Somewhere along the way, "microdosing" escaped from psychedelics research and attached itself to the most popular drug class in America. With roughly one in eight U.S. adults now taking a GLP-1 medication like Ozempic, Wegovy, or Zepbound, a growing group has decided the officially prescribed doses are more than they want — and they're taking matters, and syringes, into their own hands. On Reddit, TikTok, and Facebook, people swap protocols for taking 0.05 to 0.5 mg of semaglutide a week, when the standard starting dose is 0.25 mg and therapeutic doses run 1 mg and up.
Doctors, predictably, are split. Some call it a rational response to a real problem: full doses are expensive, sometimes brutal on the stomach, and stronger than many people need. Others point out — correctly — that microdosing GLP-1s is not FDA-approved, has essentially no clinical trial data behind it, and often involves eyeballing doses from compounded vials or counting clicks on a pen designed to be foolproof at exactly one setting. This piece isn't going to tell you to do it or not do it; that's a conversation for you and a physician who knows your history. It's going to do something more useful: lay out what's actually known, and what going off-script means for how you have to operate.
Why microdosing took off
A Johns Hopkins team recently used AI to analyze tens of thousands of GLP-1 discussions on Reddit, published in JACC: Advances. The recurring themes weren't vanity — they were cost struggles, insurance battles, side effects, and confusion about dosing. Microdosing sits at the intersection of all four. A vial or pen stretched across smaller doses lasts longer, which matters when insurance won't pay. Lower doses often mean less nausea, constipation, and food aversion. And some people don't want maximum weight loss at all — they want to lose a little, lose it slowly, or simply hold steady after finishing a full course.
That last group may be the most interesting. As we covered in our guide to keeping weight off after stopping a GLP-1, most people who quit cold turkey regain much of the weight within a year. A tiny maintenance dose feels, intuitively, like a middle path — enough to keep the food noise turned down without the full-dose side effects or the full-dose price tag. Intuitive, however, is not the same as proven.
What the evidence actually says
Here is the uncomfortable truth at the center of the trend: the clinical trials that made these drugs famous tested specific doses, escalated on a specific schedule, and the results scale with dose. In the semaglutide trials, higher doses produced more weight loss on average — that dose-response relationship is one of the most consistent findings in the literature. Below the lowest studied doses, there is no trial data. Nobody has run a randomized controlled trial of 0.1 mg weekly semaglutide, so nobody can tell you what it does on average, let alone what it will do to you.
What pharmacology can tell you is why the answers vary so much from person to person. At sub-therapeutic doses, you're operating on the steep, unpredictable part of the dose-response curve, where individual differences in receptor sensitivity, body weight, metabolism, and baseline appetite swamp the drug's average effect. That's exactly what the anecdotes reflect: one Redditor swears 0.1 mg silenced their food noise completely; another feels nothing below 0.5 mg. Both are probably telling the truth. Medical reviewers at Drugs.com summarize the state of play drily: microdosing may reduce appetite and cravings, but the effects are highly variable and not clinically proven.
The part nobody tells you: you just became the trial
When you take a studied dose of a studied drug, you're borrowing the statistical confidence of a trial with thousands of participants. You know roughly what to expect because thousands of people walked the path first. The moment you drop below the studied range, that borrowed confidence evaporates. There is no average result to lean on. There is only your result — and the strange thing is that most people microdosing have no systematic way of knowing what their result even is.
Think about what "is it working?" actually means at 0.1 mg a week. Appetite suppression at that level isn't a sledgehammer; it's a subtle shift in how often you think about food and how quickly you feel done with a meal. Human memory is spectacularly bad at detecting subtle shifts. If you don't measure, you'll evaluate a months-long experiment on vibes — and vibes are exactly how people end up paying for a dose that does nothing, or crediting a drug for what was really a changed routine.
A meaningful self-experiment needs the same three things every trial needs: a baseline, consistent measurement, and enough time. Concretely, that looks like knowing your typical daily intake and hunger pattern before changing anything, then tracking a handful of signals — what you eat, rough protein (muscle loss is the quiet risk of every GLP-1 protocol, at any dose), weekly weight trend, and a one-line note on hunger and side effects — through each dose change, for weeks, not days. That's how you distinguish "0.25 mg every ten days holds my weight" from "I think it's maybe doing something."
