A day with missing entries should look different from a day you under-ate#
Rough days shipped on September 23, 2026. It is a small change with a narrow job: when a past day in your log looks unfinished, the app marks it with a yellow half-disc instead of scoring it like any other day. The marker appears in the Plan week strip and in History, on both the recent-day cards and the calendar. It sends no notification, shows no banner and asks nothing of you. It sits there, the way a pencil tick sits in the margin of a paper diary.
The reason for the marker is a problem every food log has and almost none admit to. A day where you stopped logging at four in the afternoon and a day where you genuinely ate very little produce the same number at the bottom of the screen. Treated as equal, the first one quietly flatters your weekly picture: it reads as a big deficit, when all it records is that the log went quiet. The research on self-reported intake says this is not an edge case. It is the most common way food records go wrong.
Food records fail by omission more than by miscounting#
The study most people cite here is Lichtman and colleagues' 1992 paper in the New England Journal of Medicine. They followed people who described themselves as diet-resistant, reporting that they could not lose weight on 1,200 calories a day or less. Measured with doubly labeled water, their actual intake was far higher than their records showed: they under-reported what they ate by an average of 47% and over-reported their physical activity by 51%1. Their metabolism was unremarkable. The gap was in the record.
What gets lost matters as much as how much. In a metabolic-facility study where everything participants ate was known, the meals themselves were reported reasonably well, but energy from snacks eaten between meals was significantly under-reported2. Missing entries cluster in the unstructured parts of the day: the grazing, the late bite, the evening that never got written down. The longer argument on this, with the rest of the doubly-labeled-water literature, is in what the underreporting studies actually show, and the practical side of it in the hidden calories that stall weight loss.
So when a record runs out early, the likeliest explanation is not that you stopped eating. It is that you stopped writing. A tracker that silently accepts the short total as the truth is building your week on the part of the record most likely to be missing.
Why the rule needs two conditions, not one#
A short day on its own proves nothing. People fast, get sick, travel, skip dinner after a big lunch. An early last entry on its own proves nothing either; plenty of people finish eating at five. Flagging on either one alone would put yellow on days that were perfectly complete and train you to ignore the marker.
So a past day turns yellow only when both are true: you stopped logging noticeably earlier than you usually do, and the day's total came in well below your usual. "Usual" is yours, not a population average. It is learned from your own history, and until there is enough of it the app falls back to a sensible evening time and your daily goal. The default for "well below" is 35% under your usual total.
That combination is the fingerprint the underreporting research describes: a record that ends early and comes up short. Either signal alone is ordinary life. Both together is most often a log that went quiet.
A few days are deliberately left out. Today and future days are never marked, because they are not over. A day with nothing logged at all shows as empty rather than yellow, since there is nothing partial about it. And if you logged a late meal after the fact, that counts: filling in yesterday's dinner the next morning makes the day look finished again.
What you see when you tap it#
Open a past day and, if it looks unfinished, an Unfinished? chip sits next to its total. Tap it and a sheet titled Why Tuesday is yellow (or whichever day it is) lays out the two reasons in plain rows: when your last food was logged compared with your usual cutoff, and what the total was compared with your usual. No score, no judgment, just the two facts that triggered the marker.
Below the reasons are three ways out, and all of them are optional:
- Add food to Tuesday opens the composer already unlocked for that day, so whatever you remember goes to the right date instead of today.
- Mark complete tells the app the day really was that small. The mark sticks: the day stays finished even if you later edit or delete a meal from it.
- Adjust rules takes you to Profile → Unfinished days.
Doing nothing is also a fine answer. The marker does not escalate, does not come back as a reminder and does not block anything.
You can tune it or switch it off#
Profile → Unfinished days holds the whole feature. There is an on/off switch; turned off, every past day is simply treated as finished. There are two dials. Last food before is the time cutoff; it follows your own habits automatically until you set it yourself. Total below usual by is the shortfall, 35% by default; raise it if you often eat light on purpose and do not want those days marked. Restore defaults puts both back.
The same signal carries into the weekly report, so a week with a couple of unfinished days reads as a week with gaps rather than a week of triumphant deficits. And a day you closed with the kitchen-closed checklist shows a ✓ instead: you already said the day was done, so the app believes you. That ritual is described in the Finish Today redesign.
Why a passive marker, and not a nudge#
The obvious design would have been a push notification at ten at night: "You haven't logged dinner." We did not build it, for reasons the evidence supports.
