snapkin
The research behind Snapkin
Snapkin analyses what you have eaten and puts a number on it. You should be able to check where that number comes from. So this is where we write down our reasons.
Last updated .
16 decisions, 4 questions, 161 sources. Every one of them linked.
None of this is medical advice. Snapkin is a general wellness tool. It does not diagnose, treat, prevent or monitor any disease. It is not a medical device. The terms say what that means in practice.
Question 1 of 4
The camera
What a picture of a plate can tell you, and what it cannot.
How close Snapkin gets to food on a scale
Snapkin against 500 meals weighed on a scale: how close a single photo gets, where it does best, and what we have not tested yet.
83% · 90% median accuracy per meal and over a simulated week, on phone photos; 80% and 89% across all sets
Why we think a photograph is good enough
Why a photo is enough to start from, what other people measured, and why knowing the portion moves the number most.
88% median accuracy on what the food is, from the photo alone: calories per gram within 12% of the weighed food on phone photos; 87% across all sets
Why we track added sugar, and what we mean by it
Why free sugars are not the same as the US added-sugars label, and why tests of experts on packaged food say nothing about photos.
10% of your calories: the WHO’s limit on free sugars, and where Snapkin’s daily sugar budget comes from
Question 2 of 4
The targets
What the app does with the number once it has one.
How we calculate your calorie target
The sums for losing, gaining and keeping your weight, the evidence behind them, and why eating extra cannot promise kilograms of muscle.
0.5% a week of your body weight: the default pace for losing, always kept inside a safe range of 0.25–0.8%
Why protein comes first
What eating more protein does to appetite, what it protects while you are losing, and where the effect stops.
−441 kcal/d fewer calories eaten a day on average when protein went from 15% to 30% of the diet
Why your targets change at 65
From 65 the app sets one protein amount for everybody, a smaller deficit and a higher lowest goal weight, and it may suggest keeping your weight. Why, and where you can still choose against the advice of geriatricians.
1.2 g/kg of protein a day from 65, with or without strength training: the top of the range experts advise for healthy older people, and under the 1.3 g kidney limit
Why your steps and workouts do not change your calorie target
Most other trackers add your workout to what you may eat. The review behind that choice, and the fair complaint against it.
6 of 6 studies found wearable calorie estimates more than 20% off in at least one test, so your target does not follow them
Why the meal it picks for you depends on where you are
Meal times come from national time use diaries, not from one country’s habits. So lunch is set for when your country eats. And what that still cannot know about you.
23 countries of real time use diaries decide when the app calls it lunch
Where your fibre, salt and mineral targets come from
Europe and the US publish different daily targets for fibre, salt and minerals. For each one we used the number with the better evidence for your health.
5 g of salt a day at most, the European and WHO limit; the US allows 5.75 g
Question 3 of 4
The habit
Whether logging food is worth doing at all, and who it is bad for.
Why track anything at all
What logging is worth in kilograms, why the app asks for a photo instead of a database search, and the honest limits of the evidence.
−2.87 kg more weight lost with self-monitoring than without, pooled across 12 randomized trials
Why we read your week back on Monday
Why the Monday report reads your week as one total, expects a higher weekend, counts the days you logged, and leaves your weight out.
−6.4 vs −5.6 kg lost with two very low days a week against a smaller cut every day, the same weekly deficit, in a 6-month trial
Tracking and eating disorders
This page argues against our own product. What the link between tracking and eating disorders really is, what the experiments can and cannot show, and what the app refuses to do.
No increase in eating-disorder symptoms after a month of calorie tracking, compared with no app, in a randomized trial
Rewards, and what we will not give one for
Nothing here can be earned faster by eating less. The studies behind that rule, and what the app counts instead.
−0.70 kg more weight lost on average when health apps add game elements, across trials
How much setup is worth asking for
Which setup questions change the plan, what can wait, and the evidence for testing a shorter introduction.
12% easier how much easier the app felt to people who skipped the tutorial, in a study of 70; the case for a short setup
Question 4 of 4
The scale
How fast weight should move, and how to read a scale that jumps around.
How fast the app will let you lose weight
Why the pace stops at 0.8% of body weight a week. The small trial behind it, where the slower group gained more lean mass, and why the default sits well under the limit.
+2.1% more lean mass, not less, in athletes losing 0.7% of their weight a week; Snapkin’s pace stops at 0.8%
Why the weight chart draws a line you did not measure
The heavy line is not your weighings. What one morning on a scale really tells you, and why smoothing cannot remove every reason the scale moves.
0.5% of body weight is normal day-to-day noise; the trend line smooths it out, so one heavy morning is not a setback
How these pages are maintained
These pages are in English only, on purpose. Machine-translating a medical claim into ten languages is ten chances to get one wrong.
Who writes them, and what the company behind them is registered as, is on the about page. Every decision on these pages has a feature page describing what it does in the app: Photo logging, Repeat meals, Protein first, Widgets and watch, It learns you, Honest numbers, Weight tracking, Monthly check-in, Your language, your food.
Every figure here was read from the paper or its abstract, not remembered. Where we could only reach an abstract, the page says what the abstract says and nothing more. Where a claim rests on a preprint or on one small study, it is labelled. If you find something wrong, tell us at hello@getsnapkin.app and we will correct it here.