Buying guide

How to read a shilajit certificate of analysis

Most shilajit lab reports are built so they cannot tell you anything. Here is what a useful one contains, what "pass" hides, and how to check the numbers yourself.

Affiliate disclosure.We earn a commission if you buy through links on this page, at no extra cost to you. It does not change what we recommend, and we say so when we think you should not buy.

Why this document matters more here than almost anywhere else

Shilajit is a rock exudate. It seeps out of stone, is collected, purified to varying standards, and sold. Which means its mineral content — including the minerals nobody wants — is inherited from the geology it came out of.

That single fact reorganises the whole category. Metals in shilajit are not primarily a manufacturing contaminant to be cleaned up. They are a property of the source. And it means the words that usually reassure people — organic, pure, wild-harvested, traditional, ancient — tell you nothing whatsoever about what is in the jar.

The only thing that does is a laboratory report. So it is worth knowing how to read one.

What the evidence actually shows

We want to be careful here, because this is a topic where it is easy to frighten people and hard to inform them.

The published record is genuinely mixed. A 2008 study in JAMA analysing 193 Ayurvedic products found about a fifth contained detectable lead, mercury or arsenic — including one product sold as shilajit. Those samples were bought in 2005, and describe one lot of one product two decades ago. A 2025 analysis in BMC Chemistry looking at thallium — an element that appears on essentially no commercial certificate — found some finished supplements testing higher than the raw material they came from. But an independent testing programme in 2024 examined eight shilajit products and found no metal contamination concerns; its single failure was for containing more fulvic acid than the label claimed.

Those findings are not reconcilable, and we are not going to pretend otherwise. Different products, different years, different methods, and — the part that matters most — different ideas about what counts as a concerning number.

The honest summary is that metal content in this category is variable and not well characterised. Not uniformly dangerous. Not uniformly fine. Which is exactly why the document matters.

The reference points, and why they disagree by twentyfold

Before you can read a lab report you need to know what to compare the numbers against. There are three frameworks in common use, and they do not agree.

California’s Proposition 65 sets a lead level of 0.5 micrograms per day. It is state law, and it triggers a warning label, not a prohibition. It has no adopted level for mercury at all — so “meets Prop 65” tells you nothing about mercury.

USP, a non-governmental standards body, sets a permitted daily exposure for lead of 10 micrograms per day. Voluntary unless a product claims to comply.

The FDA has no general numeric heavy-metal limit for dietary supplements. Under federal manufacturing rules, the manufacturer sets its own specifications.

Read that last line again, because it is the key to everything below. On most certificates, the “limit” column is the company’s own number.

And the two named frameworks differ by twentyfold on lead. A product can comfortably pass one and require a warning label under the other. Which brings us to the single most important thing on this page.

A “PASS” tells you nothing

A pass is a comparison. Without knowing what it was compared against, it is not information.

Given that two legitimate reference points for lead differ by a factor of twenty, and that the specification on the page may simply be whatever the company decided, a certificate reporting pass or fail rather than measured numbers is structurally incapable of being evaluated. Not suspicious. Not necessarily hiding anything. Just useless for the purpose you are reading it for.

Always prefer numbers.

The checklist

Work through these in order. It takes about three minutes once you know what you are looking for.

1. Is the laboratory named? An unnamed lab is an unverifiable result. And check whether the lab and the brand are the same entity — an in-house result may be perfectly competent, but it is not third-party testing, and the difference is who chose the sample.

2. Is there a lot or batch number, and does it match your jar? This is load-bearing. A certificate with no lot number cannot be connected to any particular product. One with a lot number that does not appear on what you bought is equally disconnected. Everything else on the page depends on this.

3. How old is it? A certificate describes the sample on the day it was tested. Compare the date against the production date and the shelf life.

4. Are there measured values, or just verdicts? See above. Numbers let you apply your own standard.

5. Which analytes were actually tested? Lead, arsenic, cadmium and mercury are the four people usually look for. A report covering two of them is silent on the other two. That silence is not reassurance — but it is not necessarily concealment either. It is an unanswered question, and should be read as one. The thallium finding above is the sharp illustration: measured in commercial supplements, and on virtually no certificate anywhere.