Run your n=1 like it matters
The reason most self-experiments fail isn't willpower — it's that logging is tedious exactly when your data matters most. VoiceLog strips the friction out: say "chicken salad and a flat white, hunger was maybe a 4 today" and it files the meal, calories, protein, and the note, transcribed on your iPhone, no food-database spelunking. Weeks later, you can actually see whether the smaller dose changed your intake, your protein, your weight trend — instead of guessing. If you're going to be your own trial, at least collect the data.
Get VoiceLogWhat a sane protocol looks like (if you and your doctor go there)
- Involve a real physician — ideally one comfortable with obesity medicine — and be honest about the dose you're actually taking. Off-label dosing with supervision is medicine; off-label dosing from TikTok is gambling.
- Establish a two-week baseline before changing anything: typical intake, hunger pattern, weight trend. Without a baseline, no later comparison means anything.
- Change one variable at a time and hold each dose at least four weeks. GLP-1s reach steady state slowly; judging a dose after five days tells you nothing.
- Track intake, protein, weekly weight average, and a daily one-line hunger/side-effect note. The weekly average matters more than any single weigh-in.
- Decide your success criteria in advance — "weight stable within a kilo and food noise manageable" — so you can't retroactively declare victory on whatever happened.
- Know your exit: if the data shows the microdose doing nothing after six to eight weeks, stop paying for it or discuss a proper dose. A negative result is still a result.
The honest bottom line
Microdosing GLP-1s is what happens when an effective drug collides with a broken pricing system and bodies that don't read the prescribing information. The impulse behind it — right-sizing a powerful medication to your own goals — is reasonable. The evidence beneath it is thin to nonexistent, and the practice carries real risks that deserve a doctor's eyes. What's not in dispute is this: the further you step off the studied path, the more the burden of proof shifts from the pharmaceutical trial to you. The trial had thousands of participants, control groups, and statisticians. You have one participant. The least you can do is take good notes.
One sentence a meal. That's the whole system.
VoiceLog turns spoken sentences into a food, protein, weight, and habit log — on-device, private, and fast enough that you'll still be doing it in week six when the data finally means something. Free to start.
Get VoiceLogFrequently asked questions
What is microdosing Ozempic or other GLP-1s?
Taking doses below the standard prescribed range — often 0.05 to 0.5 mg of semaglutide weekly, versus the 0.25 mg starting dose and 1 mg-plus therapeutic doses. People do it to cut costs, reduce side effects, lose weight more slowly, or maintain weight after a full course. The practice is not FDA-approved.
Does microdosing a GLP-1 actually work for weight loss?
Nobody knows on average — there are no clinical trials below the studied doses. Trial data shows weight loss scales with dose, so smaller doses likely produce smaller effects, and responses at tiny doses vary enormously between individuals. Some people report real appetite reduction; others feel nothing. The only way to know your response is to track your intake and weight systematically.
Is microdosing GLP-1s safe?
Lower doses generally mean milder side effects, but the practice adds its own risks: dosing errors from splitting doses or counting pen clicks, compounded products of uncertain concentration, and reduced medical oversight. Doctors are divided on the trend, and every version of it should involve a physician who knows what you're taking.
Can you use a microdose of Ozempic for weight maintenance after losing weight?
It's one of the most common reasons people microdose, and mechanistically it's plausible — but unproven. Most weight regain happens after stopping GLP-1s entirely, so a small maintenance dose is an active area of experimentation. If you try it with your doctor, track your weekly weight trend and intake so you can tell within weeks whether it's holding.
How do I know if my GLP-1 microdose is working?
Define what "working" means before you start — for example, stable weight and manageable hunger. Record a baseline first, then log meals, protein, a weekly weight average, and a daily hunger note for at least four to six weeks per dose. Subtle appetite changes are almost impossible to judge from memory; the data will tell you what vibes can't.