This is not medical advice. General information about food tracking tools, not a treatment plan. If you are under 18, pregnant or breastfeeding, have ever had an eating disorder or a difficult relationship with food, or are managing a medical condition, talk to a doctor or a registered dietitian before you start counting calories or change your intake. The full list is here.

The minute nobody budgets for

Every calorie counter app asks you for the same thing, and it is never the thing advertised on the download page. What it wants is a minute of your attention, four or five times a day, for as long as you intend to keep going.

Somebody has measured this. In a 24-week online weight-control programme, 142 participants logged their food through a web-based diary while the software recorded how long they spent doing it. In month one the average was 23.2 minutes a day. By month six it was 14.6, and only 65.5% of them were recording any minutes at all.1

Twenty-three minutes a day is not a small ask. It is a commute, or a phone call to your mother, and it is being spent on data entry. Spread across four eating occasions, it is close to six minutes per meal.

Here is the part of that study that should have changed how these apps are built. Among the people still logging at six months, time spent made no significant difference to weight loss. What did make a difference was how often they opened the diary: people who lost 5% or more of their body weight logged in 2.4 times a day against 1.6 for those who lost less, and the gap for 10% was wider still.1 How often people logged predicted the outcome. How long they spent doing it did not.

Which makes the minute overhead rather than effort. A basic calorie counter app that reaches the same number in fifteen seconds has removed the overhead and kept everything the study found useful.

Where the minute actually goes

Pick up a typical food logging app and log a bowl of porridge with berries and a coffee. Count the decisions.

You open the app and choose a meal slot, because it wants to know whether this is breakfast or a snack. You search "porridge" and get back forty results, most of them user-submitted, several of them wrong. You pick one. It defaults to a serving size of "1 container" and you have to change it to grams, then guess the grams. Then you do the berries. Then you do the coffee, and decide whether the splash of milk is worth a third entry. Then you tap save.

The barcode scanner is sold as the fast path and it is not, quite. It gets you a correct product in about two seconds and then hands you straight back to the portion question, which is the slow part, because the packet holds 500g and you ate some unmeasured fraction of it. A scan-and-adjust flow is really a two-second win wrapped around a thirty-second decision.

None of these steps is unreasonable on its own. That is exactly why they accumulate. Each one was added by somebody who could justify it, and nobody was measuring the total.

A useful test for any calorie logger: time yourself logging one ordinary meal, from unlocking your phone to putting it down. Not the meal the marketing screenshots show. The one you actually ate, with the ingredient you are not sure about. If it runs past thirty seconds, ask what you got for the extra time, and whether you will still be paying it in March.

Friction is what ends it, not willpower

The standard explanation for quitting is motivation, and it is mostly wrong. When researchers ask people directly, they name the time.

A Dutch mixed-methods study ran focus groups on six commercial nutrition apps and then surveyed 1,420 people. Among those who had abandoned an app, the single most common reason given for quitting early was the high time investment, at 14 of 38 responses.3 It came up in the focus groups too, before anyone was offered it as an option on a form.

You can also see it in what people keep doing versus what they stop doing. A six-month trial gave 54 adults three things to self-monitor: their weight on a connected scale, their steps on a Fitbit, and their food in an app. The median time before they disengaged from step tracking was 19.5 weeks. For food logging it was 10 weeks, and for weighing, 4. The authors put the pattern down to burden: physical activity tracking is the one that happens whether you participate or not, and it is the one people sustained.2

That study is small, and it was not designed to prove that burden causes disengagement, only to describe the pattern. But it also found that consistency and time-to-disengagement both predicted weight loss for food logging.2 Put the two together and the friction stops being a usability annoyance sitting alongside the outcome and starts being one of the things that determines it. The logging works. The friction ends the logging.

This is a different claim from the one I made in the piece on four weeks of tracking, and worth separating. That one is about the fact that persistence takes longer than people expect. This one is about the fact that a tool can make persistence cost more than it needs to, and most of them do.

More features do not make a better calorie logger

The intuition behind a maximal app is that every feature is optional, so more of them can only help. Somebody checked.

