Kriterion vs Nutracheck
This is the comparison a British person actually makes. Nutracheck is a UK company with a UK food database. Kriterion was built for British food too, with UK own-brand and chain entries badged by source. The difference that matters most is what happens after you log it.
| Kriterion | Nutracheck | |
|---|---|---|
| UK food data | UK government lab data (CoFID), plus UK chain menus and supermarket products, each entry marked with where it came from | A large, verified UK database with supermarket own-brands and chain menus. Their strongest suit, and genuinely theirs |
| Product photographs | Not offered | Yes, on a large share of the database. A useful touch |
| How your calorie target is set | Measured from your logged intake against your weight trend. No activity level to pick, because it is derived rather than declared | Estimated formula: resting metabolism from your stats, times an activity multiplier you choose at signup |
| Target adapts to your results | Continuously, weighted towards your recent weeks of weigh-ins | Recalculates when your weight or your stated activity level changes; the multiplier itself is still the one you picked |
| Shows its error margin | Yes, as a spread either side of the figure, so you can see how firmly the number is pinned down | Not published. The target is presented as a figure |
| Checks its own past predictions | Yes. Replays what it believed across your history against what the scale did, and tells you the gap | Not offered |
| Barcode scanning | Yes | Yes |
| Salt and fibre tracked | Yes, alongside the macros and micronutrients | Yes |
| Strength training | Built in: sets, reps, RIR, readiness, a HIIT timer with spoken cues | Exercise is logged as calories burned rather than as training |
| Menstrual cycle tracking | Built in, shown beside readiness as context for training | Not offered |
| Free tier | Logging is free and unlimited, permanently, with no card required | A free tier exists with limits on the diary; the full app is a subscription |
| Platforms | iOS, Android, Apple Watch, Wear OS | iOS, Android, Apple Watch |
Why the activity multiplier is the whole argument
Every formula-based calorie app asks you the same question near the start: how active are you? Sedentary, lightly active, moderately active, very active. Your answer picks a number, usually somewhere between 1.2 and 1.9, and your resting metabolism is multiplied by it. That product is your maintenance calories, and every target the app ever gives you is built on it.
The trouble is that the gap between the bottom and the top of that scale is enormous. On the same person, the difference between “lightly active” and “very active” can be six or seven hundred calories a day. Nobody knows which one they are. Desk workers who go to the gym four times a week pick different answers for identical lives, and people who are dieting tend to pick optimistically. You are being asked to supply the single most important input, about yourself, before you have any evidence.
Then, crucially, nothing checks it. If you picked wrong, the target is wrong by that much for as long as you use the app, and the only signal you get is that the weight is not moving the way it said, which most people read as a personal failure rather than as a wrong multiplier.
Kriterion never asks the question. It watches what you logged and what the scale did, and solves for maintenance from those two things. If you eat 2,100 a day for three weeks and lose half a kilo, your maintenance is not a category you selected; it is arithmetic. That is the whole of the difference, and it is also why the food database has to be good: the measurement is only as honest as the logging underneath it, which is the one thing Nutracheck has always understood.
What we will not claim
We do not have the only UK food database, and it would be a lie to imply it. That claim gets made a lot in this category and Nutracheck is the standing refutation of it. What we have is UK government lab data, UK chain menus, and a source label on every entry so you know whether a number came from a laboratory, a manufacturer or a stranger. What we have that they do not is the engine that turns your logging into a measured figure with a stated margin of error.
We are also not claiming a bigger UK grocery catalogue. Theirs has years of packaged groceries behind it and photographs that help you log the right thing, and it is the benchmark ours is being built against. What we claim is the source label on every entry, so you can see how each number was produced.
And the measured approach has a real cost. It needs a few weeks of consistent logging and regular weigh-ins before it says anything confident, and it tells you so rather than pretending otherwise. You still get a number on day one: until your own data has earned the weight, Kriterion leans on the same standard formula and labels the figure as still calibrating. If you rarely weigh in, it has less to measure from, and the margin beside the figure says so.
What the ± is for
One number hides how much of a guess it is. “1,870” looks identical whether the app has three months of your data or three days. Kriterion shows the spread as well, so a tight range tells you the weigh-ins agree with each other and a wide one tells you there is too much scatter to act on yet. Under the hood it is a 95% confidence interval on the expenditure estimate itself, taken straight from what the engine currently believes rather than approximated from a fitted window, which is why it widens on its own through a gap in your logging, where an interval drawn around a trend line would have held still.
What it does not claim: a tight spread means your scale data is consistent, not that the number is true. The larger error in any app like this is under-reported food, which the tracking literature puts at 20% or worse. That is a bias rather than random scatter, so no honest interval can include it. We would rather say that than hand you a range that quietly implies more than it knows.