How your plan adapts each week

Once a week, Kinra quietly checks how things are going and, if needed, nudges your calorie target a little. Here's what happens in that check-in, and why it's built to move gently rather than react to every wobble.

What the weekly check actually compares

Your plan started with a target pace — say, losing weight at a modest, steady rate. Each week, Kinra compares that intended pace against what your trend weight actually did. Trend weight isn't your scale reading from this morning; it's a smoothed line built from roughly the last ten days of weigh-ins, so a big dinner out or a salty weekend doesn't throw it off. Day-to-day weight moves for reasons that have nothing to do with fat — water retention, sodium, glycogen stores, what's currently in your gut — so looking at any single number would be like judging the weather from one gust of wind¹. The trend is what matters.

Alongside trend weight, Kinra keeps a running, day-by-day estimate of what you actually burn, learned from your logged meals and your weigh-ins together — the estimator behind it tracks your de-noised weight and your daily burn as a pair, and only a burn estimate that fits both your intake and your trend survives. That estimate updates smoothly in the background every day, but your target never chases it day to day: the plan only ever changes at the weekly check-in. Combining what you actually ate with what your weight actually did gives a sturdier signal than either one alone.

From gap to correction, and why it's damped

If your trend weight moved faster or slower than planned, that gap gets translated into a calorie correction using roughly 7,700 kilocalories per kilogram of body tissue. That number is a useful approximation over a short window, not a fixed law of physics — the true energy cost of a kilogram of weight change shifts over the course of a diet, since early weight loss carries more water and glycogen and less fat than later weight loss²³. Kinra treats it as exactly what it is: a working approximation, not a precise measurement.

Here's the part that matters most: Kinra doesn't apply that full correction. It damps it by half. If the numbers suggest you're 200 kcal a day off pace, the adjustment applied is roughly 100 kcal a day, not 200. Then it caps the result — no single weekly adjustment moves your target by more than about 150 kcal a day in either direction. Finally, if the correction is tiny, Kinra ignores it rather than fiddling with your target over noise that isn't meaningful. And an adjustment is a proposal, not a decree: Kinra suggests the new target and you accept it — a change to your goal or your pace is always a confirmed action, never something patched quietly in the background.

Put together, one week's adjustment looks like this:

observed vs. planned pace  →  raw correction
raw correction ÷ 2         →  damped correction
capped at ±150 kcal/day    →  final adjustment
rounded to the nearest 10

Why gentle beats big swings

It would be easy to build a system that reacts hard — see a gap, close it fast. Kinra deliberately doesn't, because the evidence points the other way. A meta-analysis comparing gradual and rapid weight loss for the same total amount lost found that, on average, gradual loss better preserved resting metabolic rate and came with a somewhat larger share of fat loss relative to lean tissue⁴. Slow, steady correction tends to protect the things you actually care about — your energy, your muscle, your ability to keep going — better than a sharp course correction does.

There's a second reason for restraint. Your body doesn't sit still while you diet or bulk: sustained calorie deficits tend to nudge your energy expenditure down over time, and sustained surpluses tend to nudge it up. That's a normal, expected adaptive response, not a sign that anything has gone wrong⁵. Because that drift is expected, Kinra's job isn't to fight it hard each week — it's to keep gently re-aiming, correcting only for what the data actually shows and leaving room for your body's own adjustment to settle.

Why it holds steady when data is thin

Some weeks, you won't log much, or your weigh-ins will be sparse or irregular. Kinra's burn estimator computes its own uncertainty as it learns — a real statistical range that narrows with complete days and weigh-ins and widens through sparse stretches — and the confidence level you see (high, medium, or low) comes from it. When confidence is low, when too many recent days look missing or only partly logged, or when your logs and your trend disagree too much to reconcile, the plan holds your current target rather than reacting to a shaky signal.

Partly logged days deserve a special mention, because they're the most common failure mode of this kind of math. A day where you logged breakfast and lunch but the evening went unrecorded looks, numerically, like a day you barely ate — and a naive system would read that as a lower burn and cut your target. Kinra's estimator instead treats such a day as "at least this much," a floor rather than a total, so a partial day can never drag your numbers down. Skipped days are simply absent. Neither costs you anything except a slightly wider uncertainty range until the data fills back in.

A held target, in other words, is not the plan stalling. It's the plan declining to move on evidence that isn't there yet — which is exactly what you'd want from a coach.

This isn't about not trusting you. It reflects something well documented in nutrition research: people, even when motivated and trying carefully, tend to under-report what they eat when logging by hand, sometimes by hundreds of calories a day, and this shows up even in people actively working on their weight⁶. That's exactly why Kinra leans on your smoothed weight trend as a cross-check rather than taking any single week's log at face value — logs are one signal among several, not a test you can pass or fail. When the picture is unclear, the calm move is to wait for more information, not to guess.

The same caution applies to your starting number. Your very first calorie target comes from a widely used formula (Mifflin-St Jeor) adjusted for activity. In studies, that formula tends to be reasonably accurate on average across large groups of people, but it's not exact for any one individual — it's a solid starting estimate, not a verdict⁷. As real weeks of data accumulate, Kinra leans more on what's actually observed from you and less on the formula, but always through this same damped, capped process.

The adjustment is not a judgment

If your target shifts by 40 kcal one week, that's not a scorecard and it isn't a reaction to anything you did "wrong." It's the plan doing its one job: staying close to a moving target using real signals, in small enough steps that it never overreacts to noise. The point of damping, capping, and holding steady when data is thin is to keep the plan working quietly in the background, for you, without whiplash.

If you're dealing with a medical condition, pregnancy, or a difficult relationship with food, it's worth talking to a qualified clinician or registered dietitian — Kinra is built for general wellness support, not diagnosis or treatment.