How do I log food quickly?

Save the day you usually eat as a day template, then log the whole day in one tap. Base is built around this rather than around search, because a serious athlete eats close to the same thing most days and searching a database five times a day is the cost that kills the habit.

Does Base have a food database?

Yes, with barcode lookup through Open Food Facts, but it is deliberately the fourth option in the add sheet rather than the first. Your day templates come first, then your own meals sorted by how often you eat them at that hour, then recent items, then search and barcode, then a bare macro entry.

Can I log different meals on training and rest days?

That is exactly what day templates are for. A training-day template and a rest-day template is the common pattern, and it is what makes carbohydrate periodisation practical — one tap for the high-carb day and one for the low.

What if I eat something not in the database?

Enter calories and macros directly with no name required. The quick macro entry exists because the goal is always to log the food, and an app that demands a product match for a homemade meal is an app you stop using on the third Tuesday.

Does Base track micronutrients?

Eighteen of them, and it reports explicitly what share of your day's food it actually has micronutrient data for. A micronutrient total computed over 40% of your intake is not a micronutrient total, and showing it without that caveat would be a lie by omission.

The Food Log: Template Replay Instead of Database Search

Most food trackers are built for discovery: a big search field, a database of two million products, browsing. That is the right design if you eat differently every day and want to explore.

A serious athlete's diet is repetitive by design. The same breakfast, the same post-training meal, the same dinner rotation, for months. Optimising that workflow for search is optimising the wrong thing, and the cost compounds: five searches a day at forty-five seconds each is nearly four minutes, every day, forever.

Four minutes a day is what kills a logging habit. Not lack of discipline — friction, multiplied by three hundred and sixty-five.

One tap does not.

The add sheet, ordered by likelihood

Tapping + on any hour opens the add sheet, and its order is the whole design:

Day templates — "Log Training Day", "Log Rest Day"
Your meals, sorted by how often you eat them at this hour
Recent items
Search and barcode (Open Food Facts)
Quick macro entry — calories and P/F/C, no name required

Item two is the piece that does the most work and gets noticed the least. The meal you always eat at 1pm is the first thing you see at 1pm. Not alphabetical, not most-recent-overall — most likely given the hour. The list reorders itself through the day without being told to.

Item five exists for a specific failure. Every tracker that demands a product match makes homemade food painful, and painful food is unlogged food. If you know it was roughly 600 calories with 45 g of protein, enter that. A rough entry beats a missing day by an enormous margin, and the arithmetic that matters — adaptive maintenance — is robust to a rough entry and destroyed by a blank one.

The hourly timeline

The left rail is hours. Active hours get a dark chip; empty hours a muted chip with a +. To the right of each chip, that hour's macro summary. Meal rows are cards with a thumbnail, the name, and a metadata line: 295 kcal · 17P 21F 12C · 1 serving.

Time-of-day is shown rather than hidden because for a hybrid athlete it is a real variable, not decoration. Carbohydrate before a long session and protein distribution across the day both matter, and neither is visible in a flat daily total. Roughly 25 to 35 g of protein clears the leucine threshold and the anabolic response lasts two to two and a half hours — so the same 180 g across two meals is a different stimulus from 180 g across four, and the timeline is where you can see which one you did.

Editing without leaving

Servings adjust inline. 1.5 servings without opening a detail screen.
Long-press a row for multi-select, then View · Copy · Move · Remove. Copy targets another day or another hour, which is how you build a template from a day that went well.
Swipe a row: left deletes, right duplicates.

Every one of these exists to avoid a screen transition. A screen transition is a quarter-second and a loss of context, and the daily budget is sixty seconds.

Micronutrients

Eighteen nutrients, on two tiers.

Build — shortfall demonstrably costs training adaptation: vitamin D, magnesium, zinc, iron, calcium, potassium, omega-3 (EPA+DHA).

Baseline — shortfall costs health and a real share of adults are short: fibre, sodium, B12, folate, vitamins A/C/E/K, selenium, iodine, choline.

Everything else — thiamin, riboflavin, niacin, B6, biotin, pantothenic acid, phosphorus, copper, manganese, chromium, molybdenum, chloride — is deliberately excluded. Population inadequacy is near nil in anyone eating enough calories with any protein at all, and an always-green row costs attention while teaching nothing.

Coverage is reported explicitly. If Base has micronutrient data for 62% of today's calories, it says 62% rather than presenting the total as complete. A micronutrient figure computed over part of your intake is not a micronutrient figure, and showing it without the denominator is a lie by omission. Full detail on the micronutrients screen.

Two things hybrid athletes need that a generic tracker does not give them

A carbohydrate number that moves with training. Because fat and protein are set from bodyweight and calories and carbohydrate takes the remainder, a heavier endurance week automatically produces a higher carbohydrate target. Fix carbs at a percentage instead and adding a long run silently takes calories out of protein too, which is the opposite of correct.

Two templates, not one. Carbohydrate periodisation — more carbs on hard days, fewer on easy, protein and fat held constant — is well supported and completely impractical if it means re-planning a day every morning. Two templates and two taps makes it a habit rather than a project.

What it will not do

No "food score" or letter grade. A colour-coded moral judgement on a meal is a different product.
No barcode-first flow. Scanning is a fallback, and it degrades silently to manual entry when the lookup fails or you are offline.
No photo requirement. The AI plate scan exists, is off by default, and needs your own API key. Why.

What to do next

Build one day template from a day you actually ate. It takes about five minutes once, and it is the single change that makes the difference between logging for a fortnight and logging for a year.