6. Are detection limits stated? This one is missed almost universally. “Not detected” does not mean “not present.” It means below what this method could see. A method that can see down to 0.01 and one that can only see down to 5 will both report “not detected”, and they mean completely different things. A “not detected” with no stated detection limit is not a clean result — it is no result.

7. Check the units, and convert to your actual dose. Parts per million and parts per billion differ by a thousandfold and look almost identical on a page. Results may be given per gram of material or per serving, and the conversion depends on a serving size the certificate usually does not state. Convert everything to micrograms per daily serving before comparing it to anything.

8. Is a method stated? Rarely. It should be.

What a certificate cannot tell you

It cannot tell you the product works — no laboratory measures efficacy. It cannot tell you the product is safe at the dose you will take; it reports composition, and safety is a separate judgment. It cannot tell you anything about consistency between batches, because it describes one batch. And if the lot does not match what you bought, it tells you nothing about your jar at all.

A certificate is a narrow, dated, sample-specific measurement. It is genuinely valuable, and it is far narrower than the confidence it usually gets deployed to create.

Applying the checklist to a real product

We will run this on the shilajit product we happen to earn a commission on, because a checklist you never apply to yourself is decoration.

What its certificate does well. Two laboratories are named with accreditation numbers. There is a batch number and a production date. Crucially, it reports measured numbers rather than pass/fail — and it includes measurement uncertainty, which is better practice than most consumer-facing certificates in any supplement category. Specification limits are shown alongside results. It also covers assay, moisture and pH, and two pesticides.

Where it falls short of the checklist. Cadmium and mercury are not among the reported analytes — only arsenic and lead — so on two of the four metals people look for, the document is silent. No analytical methods are stated. No detection limits. The specification limits shown are considerably more permissive than the voluntary standards body’s guidance, and the document does not say where they come from. One laboratory’s cited accreditation validity dates fall before the production date on the report; whether it was renewed cannot be determined from the document.

And the part the checklist was built for. At the product’s own stated serving, its reported lead figure works out to roughly 0.68 micrograms per day. That sits comfortably under the voluntary standards body’s limit of 10, and above California’s 0.5 threshold.

Both readings are legitimate. Neither is a scandal. We are showing you the arithmetic rather than a verdict because that is the entire point — the same measured value passes one reference point and exceeds another, and only the number lets you see that.

It is very much to this document’s credit that it reports a number at all. A pass/fail certificate would have made the calculation impossible, and you would have had no way to know.

For contrast, we looked at another brand’s published report. It has a clean batch number and a full received-tested-reported date chain, which is good discipline. But lead, arsenic, mercury and cadmium are each reported only as “Pass” — no values, no units despite a units column, and specifications given only as “per specifications”, unnamed. It is a real, published, publicly available certificate that cannot be independently evaluated for metals, because there is nothing in it to evaluate.

We are not naming that brand as dishonest. It is doing more than most, which publish nothing at all. It simply illustrates the difference better than any argument could.

What we would actually do

1. Look for the certificate before you look at the price. If a brand publishes none, you are relying entirely on its word, and you should know that is the position you are in.
2. Check the lot number matches. If it does not, the document is about a different material.
3. Insist on numbers. A pass/fail report is not evidence.
4. Note what is missing, and treat it as an open question rather than either reassurance or alarm.
5. Convert to your daily dose before comparing anything to any limit.
6. Decide which reference point you want to use — and know that reasonable people use different ones.

None of the certificates we could find, from any brand, meets this checklist in full. Every one omits methods and detection limits. That is the industry norm, not a failing unique to anyone.

Which is really the finding here: the problem with shilajit certificates is not that they reveal alarming results. It is that most are constructed so they cannot reveal anything at all.

If you have decided you want shilajit and you want the format question rather than the documentation question, that is the next page — and it will probably tell you the thing you were about to base the decision on does not matter.

Written and reviewed by CJ. Last verified 2026-09-03. Next review due 2027-03-03.