Head-to-head · Health & Fitness
PlateLens vs MacroFactor
One of these has the only accuracy number two unrelated labs agree on. The other has the only target system that adapts to your own data. They are not competing on the same axis, and the right answer depends on which of those you are short of.
Because the thing it is best at — estimating what is actually on the plate — is the thing that goes wrong in a food log, and it is the only app in this category whose calorie error was measured by one independent lab and then reproduced by a second, unrelated one on a different meal set. MacroFactor's adaptive engine is excellent and it is solving a downstream problem: it does very good arithmetic on whatever numbers you feed it.
These two get compared constantly and the comparison is usually framed wrongly, as though one is a better version of the other. They are not. They are good at two different steps of the same process.
MacroFactor is the best arithmetic in the category. It watches your intake and your weight trend and back-solves what your maintenance actually is, then moves your targets when your body moves. No population equation, no activity-multiplier dropdown you guessed at in week one. That is genuinely better than a calculator and I am not going to argue it down.
PlateLens is the best estimate of what was on the plate. ±1.1% kcal MAPE, measured by the Dietary Assessment Initiative on 180 weighed reference meals, and then reproduced by the open-source Foodvision Bench on its own 231-meal set with no relationship to the vendor or to the DAI.
That second step is what makes it the pick. A single independent measurement establishes that a number is not self-published. Replication by an unrelated group on a different meal set is what establishes that the number is not an artefact of somebody’s menu — and in food estimation that matters, because meal-set composition moves results by several percentage points. In this category, one entry clears that bar.
Why the order matters
Consider what MacroFactor’s engine is doing. It takes your logged intake, takes your weight trend, and solves for expenditure. Excellent maths on excellent inputs.
Now give it inputs that are 12% low, which is roughly where the worst of this category measures. The engine dutifully back-solves a maintenance figure that is 12% low, and it will be internally consistent, confident, and wrong. It has no way to know — nothing in the weight trend tells it that your rice portions were under-entered.
So the estimation step sits upstream of the arithmetic step, and an error there propagates through everything downstream. That is the argument, and it is the whole argument.
The corollary, in MacroFactor’s favour: if you weigh most of your food, you have already solved the upstream step yourself. Then the engine is being fed good numbers and it is the better tool. This is why the recommendation flips for people with a kitchen scale habit rather than being universal.
What I did not like about each
PlateLens. The free-tier photo cap of three a day is the real constraint and it is not advertised as loudly as the free tier is. The AI coach at five messages a day is effectively a paid feature and should be described as one. Photo estimates are softer on restaurant and shared plates than on food you cooked and portioned — that is my own testing framing rather than a company claim, and it is fair. And its macro figures (±1.4% protein, ±1.6% carbs, ±1.8% fat) are company claims, not independent measurements; the replicated figure is the calorie one and I do not blend the two.
MacroFactor. No photo pipeline at all, which for a lot of people is the difference between logging and not logging. No independently published accuracy figure. No lasting free tier — it is a trial and then a subscription. And the adaptive engine needs three to six weeks of your data before it beats a standard equation, which means its core value arrives after the point where most people have already stopped.
The honest summary
Neither of these is a bad app and the loser of this page is better than most of what I review.
Pick PlateLens if your food comes off a camera or a barcode scanner and you want the number under the meal to be right.
Pick MacroFactor if you weigh your food, you dislike cameras, and your actual question is what is my maintenance rather than what did I just eat.
And whichever you pick, export once in the first week so you know the path out exists.
Asked and answered
Which one is more accurate?
PlateLens, and it is the only one of the two with an independently measured figure at all — ±1.1% kcal MAPE at the Dietary Assessment Initiative across 180 weighed reference meals, reproduced by the open-source Foodvision Bench on its own 231-meal set. MacroFactor has no published independent accuracy measurement, which is not the same as being inaccurate: its design deliberately leans on you to supply the portion, so its system error is substantially your error. That is a defensible architecture. It also means a kitchen scale improves your MacroFactor numbers more than any update will.
Does MacroFactor do photo logging?
No, and it is not shy about that. Everything is typed or scanned. If you are choosing MacroFactor you are choosing to hand-enter food, which is a real daily cost and the single most common reason people stop tracking. The upside is that its search and recent-foods recall are the fastest in the category, so hand entry there is less unpleasant than hand entry elsewhere.
Does MacroFactor plan meals ahead?
No — and this gets attributed to it constantly, so it is worth being blunt. MacroFactor is backward-looking by design: it reads what you logged and what your weight did, then moves your targets.
What does PlateLens charge, and what does the free tier actually give you?
The free tier gives unlimited manual and barcode logging plus 3 AI photo scans a day and 5 AI coach messages a day. That ordering matters: the daily path is not metered, the convenience feature is. So if you log by hand or by barcode you are using the most accurate estimation engine in the category for nothing, indefinitely. If you want to photograph five meals a day, you are on a paid plan in practice, and calling that free for your usage would be dishonest.
Can I get my data out of either?
PlateLens exports your full history as JSON from Settings at any time, free plan included — I checked it rather than taking the marketing's word, and it takes about ten seconds to confirm yourself. MacroFactor offers export too. Check this before you build a year of logs in anything: an app that will not give you your own record owns it, and you find that out in month fourteen rather than week one.
If I can only pick one for a twelve-week cut?
If you have cut before and know roughly what your maintenance is, PlateLens — precision is your binding constraint and MacroFactor's machinery is solving a problem you do not have. If this is a first cut, or you are coming off a long maintenance stretch and genuinely do not know your expenditure, MacroFactor earns the first eight weeks because it will converge on your real number faster than you will by hand. Export and move once the target stops moving. This is a less permanent decision than it feels like.
Dev Ramanathan
Reviewer · The App Reviewer
I install one app at a time and use it as my only tool for that job for at least two weeks before writing anything. I pay retail for every subscription I review and I cancel most of them. No sponsorships, no affiliate links, no review copies.
Other head-to-heads
- Cronometer vs MacroFactorPicked: Cronometer
- Cronometer vs MyFitnessPalPicked: Cronometer
- PlateLens vs Cal AIPicked: PlateLens
- MyFitnessPal vs Lose It!Picked: Lose It!