MacroFactor alternatives

A macro tracker that works out what you actually burn and adjusts your target.

5 of its 6 core features can be rebuilt with AI. Here is one, working, below.

They makeRevenue estimate on Sensor Tower: $3M a month
You pay: $72 a year
AI can rebuild: 5 of 6 core features

Try it

Visual demo only. Copy the prompt below to get the app.

Build it yourself

Real files on your machine, a real database, and it is yours. This one carries the interface as well — the layout, the colours and the shape of the real app, read off its own screens.

Each opens with the prompt loaded and waiting — nothing is sent until you press enter. It is copied at the same time, in case the app is not installed.

Build me a local macro tracker I own, to replace MacroFactor.

- A web app I run on my own machine: a single Node process serving one page, data in a SQLite file next to it.
- Model it properly: foods with macros per 100g, entries referencing a food and a quantity, weigh-ins, and a targets row per week, so what it told me and why stays auditable.
- The food database is what the subscription really sells, and a version is downloadable: import the Open Food Facts dump, index it with FTS5, and resolve barcodes locally.
- Trend weight as an exponentially weighted moving average, robust to one daft weigh-in.
- Expenditure from my own data: intake over a rolling fortnight plus the energy implied by the change in trend weight, at 7,700 calories a kilogram. Fit across the window, and report nothing under ten days.
- Targets set weekly from that estimate and a chosen rate, each change stored with a reason.
- A day with no food logged is missing, not zero. Exclude it from the window or the estimate collapses.
- One command to export every entry and weigh-in to CSV, one to back up the SQLite file.
- Binds to localhost only. No accounts, no telemetry. The only network call is the one-off dump download.
- Skip: hand-verified branded nutrition. Somebody checks those labels and keeps checking them.
- Include a README with setup, how to import the dump, and where the database lives.

## Look and feel

MacroFactor's own — dark charcoal, sharp amber gold on the numbers that matter and the primary button, charts given real room, tabular figures. An instrument, not a lifestyle app. Generous spacing, clear hierarchy, big tap targets, phone-readable type, system font. Keep the text dark enough to read at a glance: the softness of a palette belongs to the shapes and the fills, never to the words. Nothing I need to read — dates, numbers, labels — in pale grey on white.

Did this prompt work for you?

More subscriptions die every week.

New apps and the prompts that replace them, as they land. One email, no spam.

Free forever · unsubscribe in one click · your address is never shared

Feature by feature

Every feature, whether AI can rebuild it, and what you would be signing up to keep running if you did. The paid app does that last column for you, and that is most of what the money buys.

Feature	Core	AI can rebuild	You'd maintain
Log food with calories, protein, carbs and fat	Yes	Yes	You would type the numbers off the packet the first time you eat anything
Daily totals against your targets	Yes	Yes	Nothing
Weight logging with a smoothed trend weightScale weight bounces two pounds a day on water alone. The smoothing that makes it readable is a weighted average.	Yes	Yes	Nothing
An expenditure estimate from your intake and your weight trendThis is the app's whole reason to exist, and it is arithmetic: what you ate, plus the energy the weight you lost must have been worth, at roughly 7,700 calories a kilogram. The hard part was deciding to do it, not doing it.	Yes	Yes	Nothing
Targets that adjust every week from the estimate	Yes	Yes	Nothing
A searchable database of branded foods with checked nutritionThe moat, and it is not code. Somebody has to check hundreds of thousands of labels by hand and keep checking them when the recipes change.	Yes	No	You would build your own list of foods, one packet at a time
Barcode scanningA barcode is only useful with a database behind it, and a page cannot reach one.	No	No	A phone can read the barcode; nothing tells you what the number means without the database behind it
Charts of weight, intake and expenditure over months	No	Yes	Nothing
Fibre and micronutrients beyond the big three	No	Partly	Only as good as the numbers you typed in

What you give up

the food database and the barcode scanner
years of logs backed up and searchable
the properly tuned version of the expenditure model
logging a branded meal in two taps

Why people still pay

The adaptive maths fits on a napkin: what you ate, plus what the weight you lost was worth in calories. What does not fit on a napkin is hundreds of thousands of food labels checked by hand, which is why every calorie app that ever mattered was really a database company.

Or use one somebody already built

Free and open source. No prompt, no building, no subscription.

Cronometer free tier: The most carefully sourced nutrition database of the free trackers, with real micronutrients rather than crowd-typed guesses. No adaptive target, but the data underneath is better. (web, ios, android)
Waistline: A free, open-source calorie and macro tracker that reads its food data from Open Food Facts, the community-maintained food database. No account and no subscription. (android)
Open Food Facts: The open food database itself, with barcodes and labels contributed by the public. It is the closest free thing to the part of MacroFactor you genuinely cannot rebuild. (web, ios, android)

Other apps people are freeing

Questions

Can AI really replace MacroFactor?: 5 of its 6 core features, yes. The demo above is what one prompt produced. What it cannot do is on this page too.
What does MacroFactor cost?: About $72 a year on the MacroFactor subscription plan.
Is there a free alternative to MacroFactor?: Yes: Cronometer free tier, Waistline, Open Food Facts. They are listed below, and they are somebody else's finished work rather than yours.