Search shilajit and training on any app and you’ll get reel after reel of someone deadlifting under a caption about mountain resin. The human research behind those captions comes down to two small trials, 63 active men in one and 25 in the other. That’s the lot. We sell the stuff and we’ll still say it plainly: the gym marketing has run years ahead of the evidence. Here’s what the two trials measured, and where they stop.
The most useful study on shilajit and gym recovery is Keller 2019, published in the Journal of the International Society of Sports Nutrition. A randomised controlled trial in 63 active men, 500 mg a day for eight weeks, using a standardised extract rather than raw resin. The men were put through a fatiguing protocol and the researchers measured how much of their maximum strength survived it.
By week eight, the shilajit group held onto more of their strength under fatigue than the placebo group did. But their one-rep max didn’t climb higher and their gains didn’t arrive faster. They kept more output when tired. A narrow finding, and exactly the sort of thing a training study can measure properly. The abstract is here if you’d rather read it than take our summary of it.
The same trial reported a lower level of hydroxyproline, a marker of collagen breakdown, in one subgroup of participants. That second result is the one quoted all over the internet, usually with the word “subgroup” removed.
A subgroup result means the effect turned up in a slice of the participants rather than across all 63. Cut a sample that small into slices and go hunting, and you’ll find something eventually. So researchers treat a subgroup finding as a question worth testing again, not a result you can build a product page on.
If we ran an ad saying shilajit lowers collagen breakdown in athletes, we’d be doing the thing we complain about when the budget jars do it. One clear primary finding in 63 men, one soft secondary finding in fewer. That’s the strongest study on the training side, and we’d still call it a starting point.
Yadav 2026 in Cureus was an open-label pilot in 25 healthy young men taking shilajit resin for 28 days. They improved on strength and fatigue measures over the month. And the study reported that testosterone was unchanged in that group across those 28 days.
We’re keeping that last sentence in. It’s the one a lot of this industry would quietly drop, and part of why we print nothing about hormones on our jars. The design caveat matters too: open-label means everyone knew what they were taking and there was no placebo group, so expectation is doing some unknown share of the work. What the pilot has going for it is the format. It used resin, which is what most South Africans actually buy.
The study quoted far more often is Pandit 2016 in Andrologia, a randomised controlled trial in 75 healthy men aged 45 to 55 on 500 mg a day of purified shilajit for 90 days, which reported total testosterone up 20 per cent and free testosterone up 19 per cent against placebo. Different age band, duration, preparation and outcome. If you’re 25 and in the gym, Pandit 2016 is not a study about you, whatever the reels are captioned.
| Study | Who, how long | Form used | What it found |
|---|---|---|---|
| Keller 2019, JISSN, randomised controlled | 63 active men, 500 mg/day, 8 weeks | Standardised extract | Retained strength under fatigue, plus a lower collagen-breakdown marker in a subgroup |
| Yadav 2026, Cureus, open-label pilot | 25 healthy young men, 28 days | Resin | Strength and fatigue measures improved, testosterone unchanged |
Four things sit between those results and the claims on a supplement ad.
So the evidence base is thin, with two decent signals in it. We think that’s still worth something. Anyone selling shilajit as a recovery breakthrough is describing a study nobody has run yet.
Independent testing in the United States found fulvic acid per serving ranging from under 7 mg to over 2 200 mg across shilajit products. A front label on its own proves close to nothing. Genuine raw resin usually measures 15 to 35 per cent fulvic acid depending on the assay, and our batch 250915 came back at 35 g/100g by titration, with the method printed beside the number.
Contamination matters more than potency here, because you’d be taking this daily for months. Kamgar 2025 in BMC Chemistry analysed six commercial shilajit products and found thallium in five of them. That same resin went to Qarshi Research International, an ISO/IEC 17025 accredited lab and ILAC member, between 28 November and 8 December 2025: lead, arsenic, mercury and cadmium all not detected by AOAC 999.10/11. The results sit on our lab tests page, and if you’re comparing brands, the checks we’d run on any jar are written up separately.
Start with the dose gap. The trials used 500 mg a day of a standardised extract. A pea-sized serving of resin is roughly 0.4 g. Close by weight, not the same material, and pretending otherwise would be dishonest maths.
Then give it the window the research used. Keller ran eight weeks, Yadav ran 28 days, and neither reported anything meaningful at day four, so judging a jar after a long weekend tells you nothing. Our 15 g jar at R500 covers about 37 days, which is shorter than the Keller window. Long enough to find out whether you can stand the taste, too short to answer the training question. The 40 g jar at R999 works out to R25 a gram and runs about 100 days at that dose, so that’s the Keller window covered with room to spare.
Change one thing at a time. Start shilajit the same week you switch programmes, add creatine and finally start sleeping seven hours, and you’ll learn nothing about any of them. Write down two numbers, a working set weight and how you feel on your third session of the week, then look again in eight weeks. And if your sleep and your food are a mess, shilajit is the least interesting variable in your week.
One randomised trial in 63 active men found they retained more strength under fatigue after eight weeks on 500 mg a day. That’s close to the whole evidence base. It’s a real finding in a small study rather than a settled effect, and nobody has measured what shilajit does to muscle soreness.
Eight weeks, going by the trial durations. Keller measured at eight weeks and the 2026 pilot at 28 days, so whatever you notice in week one is probably you paying attention.
Nobody has tested that head to head. Most of the positive research used a standardised extract rather than raw resin, while the 2026 pilot used resin. We’ve written up how resin, capsules and gummies compare on cost per serving, and we sell resin because it’s the form we can purify and test as a whole batch, not because a study says it wins.
No trial has tested timing. Both training studies dosed once a day, so consistency is the only part with support behind it. Most of our customers take theirs in the morning with something warm.
Read the lab report before you buy anything, ours or anyone else’s. Look for a batch number, a named lab, an accreditation, dates, and a method beside every figure, because a PDF without those is a design job rather than a test. Ours is linked further up this page, and it’s the document we’d want you reading before any of our sales copy. The 40 g jar holds enough resin to run the same eight weeks the researchers did.
A complementary product, not a medicine. Pregnant or breastfeeding women, and anyone on chronic medication, should speak to a health professional first.
We publish the full laboratory report for the batch shipping now, with the method named beside every result. Read it in full.
See the certificate
PurePharm