GLP Fish is our iPhone app for recording a GLP-1 course of treatment — semaglutide or tirzepatide, weekly injection or daily tablet. Version 1.0 was finished and submitted to the App Store in mid-August; the first review round came back with three things Apple wanted done differently (community-content protections, an age-rating detail, and making account deletion easier to find), the fixes are written and tested, and the resubmission is being prepared. We would rather explain what 1.0 is while that finishes than stay quiet, so: what the app does, and the reasoning behind the part people ask about most — the forecast.
This post is the why. For step-by-step use, there is now a guide page; for what comes after 1.0, see the companion post on the 1.1 update.
One record, four stages
A GLP-1 course of treatment runs in months and it does not run in a straight line. You spend weeks deciding; you use it; you may pause — because you hit your goal, because of side effects, cost, or your clinician's advice; and you may resume, onto the same medication or a different one. Most weight apps handle this by flattening it into a "current diet" that starts when you install the app and breaks when you stop logging.
GLP Fish instead models the course as episodes on one timeline. A pause records its date and its reason, makes the finished episode read-only, and keeps it on the chart as its own chapter — and you keep weighing in through the pause, because whether the weight holds without the medication is exactly the question a pause exists to answer. Resuming is a real step too: you re-confirm the medication and dose, and a new episode starts on top of the old one. The record survives every stage change, which is the whole point of having it.
Around the timeline: doses are logged in seconds with their cost in your own currency (shots and tablets keep separate records, and late doses keep both the schedule and the actual time), weigh-ins carry the context you weighed in under, and official label information for your market — FDA, PMDA, EMA, MFDS, TFDA — opens inside the app instead of our paraphrasing it. Everything works without an account; an account adds sync, and deleting it deletes the cloud copy.
The forecast, and why it looks borrowed
While you are still deciding, GLP Fish draws study curves from your starting weight at 12, 24, 52 or 72 weeks, for the medications you are considering. Every number comes from a named published trial, shown on screen next to the curve:
- STEP 1 — semaglutide 2.4 mg weekly injection, 68 weeks: the group averaged −14.9% body weight. The placebo group, on the same diet and activity counselling, averaged −2.4%.
- SURMOUNT-1 — tirzepatide 10 mg weekly injection, 72 weeks: −19.5%, against a placebo group at −3.1%.
- OASIS 4 — semaglutide 25 mg daily tablet, 64 weeks: −13.6%.
The placebo arm is on screen on purpose. Everyone in those trials had counselling and lifestyle support, so the drug arm's number is not the drug's doing alone — and a curve that quietly absorbed the placebo's work would flatter the medication. We show both arms so the comparison is honest.
Two deliberate refusals shape the rest of it. First, the app only draws those personalised-looking curves when the trial population plausibly covers you; if your recorded profile falls outside it, the published endpoints stay readable as text and nothing pretends to be yours. Second, once treatment starts, the app does not restart a trial curve from your latest weight and call it "your remaining trajectory". That number is not knowable from a group average, so you get your recorded trend beside the published endpoints — a comparison you can think with, not a prophecy.
Weigh-ins that can be compared
The other half of the forecast is the trend it is drawn against, and daily weight is noisy in a specific way: the same body reads differently at different hours. GLP Fish asks when and how you weighed in — morning before food, after a meal, bedtime — and corrects each reading against the morning fasted baseline, showing a confidence range with every corrected number: narrow in the morning, wide late in the day.
We want to be clear about what those context corrections are: rough population starting points, not a validated personal formula. That is why the uncertainty is shown rather than hidden, and why the next update (see the companion post) makes the correction learn your own daily rhythm from paired readings instead of trusting the population's average forever.
Where the real-world numbers come in
Trial averages describe people with dedicated staff, fixed dose increases and high adherence. The published real-world data is less flattering and more like life: in a study of 7,881 adults who started injectable semaglutide or tirzepatide in 2021–2023 (Gasoyan and colleagues, Obesity, 2025), the one-year average loss was −7.7% on semaglutide and −12.4% on tirzepatide — and roughly half the participants had stopped within a year; those still on treatment averaged −11.9%. Trials tell you what the medication can do under near-ideal conditions. Real-world data tells you what usually happens, and it is the fairer thing to show someone deciding whether to start. Version 1.1 turns that dataset into the app's prediction, alongside the trial curves, with the difference between the two stated rather than averaged away.
What GLP Fish is not
It is not a medical device. It does not diagnose, does not treat side effects, and never recommends a dose change — the dose lists and missed-dose windows it shows come from the official labels, as education. Study scenarios are group averages, not promises about any one reader, and the app says so where the numbers appear. What the app can promise is the record: every reading, dose, cost and pause, kept in one place, comparable across a year — and a forecast you are welcome to check the sources of, because we name them.
GLP Fish is listed on the App Store. Questions or feedback? Support is the fastest way to reach us.