Start at the smallest deficit that works, and change one thing at a time.

It will not be a straight line, and any answer that gives you a date is guessing. Weight comes off in steps: weeks where the scale moves, a fortnight where it does not move at all, a morning where it jumps because of salt, a late meal or a hard session. The app acts on the direction over months, never on one reading.

What it does instead of promising a date is work at a pace you can hold, as a small share of your body weight a week, starting at the smallest deficit below your estimated maintenance rather than at a dramatic one. Forty weeks of that beats a fortnight you abandon.

The plan screen shows an estimate of the last stretch, worked out from the pace you chose, and it moves as your real trend appears. It is an estimate, not a promise, and the app never adjusts its way towards a date.

Expect the first week to flatter you. Eating less drops water along with fat, so the first few days can show more than a kilogram, and then the number settles into something slower. That early drop is not a bonus you can keep spending.

Expect the app to be quiet about scoring: for a fortnight it says that nothing is scored yet, because a percentage over days that had no plan to follow is invented. Log the weight, the food, the steps and the training, and the picture appears.

Expect the maintenance figure to be an estimate at this point. It is a population average applied to you, and the app replaces it with what your own logs measure once there are two weeks to read. Weigh in most mornings: a week with one reading in it gives the app nothing to average, and it says there is not enough to read rather than guessing.

No. A 0.9 kg gain of fat would need about 7,000 kcal more than you spent, and no single weigh-in changes anything the app decides. What moved is water and the contents of your gut: salt, alcohol, a late meal, a hard session, short sleep or the point in your cycle.

The app reads a weekly average, which is why one high morning cannot trigger a change. Log it, log the reason if there was one, and leave it.

If seeing the daily number is the problem, the settings screen hides it from the chart and draws the weekly average instead. Every weigh-in is still recorded and still counts, so nothing about the plan changes.

Log it honestly and carry on. One day above target does not undo a week, and the app will not mark you as failing because of it. It does not ask you to compensate with a very low Monday either.

What it does watch is the pattern: if your weekends consistently sit well above your weekdays, it says so, because that is often the whole explanation for a month that looks like a stall.

The one thing worth avoiding is deciding the week is ruined. A plan that cannot survive a wedding is not a plan.

No. Start again today, and fill in the gap if you remember enough to be worth recording: the past-data screen takes a day at a time from notes, and a rough number you are confident in beats a precise one you invented afterwards.

Logging matters for what the app does next rather than for its own sake. When progress stalls, the first question is whether the last week ran as planned — the steps walked and the food logged, in your own numbers. If the record is too thin to read, the app says that before it says anything about your metabolism. A missed week is missing evidence, and the next week supplies it.

Every instruction comes from your own record rather than from a template. There is no fixed deficit and no prescribed cardio block here: the app reads your weekly average, your tape measure, the food and steps you logged, and your training, then proposes the smallest change that could move the number, and says which of those readings it used.

The order is fixed and it does not change: more activity first, a small reduction in calories after that, and more daily movement last. One change at a time, and only after the app has asked whether the week ran as planned and whether the tape is still moving. A slow week that is still moving is not a reason to change anything.

The figures that come out of that — the calorie target, the cardio minutes, the step goal — are worked out for the body in front of it, which is why they are on your own plan screen rather than on a page written for everybody. You see each one, with the reasoning attached, before anything changes. On your own, you approve it or decline it yourself; with a coach attached to you, the decision is theirs — and every decision is recorded on your own plan screen either way, so you always read what changed and why.

First it checks the week rather than the metabolism. Steps and calories are read over the past week, and if either was off the plan it names them in your own numbers: your average day was well short of your step target, or your logged days averaged above your calorie target. A week that did not run as planned is answered with a properly run week, and the app looks again after a full week of it.

If the week did run as planned and the weekly average has still not come down by any real amount, over two consecutive weeks, it reads your measurements next. A waist still coming down means the plan is working and nothing changes.

If the tape has not moved either, it asks you — see the next question — and only then proposes one change, with the reasoning on the page before anything is decided. Nothing is ever applied automatically.

Because the scale cannot tell fat from muscle, and the tape and the photos can. Somebody losing fat and adding muscle at the same time holds their weight steady, and that is common in the first months of training rather than rare.

