number of calories Wycherley et al. (2012) · whole review: 24 trials, 1,063 people; outcome samples differ
+1.2 kg mean lean-mass gain with 2.4 g/kg/day and training six days/week, versus +0.1 kg at 1.2 g/kg/day Longland et al. (2016) · 40 men, four weeks, all food provided

What this looks like in the app is on the protein first page. This page has the research behind it.

What more protein actually gets you

Wycherley et al. (2012) is the a meta-analysis of energy-restricted, higher- versus standard-protein diets. It pooled 24 randomised trials with 1,063 people, all on diets with the same number of calories. Three of its four outcomes are in kilograms, so they can go on the same chart.

Higher protein against lower protein, at the same calories Whole review: 24 trials, 1,063 participants, mean duration about twelve weeks. Different outcomes use different subsets. Estimates and 95% confidence intervals.
no difference
Body weight −0.79 kg

95% CI −1.50 to −0.08. Under one kilo of scale weight over about twelve weeks

Fat mass −0.87 kg

95% CI −1.26 to −0.48. Separate pooled outcome; do not subtract it from the pooled weight change

Fat-free mass +0.43 kg

95% CI 0.09 to 0.78, kept rather than gained

−1.6 kg +1.0 kg
The fourth outcome is resting energy expenditure, which is in different units: +595.5 kJ/day preserved (95% CI 67.0, 1124.1), so it is not on this chart. The authors call the whole effect “modest”, and they are right. The gain is in what the lost weight is made of, not in how much is lost.

Their finding on fullness is weaker still: people felt fuller on higher protein in only 3 of the 5 studies that measured it.

Longland et al. (2016) is a tightly controlled proof-of-principle trial: 40 young men, four weeks, 2.4 vs 1.2 g/kg/day of protein during a 40% calorie deficit, with all food provided and supervised training six days a week.

+1.2 kg lean mass, at 2.4 g/kg/day against +0.1 kg at 1.2 g/kg/day
−4.8 kg fat mass, at 2.4 g/kg/day against −3.5 kg at 1.2 g/kg/day
40 young men, four weeks, all food provided The authors call it a proof-of-principle trial

It does not predict what happens to somebody logging food on a phone. The supervision, the training six days a week, and the food handed over are exactly the things a phone cannot give you. Lean mass also includes water and other non-fat tissue; it is not a direct measurement of muscle growth.

On fullness specifically

The trial the app quotes during setup is Weigle et al. (2005). Nineteen people ate a weight-maintaining diet at 15% protein for two weeks, then the same number of calories at 30% protein for two more, then twelve weeks at 30% protein with no calorie target at all. Carbohydrate was held at half of energy throughout, so the only change was more protein and less fat. Satiety rose sharply in the second phase, and in the third they ate 441 ± 63 kcal a day less (mean ± standard error) without being asked to.

Twelve weeks at 30% protein, eating as much as they wanted Weight lost during the twelve free-eating weeks, after the four weeks at fixed calories that came before them.
Body weight −4.9 kg

±0.5 kg standard error, over twelve weeks

Of which fat mass −3.7 kg

±0.4 kg standard error

0 6 kg
Both hormones moved in the direction that normally makes people hungrier: leptin fell and ghrelin rose. They ate less anyway. The caveats are serious: 19 people, no control group, phases in sequence rather than randomised, and every meal provided. It is a study of mechanism, not a diet trial, and one group of nineteen people is much weaker evidence than 24 pooled trials.

Kohanmoo et al. (2020) pooled many more trials on the same question: 49 short-term and 19 long-term, all at matched calories. The effect they found is smaller.

How much less hungry, on a 100 mm scale Pooled across 49 short-term trials at matched calories. People mark how hungry they feel on a 100 mm line.
Hunger, how far it fell ≈7 mm

Out of the full 100 mm the scale runs to

Fullness, how far it rose ≈10 mm

The larger of the two, and still a tenth of the scale

0 100 mm, the whole scale
Real, clearly measured, and small. The long-term trials, 19 of them, found no statistically significant pooled appetite effect.

The evidence against

No clear protein-group advantage at two years

POUNDS LOST (Sacks et al., NEJM 2009) assigned 811 adults to diets with 15% or 25% protein for two years: −3.0 kg vs −3.6 kg, p > 0.20. “Satiety, hunger, satisfaction with the diet, and attendance at group sessions were similar for all diets.” What did predict weight loss was turning up: 0.2 kg per group session attended.

Assigned protein levels and weight loss after two years Weight lost at 24 months, by assigned protein percentage; actual intakes did not fully match the prescribed diets.
15% protein −3.0 kg

Two years, 811 adults

25% protein −3.6 kg

p > 0.20 against the bar above. Neither bar is drawn in the page colour, because neither group won

0 5 kg lost
The gap is 0.6 kg and it is not statistically significant. Attendance was: 0.2 kg per session. Attendance was an association, not a randomized treatment dose. A non-significant diet comparison does not prove that protein never matters.

Adding a protein drink can add net calories

Ben-Harchache et al. (2021), 22 studies and 857 older adults: a protein drink before a meal cut that meal by 164 kJ. But once the drink's own energy was counted, total intake was higher by 649 kJ (95% CI 438, 861). Protein helps when it replaces other food at the same calories. Adding a drink did not reduce total intake in these older-adult studies; this does not settle every meal or population.

Too little protein may matter more than extra protein

Gosby et al. (2011) found that cutting protein from 15% to 10% of calories made people eat 12% more. Raising it from 15% to 25% did not significantly change how much they ate. That suggests the goal is to avoid too little protein, not to eat as much as possible. Martens et al. (2013) found the opposite. The two disagree, and that is the real state of the evidence.

And a caveat on our own number

The “about 30 g of protein per meal” rule in the recipe library comes from Schoenfeld & Aragon (2018), which recommends 0.4 g/kg per meal across at least four meals. That is a narrative review about muscle growth in people who lift weights. It is not a meta-analysis, and it is not about fullness or weight loss. Its lead author also declares an advisory role with a protein supplement company. We use it because a defensible minimum is better than no minimum, not because the question is settled.

For the actual protein multipliers used by loss, gain and maintenance, including the reference-weight cap and why protein alone cannot promise muscle growth, see the calorie-target audit.