First, a missing evening is not a broken habit. In habit-formation data, automaticity built up over roughly two months on average, and missing a single day did not meaningfully set it back3. A gap is a gap, not a reset, so there is no emergency to interrupt your evening over.
Second, what keeps people logging is frequency across the week, not perfection on any single day. In a nine-month weight-maintenance study, people who self-monitored on at least five or six days a week kept losing weight, while those logging only one or two days a week regained4. A marker that helps you see which days had holes supports the weekly pattern without demanding that every day be complete. The broader picture on logging in small, frequent instalments is in how to track calories when you're busy.
Third, and most important, the tone of the reminder changes what people do next. Guilt is not a motivator for eating well. People who associated chocolate cake with guilt reported less perceived control over their eating and lost less weight than those who associated it with celebration5. In the other direction, an experiment that encouraged self-compassion after eating a doughnut found that restrained and guilty eaters then ate less in a subsequent taste test than those given no such message6. That study was small and lab-based, so hold it loosely, but it points the same way as the guilt data: reproach tends to backfire, and a neutral acknowledgement leaves room to carry on.
That is what the yellow half-disc is trying to be. It is not a failure dot; it deliberately replaces the usual day indicator so it cannot be mistaken for a warning about eating too much. It says one thing, "this record might be missing something", and leaves the decision with you. The wider case for treating a log as attention rather than audit is the pillar on mindful eating with numbers.
What the marker does not do#
It does not guess what you ate. It does not add phantom calories to fill the gap, and it does not change your totals. It does not know whether you were ill, fasting or simply busy; the two conditions describe the record, not your intentions, and Mark complete exists for exactly the days when the short record was the whole truth. It also cannot catch the other kind of underreporting, the portion that was bigger than you logged or the bite you never thought of as food. It catches the day that ran out early, which is the single most visible hole a daily log can have.
FAQ#
Why is one of my past days marked yellow?#
Because two things were both true for that day: you stopped logging noticeably earlier than you usually do, and the total was well below your usual intake (35% below by default). Tap the day, then the Unfinished? chip, and the sheet shows both reasons with the actual times and totals.
Does a yellow day mean I did something wrong?#
No. It means the record for that day may be incomplete, which is the most common way food logs go wrong for everyone. Self-reported intake has been shown to fall well short of measured intake, with snacks between meals among the items most often left out (Lichtman et al., 1992; Poppitt et al., 1998).
I really did eat very little that day. How do I clear the marker?#
Open the day, tap Unfinished?, then Mark complete. The mark is kept even if you later change meals on that day.
Will the app remind me to finish logging?#
No. There are no notifications or nudges for unfinished days. The marker is passive, and you can switch it off entirely in Profile → Unfinished days.
Can I make it less sensitive?#
Yes. In Profile → Unfinished days, raise Total below usual by above 35% so only very short days are marked, or set Last food before to an earlier time. Restore defaults undoes your changes.
Does logging a meal the next morning fix yesterday?#
Yes. Use Add food to
Sources#
- Lichtman SW, Pisarska K, Berman ER, et al. Discrepancy between self-reported and actual caloric intake and exercise in obese subjects. N Engl J Med. 1992;327(27):1893-1898.
- Poppitt SD, Swann D, Black AE, Prentice AM. Assessment of selective under-reporting of food intake by both obese and non-obese women in a metabolic facility. Int J Obes Relat Metab Disord. 1998;22(4):303-311.
- Gardner B, Lally P, Wardle J. Making health habitual: the psychology of 'habit-formation' and general practice. Br J Gen Pract. 2012;62(605):664-666.
- Arroyo KM, Carpenter CA, Krukowski RA, Ross KM. Identification of minimum thresholds for dietary self-monitoring to promote weight-loss maintenance. Obesity (Silver Spring). 2024;32(4):655-659.
- Kuijer RG, Boyce JA. Chocolate cake. Guilt or celebration? Associations with healthy eating attitudes, perceived behavioural control, intentions and weight-loss. Appetite. 2014;74:48-54.
- Adams CE, Leary MR. Promoting self-compassionate attitudes toward eating among restrictive and guilty eaters. J Soc Clin Psychol. 2007;26(10):1120-1144.
Source: BurnWeek — "Rough days: a quiet yellow mark for days your log ran out early", https://burnweek.fit/blog/unfinished-days-marker/. Licensed CC BY 4.0: free to quote or reuse with a link to this page.