So when the weight has stopped moving and the tape has not moved either, the plan screen asks two questions: are your measurements still going down, and do your photos look different from a month ago? A yes to either one keeps the plan exactly as it is, and the app asks again after another stalled week.

A no to both means there is nothing to show for the last stretch, and the first change is proposed. The answer is yours and it can be changed: it stays on the plan screen while the stall lasts, so a mis-tap is corrected the next day.

Because calories are the hardest lever to take back. Reducing food costs recovery, training performance and adherence, and it has to be reversed during the exit phase.

The order is deliberate: more activity first, then a small reduction in calories, then more daily movement. Adding cardio is easier, cheaper to recover from and removes no food from your week.

There is also a floor. The app will not build a plan below the safe minimum intake it works from, and it asks you to confirm professional guidance before it goes anywhere near the lower one. A very low target is not a faster version of the same plan; it is a harder and less reliable one. Both figures are stated on the plan screen.

The first change is cardio: the minutes after each strength session grow a little, and one longer session is added on a rest day. It adds movement without removing food, and it can be undone.

If the weight still has not moved, the app checks the tape and asks the same question again before it goes near your food. Then it proposes a small reduction in calories, taken from carbohydrate so no food group is removed and your protein target does not move.

After that the remaining lever is daily movement, through the step target. Each one arrives as a proposal with its reasoning, and nothing changes until somebody decides it: you, or your coach if one is attached to you. Either way the decision and its reason stay on your own plan screen.

A fortnight of going up is assessed exactly like a fortnight of standing still, and for the same reason: it can be muscle gained while fat comes off, water, or a genuine overshoot. A scale cannot tell those apart and the app does not guess.

It checks whether the week ran as planned, in the same numbers. If your logged days were above target, the instruction is to run the week at the target. If the week ran as planned, it reads the tape and asks you about your measurements and photos.

The first change is still more cardio, never fewer calories as a reflex. A rising scale is a reason to look closely, not proof that you are eating more than you think.

No. Faster than the pace you chose costs muscle, strength and mood, and it is more likely to end with the weight coming back. It is also usually water rather than fat, which is why a quick few days are followed by nothing.

The app treats losing too fast as a warning sign. Losing faster than your chosen pace while reporting hunger, poor sleep, low energy or falling strength produces a recommendation to eat more rather than to push on. The rate most people keep is the one they can hold for months rather than weeks.

Yes, and the plan assumes you will. Food is logged by the day and judged by the week, so a restaurant meal fits inside a week that is otherwise on plan.

Alcohol is worth counting for three reasons: it carries calories, it adds water to the next morning, and it makes the following day harder to run as planned. Not forbidden, and not free.

What makes twenty kilograms work is repetition rather than precision. A rough log every day beats a perfect log for eleven days and a blank week.

Some loss is possible in any deficit, and the app is built to keep as much as it can: protein is set from your target body weight rather than your current one, training continues at the same frequency, and the deficit stays small enough to recover from.

For a good while strength tends to hold or rise, because better technique and a lighter body change what you can do. The training history screen tracks whether your lifts are moving, and one lighter session is treated as one lighter session rather than a decline.

If strength trends down while you are in a deficit, that is information about the deficit rather than about your effort. It is one of the signals that asks for calories to go up.

You, and nobody else unless a coach has been assigned to you. There are no public profiles and no browsing: a coach sees the people an administrator has assigned to them and nobody else, and an account nobody has assigned appears in nobody's list.

A coach sees the record the plan runs on — weekly averages, food and step logs, training — because that is what they are coaching against. Your measurements and your photographs are separate switches, one each, showing the date you turned them on. Nothing is shared by default, taking a switch back closes the door immediately, and changing coach withdraws what was shared, because it was given to a person rather than to a role.

Your photographs are kept away from the public web and served only to you, re-encoded on upload so the location data in the original file is dropped. You can export everything you have entered, and deleting your account removes the records and the photographs.

The programme does not stop. The exit phase is its own part of the plan: calories rise in steps towards your new maintenance, the extra cardio comes down, and training and your step target stay exactly as they are.

Expect the scale to move up a little when you eat more. That is glycogen and water rather than fat returning, and it is why calories and activity are changed one at a time instead of both at once.

Afterwards the record stays: your weight, training and food are still there, and maintenance is a plan you keep running rather than a line you cross.