Estimation tool · not a prescription
What dose matches your goal?
Tell it where you are, where you want to be, and by when. It reads the published dose–response curves backwards and tells you which dose that implies, whether that lands in microdose territory, and at what point it has left the evidence behind.
Read this before you act on any number above.
This tool inverts published trial averages. You are not a trial average. It does not know your medical history, what else you take, your kidney function, your history with eating disorders, or whether a GLP-1 is appropriate for you at all. Doses below the lowest studied dose are extrapolation, clearly flagged as such, and no regulator has evaluated them. Bring the output to a prescriber as a starting point for a conversation, not as a decision you have already made.
How the calculation works
Three steps, and you can check each one.
- Your target becomes a percentage. Kilograms are meaningless across body sizes; every trial reports percent of total bodyweight, so the tool converts first.
- The timeframe is de-front-loaded. Weight loss on these drugs is fast early and flat late. Asking for 10% in 20 weeks is a bigger ask than 10% in 68, so the tool scales your target up to its equivalent endpoint value before touching the dose curve.
- The dose curve is read backwards. Anchor points come from the published trials. Between anchors the tool interpolates linearly. Below the lowest studied dose it interpolates toward the placebo response and flags the result as extrapolated, because that is what it is.
Anchor points the curve is built on
| Drug | Weekly dose | Observed total bodyweight loss | Endpoint | Source |
|---|
The sub-1 mg semaglutide anchors come from the phase 2 dose-ranging programme, where daily dosing was used, converted to a weekly equivalent by multiplying by seven. That conversion is an approximation, not a pharmacokinetic equivalence.
Why our tirzepatide numbers look lower than the ones you have seen. SURMOUNT-1 reports two different answers to the same question. The efficacy estimand (what happened to people who stayed on the drug) gives 16.0% at 5 mg and 22.5% at 15 mg. The treatment-regimen estimand (what happened to everyone randomised, including those who stopped) gives 15.0% and 20.9%. This calculator uses the treatment-regimen figures throughout, because they include the people it did not work for and you are not yet one or the other. Articles on this site quote whichever estimand the source reported, always labelled. Never compare a number from one estimand against a number from the other.
What "microdosing" actually means here. There is no regulatory definition. In practice people use it for any dose below the label's maintenance range, most often the 0.25 mg starting dose held indefinitely, or something smaller. The honest position: the low end of the semaglutide curve was studied in phase 2 and did produce meaningful loss, so this is not pure invention. But no microdose regimen has been through a phase 3 weight-loss trial, and compounded product used to hit these doses carries its own quality questions.
Questions people ask about this
Is a microdose safer than a full dose?
Lower doses reliably mean fewer and milder gastrointestinal side effects — that part is consistent across the dose-ranging data. Whether it is safer in a broader sense is a different question, because the rare risks on the label were characterised at full doses, and a lower dose does not automatically scale them down proportionally.
Will I regain the weight if I stop?
The withdrawal data is consistent and unkind: in the STEP 1 extension, participants regained roughly two thirds of what they had lost within a year of stopping. That applies at any dose. Plan for what happens after, not just what happens during.
Why does the tool refuse some timeframes?
Above roughly 1.5% of bodyweight per week you are outside anything these drugs produce and into territory where the loss is disproportionately lean tissue. The tool says so rather than returning a number that would be wrong.
Can I use this to work out a compounded dose?
No. Concentration varies between compounders, and a milligram figure is only meaningful against a known concentration. Converting milligrams to units on a syringe is exactly the step where dosing errors happen, and it is not a step this tool will do for you.
Keep reading