A 2022 systematic review pooled 34 randomised trials of smartphone apps for weight loss and ran a subgroup analysis on what the apps actually did. Apps beat controls, by about 2 kg at three months and 2.8 kg at six. Then the interesting line: neither the type nor the number of app features was associated with weight loss. What was associated with better results was a human being in the loop, coaching or giving feedback.4

The heterogeneity in that review is high, so read the pooled numbers loosely. The feature finding is the one I would hang something on, because it is a null result across 34 trials rather than an effect size from a few.

An older and much smaller trial gets at the same thing from the other side. Forty-two adults in a 16-week programme were randomised either to a detailed paper food record or to a simplified, abbreviated one. The abbreviated group returned significantly more diaries. Weight loss was effectively identical, 7.5 kg against 7.6 kg among completers, and weight loss tracked the number of diaries completed rather than how detailed they were.5

Forty-two people, sixteen weeks, and a paper diary in 2007. I am not going to pretend that is a definitive result about smartphone software. But it points the same direction as the 2019 minutes data and the 2022 feature analysis, and I do not know of a trial pointing the other way.

What a simple calorie tracker app looks like

Strip the argument down and you get a specification. A simple calorie tracker app is one where the cost of logging a meal is low enough that you will still do it on a bad day, and everything else is subordinate to that.

In practice that means a short list.

  • Logging a meal is one decision, not five. If the app needs to know the meal slot, the serving unit, the brand and the portion before it will give you a number, it has four opportunities to lose you.
  • The default path is the fast path. Whatever the app does when you tap the obvious button should be the thing you do forty times a week, not the thing you do once during setup.
  • Nothing blocks the log. An estimate you can correct later beats a modal asking you to be precise now, because the modal is where people put the phone down.
  • Editing exists and is cheap. This sounds like it contradicts the point, and it is actually what makes speed safe: if a wrong number is fixable in five seconds, the app can afford to guess.
  • The screens you see daily outnumber the ones you configure. Anything you touch once at signup does not get to sit in the main navigation.

Notice what is not on that list. Not a bigger database, not a barcode scanner, not recipe import, not a macro breakdown of every micronutrient. Those are all real features that some people genuinely want. None of them is the thing that decides whether you are still logging in eight weeks.

How MyPlate handles this, and what it gives up

I build a calorie app, so treat this section as interested rather than neutral. What I can do is tell you exactly what the trade is, which is harder to get from a download page.

MyPlate logs a meal from a photograph. You tap the plus button, the camera opens, you take the picture, and the result screen appears immediately with the photo in place and the numbers filling in behind a loading state. There is no search box, no meal slot, no serving-unit dropdown and no barcode scanner, because there is no food database underneath. An AI model looks at the picture and estimates calories, protein, carbohydrate and fat. Four taps, and most of the wait is the model thinking rather than you deciding.

The clearest example of what this costs to build is a screen I deleted. The camera path used to go through a confirmation view: take the photo, then look at it again on our screen, then press "Add dish". It offered no crop, no portion field and no note, so it was asking you to approve the same photograph twice. Removing it took the camera path from seven interactions to four. Nobody asks for that in a review, and it is worth more than most of what people do ask for.

Now the honest part. A review of image-assisted dietary assessment methods puts the trade plainly: image-based capture removes the need for the user to estimate portion size, which is the step that eats the minute, and the reviewers also conclude that less burdensome methods are less accurate and that fully automated food recognition with acceptable precision is not yet a reality.6 That is a fair description of what you get from a photo scan, mine included.

So a scanned meal is an estimate, not a measurement. If you weigh your chicken breast and look it up in a curated database, you will get a closer number than my app gives you. I have written separately about how accurate calorie tracking apps actually are, and the summary is that every method in this category carries error large enough to matter, including the careful ones.

The bet is the one the evidence above supports: a rougher number you log every day beats a precise number you log for eleven days. If you are the kind of person who will weigh and search and stay consistent for a year, use a database app. Most people are not, and most people know which they are.

Common questions

What is the simplest calorie counter app?

There is no single answer, but there is a test you can run in ten minutes. Install two candidates, log the same real meal in each, and time yourself from unlocking the phone to putting it down. Count the decisions the app made you make. The one that got to a usable number in fewer steps is the simpler app for you, regardless of which one has more features on its store listing. Published trials have found no association between the number of features an app has and how much weight its users lose, so a shorter feature list is not evidence of a worse tool.

