How to Lose Weight With a Food Log You Will Actually Keep

Of all the things people try for weight loss, writing down what they eat has the best evidence and the worst retention. In a trial of nearly 1,700 adults run by Kaiser Permanente, the people who kept a food record six or seven days a week lost about twice as much weight over six months as those who kept one a day a week or less (Hollis et al., American Journal of Preventive Medicine, 2008). The method works. The problem is that most people stop within a few weeks, and the usual reason is that the logging itself is a chore: searching a database, weighing, picking the right entry from forty.

So the question is not whether to log. It is how to make the log cheap enough to keep for months.

The method

1. Estimate, do not weigh. "A bowl of pasta with meat sauce" is accurate enough, and it is the difference between a log you keep and one you abandon by Thursday.

2. Log at the meal, not at night. Memory at 10pm loses the snack at 3pm, which is often the one that matters.

3. Log the bad days. The days people skip are the days with the most information in them. A log with gaps on Fridays is a log that has hidden Friday.

4. Look at daily totals once a week, not meal by meal. One high day in seven is normal; four is the pattern you are looking for.

5. Change one thing at a time, based on what the week showed. Usually it is an afternoon snack, a drink, or portion size at one meal.

Where the text log fits

Nutrition Textapp is a food log that costs one text per meal. Text Hi to its number, then text what you ate in your own words: "two slices of pepperoni pizza and a coke", "chicken stir fry, medium bowl". It estimates the calories and protein, plus carbs, sugar, saturated fat and cholesterol, logs the meal, and replies with the numbers and your running total for the day.

If the estimate is off, say so and it corrects the last entry. Tell it a daily calorie target and it keeps it in mind. Ask for your day and it gives you the totals. If you want to see the whole record, ask for a link to your data. If you would like a reminder to log, say yes when it asks; if you would rather it never nudged, it will not.

The whole thing happens in your messages, which is the reason it survives week three. There is nothing to open.

What the first month looks like

Week one is discovery. Most people find one or two things they were not counting: the oil in cooking, the second glass of wine, the handful of something from the office kitchen. Week two is when the daily totals start to look consistent on ordinary days, and the high days stand out with a reason attached. Weeks three and four are where the Kaiser trial's effect comes from: not from the logging itself but from the small changes it makes obvious, kept up because the log keeps showing them.

The target to aim for is logging most days, not every meal perfectly. The trial's biggest gap was between people who logged most days and people who rarely did, not between perfect and good.

A note on what the numbers are

Estimates from a description are estimates. They are consistent, which is what a log needs, but they are not lab values. Anyone managing a medical condition through diet should use the log as a record to discuss with a doctor or dietitian, not as a prescription. Nutrition Textapp will tell you what it tracks and what it does not, and it does not track fiber, sodium or total fat.

Try it by text

Text Hi to (609) 806-5168 to start Nutrition Textapp. Text what you eat in your own words and it logs the calories and macros and keeps your daily total. It works on any phone that can text, or over WhatsApp, with no app to install.

Your first 20 messages are free. After that, one $9.99 a month subscription covers every Textapp. Text STOP at any time to end it.