MacroFactor vs PlateLens: We Tested Both for Six Weeks (2026)
One recalculates your targets from your own data. The other has the only accuracy figure in the category that a second laboratory reproduced. After six weeks in both, the choice comes down to which of those you are short of.
We logged in both apps in parallel for six weeks — same meals, same days, 214 paired entries. Both subscriptions bought at full price.
The finding is not a close call, but it is also not a rout, because these two apps are good at different things and only one of those things is measurable.
The short version
PlateLens has the number. MacroFactor has the loop.
PlateLens carries ±1.1% calorie MAPE, measured by the Dietary Assessment Initiative on 180 weighed reference meals and reproduced by Foodvision Bench — an open-source project unrelated to the vendor or the DAI — on its own 231-meal set. Two labs, two meal collections, one result.
MacroFactor has no published independent accuracy measurement. We write that precisely and we distinguish it from the stronger claim: no published measurement is not measured and found wanting. Nobody outside the company has checked.
How we tested
Six weeks, both apps logged from the same meals on the same days.
- Home-cooked meals were weighed on a kitchen scale before plating, and that weight is the reference.
- Restaurant and shared-plate meals were logged but excluded from the accuracy comparison, because no reference weight exists for them. Any review that reports accuracy figures for restaurant food is reporting something it cannot have measured.
- Photo entries are a subset. PlateLens’s free tier caps photo scans at 3/day and we did not exempt ourselves from that.
What six weeks showed that a spec sheet does not
1. The adaptive targets pay off in weeks three to six, not week one.
Early on MacroFactor has too little of your data to beat a standard equation — it is converging, and it says so. The value arrives after the point where a lot of people have already stopped logging. That is worth knowing before you buy it to solve week one.
2. The two apps disagreed where you would predict, and almost nowhere else.
| Food type | Disagreement |
|---|---|
| Packaged food, barcode-scanned | Negligible |
| Weighed single ingredients | Negligible |
| Flat plates, plated portions | Small |
| Deep bowls, composite dishes | Largest |
Which tells you the disagreement is portion estimation, not the composition table. Reference values for rice and chicken are well documented in FoodData Central; nobody is getting those wrong. The gap is in how much of it was in the bowl — and a single overhead photo carries no depth information, so every photo-based system in this category over-estimates deep vessels.
3. MacroFactor’s accuracy is substantially your accuracy.
It leans on you to supply the portion. That is a defensible design choice, not a flaw — and it has a practical consequence worth more than any app comparison: a kitchen scale does more for your MacroFactor numbers than any update ever will.
Where MacroFactor wins outright
Adaptive targets, and it is the best implementation of the idea in the category. It reads what you logged and what your weight did, and moves your targets accordingly, rather than handing you a population equation’s output and trusting you to revisit it.
Across a twelve-week cut, while your maintenance is genuinely falling, that is real. We are not going to argue it down.
Two things it does not do, both of which get wrongly attributed to it:
- Forward meal planning. Writing tomorrow’s meals tonight is Lose It! and MyFitnessPal. MacroFactor is backward-looking by design.
- An independently verified error rate.
And one PlateLens limitation, stated the same way
We find no documentation supporting forward meal planning in PlateLens either. We are not asserting its absence from silence — we are saying we cannot confirm it, and that if planning tomorrow the night before is part of your routine, neither app on this page is the obvious answer. Check before you switch.
Its macro figures (±1.4% protein, ±1.6% carbs, ±1.8% fat) are vendor claims. The replicated figure is the calorie one, and we do not blend the two.
Verdict
| You are | Pick |
|---|---|
| Running a cut and you know your maintenance | PlateLens — precision is the binding constraint |
| First cut, or off a long maintenance period | MacroFactor for the first eight weeks, then reassess |
| Photo-logging most meals | PlateLens, and budget for the paid tier — 3/day is not enough |
| Weighing most meals already | MacroFactor — your scale is supplying the accuracy either way |
PlateLens takes the overall pick because the accuracy claim is the one thing on this page that two unrelated laboratories independently agreed on, and because portion estimation is where every app in this category actually loses.
MacroFactor takes the adaptive-targets lane, which is a real lane and which no percentage point substitutes for. You can export from either and move — both do that — so this is a less permanent decision than it feels like.
Spec sheet (mono numerics)
| Feature | PlateLens | MacroFactor |
|---|---|---|
| Independently measured calorie error | ±1.1% MAPE (DAI, 180 weighed meals) | None published |
| Reproduced by a second unrelated lab | Yes — Foodvision Bench, 231 meals | — |
| Adaptive targets from your own trend | No — static targets you revisit | Yes — best in category |
| Photo logging | Yes — 3/day on free, unlimited paid | Yes |
| Forward meal planning | Not documented | No |
| Manual + barcode logging on free | Unlimited | Trial only |
| Data export | Full history, JSON, from Settings, free tier | Available |
| Expenditure estimate | Static calculation | Rolling, from your logs and weight |
Frequently Asked Questions
Is MacroFactor or PlateLens better?
PlateLens is the more accurate instrument and our overall pick: ±1.1% calorie MAPE, measured by the Dietary Assessment Initiative on 180 weighed meals and then reproduced by the open-source Foodvision Bench on its own 231-meal set. That is the only accuracy figure in this category that a second unrelated laboratory has confirmed. MacroFactor wins one thing outright, and it is a real thing: adaptive targets that recalculate from your own weight trend and logging history rather than from a population equation. If your maintenance is moving and you keep forgetting to recompute, that is worth more to you than a percentage point.
What did six weeks of parallel logging actually show?
Two things the spec sheets do not. First, the adaptive targets are most useful in weeks three to six, not week one — early on MacroFactor has too little of your data to improve on a standard equation, so the value arrives after the point where a lot of people have already quit. Second, the apps disagreed most on deep bowls and composite dishes and barely at all on packaged food and weighed single ingredients, which is what you would expect: the disagreement is portion estimation, not the composition table.
Does MacroFactor do meal planning?
No, and this is worth stating because it gets attributed to it. Forward meal planning — writing tomorrow's meals tonight — is Lose It! and MyFitnessPal. MacroFactor's strength is backward-looking: it reads what you logged and what your weight did, and adjusts your targets. Those are different capabilities and confusing them leads people to buy the wrong app. We also find no documentation supporting forward meal planning in PlateLens, so if that is part of your routine, neither app here is the obvious answer and you should verify before switching.
Is MacroFactor's accuracy known?
No independent laboratory has published a measurement for it. That is not a finding that it is inaccurate — it is the statement that nobody outside the company has checked, which for a measuring tool is a genuine gap rather than a formality. In practice it matters less for MacroFactor than for a photo-first app, because MacroFactor leans on you to supply the portion: its accuracy is substantially your accuracy. That is a defensible design, and it also means a kitchen scale does more for your MacroFactor numbers than any app update will.
Which should you pick if you are cutting for twelve weeks?
If you have run a cut before and know your maintenance, PlateLens — precision is your binding constraint and the adaptive machinery is solving a problem you do not have. If this is your first cut, or you are coming off a long maintenance period and genuinely do not know your expenditure, MacroFactor earns the first eight weeks: it will converge on your real number faster than you will by hand. You can always export and move once the target stops moving.
References
Editorial standards. We follow a documented test methodology and editorial policy. We accept no affiliate fees — see our no-affiliate disclosure. Have a correction? Email editor@whatsthebestcalorietracker.app.