A MyFitnessPal alternative for people who actually cook

The database problem, stated fairly

A food database is the correct design for packaged goods. A tin of beans has a label; the label has exact numbers; searching for that tin and reading its numbers is precise and quick.

The design breaks down when the food has no label. To log a home-cooked chilli in a database app you either search for “chilli con carne” and accept a stranger’s guess at what that means, or build a recipe by entering every ingredient by weight. The first is fast and inaccurate; the second is accurate and slow enough that people stop doing it.

There is a second issue specific to large user-contributed catalogues: entry quality varies. Multiple entries for the same product with different values are common, and choosing between them is guesswork the interface presents as a lookup.

Approach by approach

When you should stay with a database app

This is not a page arguing that database trackers are obsolete. Keep one if:

• Most of your food comes out of packets and you scan barcodes already — you are getting exact numbers, and no estimate improves on exact.
• You need detailed micronutrient tracking for a medical reason. Photo estimates do not give you iron, B12 or potassium with any confidence.
• You have years of history in one app and value the continuity more than the time saved.
• You are on iPhone. Caloris is Android-only today.

What changes when logging takes three seconds

The interesting effect of fast logging is not the time saved directly — a few minutes a day is not life-changing. It is what happens to the completeness of the log. When logging is slow, people skip the meals that are hardest to log, which are disproportionately the large, calorie-dense, eaten-out ones. The resulting log is not just smaller; it is biased low in exactly the places that matter.

A method fast enough to use when you are tired produces a more honest picture, even if each individual entry is less precise. That is the real case for an AI calorie counter.

Switching over

1. Export your existing history if the option exists — trackers do not import each other’s logs.
2. Run both apps for a week on the same meals, and pay attention to how each handles your hardest dishes.
3. Recalculate your targets rather than copying them across: the calorie calculator and macro calculator take a minute between them.
4. Log your ten most frequent meals early so the new app becomes fast for you within days.

For a wider view of the category, see the guide to choosing an AI calorie counter app.