Do I need a barcode scanner to track meals?

No. A barcode scanner identifies a packaged product quickly, then leaves you with the portion question, which is the slow and error-prone half of the job. It also does nothing at all for a plate of food that never came in a package, which is most home cooking and every restaurant meal. It is a convenience for people who eat a lot of packaged food, not a requirement for tracking.

How long should logging a meal take?

Aim for under thirty seconds. For scale, participants in a 24-week web-based weight-control programme averaged 23.2 minutes a day on food logging in the first month, falling to 14.6 minutes by month six. Among those still logging at six months, time spent had no significant relationship with weight loss, while how often they opened the diary did. The minutes are a cost, not a benefit, so spending fewer of them costs you nothing that the research can detect.

Is a simple calorie logger less accurate than a detailed one?

Usually yes, and it is worth being clear about that. A review of image-assisted dietary assessment concluded that less burdensome methods are less accurate, and that fully automated food recognition with acceptable precision is not yet a reality. So a photo estimate is genuinely rougher than a weighed portion looked up in a curated database. The argument for the simple tool is not that it is more accurate. It is that a rougher number recorded consistently is worth more than a precise number recorded for two weeks.

Will I actually lose weight with a basic calorie counter app?

The evidence says the detail level is not what decides it. In a 16-week trial, participants randomised to a simplified food record returned significantly more diaries than those using a detailed one, and lost effectively the same amount of weight, 7.6 kg against 7.5 kg. Weight loss tracked how many records people completed rather than how detailed those records were. That trial was small, 42 people, so treat it as one piece of evidence rather than proof. No app of any kind removes the need for a calorie deficit you can sustain.

Log often, not thoroughly

If you have quit a calorie tracker before, it is worth considering that the app was asking you for something other than what you signed up for. Not discipline. Time, in small unremarkable amounts, several times a day, until one Tuesday you did not feel like it and never went back.

The research is fairly consistent on what to do about that. Log often rather than log thoroughly. Pick the tool that gets you to a number in the fewest decisions, and let it be approximately right. If an app is making you anxious about food, or the logging has started to feel compulsive rather than useful, stop and talk to your doctor or a registered dietitian. That is the right response, not a different app.

And if you build these things, the feature worth shipping this month is probably a screen you can delete.

Sources

Every figure above traces to one of these. If you find a number that doesn't match the source it claims, tell me and I'll correct it.

  1. Harvey J, Krukowski R, Priest J, West D (2019). Log Often, Lose More: Electronic Dietary Self-Monitoring for Weight Loss. Obesity, 27(3):380-384. Read on PubMed →
  2. Carpenter CA, Eastman A, Ross KM (2022). Consistency With and Disengagement From Self-monitoring of Weight, Dietary Intake, and Physical Activity in a Technology-Based Weight Loss Program: Exploratory Study. JMIR Formative Research, 6(2):e33603. Read on PubMed →
  3. van der Haar S, Raaijmakers I, Verain MCD, Meijboom S (2023). Incorporating Consumers' Needs in Nutrition Apps to Promote and Maintain Use: Mixed Methods Study. JMIR mHealth and uHealth, 11:e39515. Read on PubMed →
  4. Antoun J, Itani H, Alarab N, Elsehmawy A (2022). The Effectiveness of Combining Nonmobile Interventions With the Use of Smartphone Apps With Various Features for Weight Loss: Systematic Review and Meta-analysis. JMIR mHealth and uHealth, 10(4):e35479. Read on PubMed →
  5. Helsel DL, Jakicic JM, Otto AD (2007). Comparison of techniques for self-monitoring eating and exercise behaviors on weight loss in a correspondence-based intervention. Journal of the American Dietetic Association, 107(10):1807-1810. Read on PubMed →
  6. Höchsmann C, Martin CK (2020). Review of the validity and feasibility of image-assisted methods for dietary assessment. International Journal of Obesity, 44(12):2358-2371. Read on PubMed →

The formulas and limits behind MyPlate's own numbers, with their citations, are on the health sources page.