A cobalt assay report rarely arrives with a fanfare. It turns up as a PDF: a few pages of columns, footnotes and sample numbers, and somewhere in there is a figure that will be quoted in a press release, a broker note and a chat forum for months afterwards. Reading the whole page — not just the headline grade — is one of the quickest ways to tell a solid drilling result from a lucky one.

1. Start with the header, not the grade

The top of the certificate tells you who did the work and what was sampled. It usually lists the project and tenement, the drill hole identifier, individual sample numbers, the sample type (diamond core, reverse circulation chips, channel or soil), the laboratory's name, and the dates the samples were received and reported.

Check three things here. First, that the sample numbers on the certificate match the intervals quoted in the announcement — a mismatch is often carelessness, but it can be worse. Second, that the sample type suits the claim. A soil geochemistry result is not a drill intersection, no matter how impressive the number looks. Third, that there are no unexplained gaps. If the mineralised zone was sampled from 88 m to 104 m and the certificate jumps from sample 4471 to 4474, ask why.

2. Grade: mind the units

Cobalt is reported as a percentage, in parts per million, or occasionally in grams per tonne. The conversions matter because they are easy to get wrong: 1% equals 10,000 ppm, and 1 ppm equals 1 g/t. So 0.12% Co is 1,200 ppm. Anyone who mixes these up is out by a factor of ten thousand, which is the difference between a mine and a rounding error.

Is cobalt the point, or a by-product credit?

Most cobalt comes out of the ground alongside something else. In the Central African Copperbelt it rides with copper; in many laterite and sulphide nickel projects it comes with nickel; in some vein systems it appears beside silver, arsenic or gold. A 0.4% Co interval means something very different if it also carries 2% Cu than if it carries nothing else worth recovering. Look at the full element suite — copper, nickel, arsenic, iron, manganese, sulphur — and at the reported mineralogy. Cobaltite, carrollite, heterogenite and asbolane behave very differently in a processing plant, and a report that never mentions mineralogy is telling you the metallurgy is still unknown.

Cut-offs and compositing

Laboratories report individual samples: one number for every metre or two of core. Companies then composite those into wider intervals using a chosen cut-off, for example "a 0.02% Co lower cut-off with a maximum of two metres of internal dilution". That cut-off is a judgement, not a fact. Two companies can drill the same ground and report different widths depending on where they set it, so compare results built on similar assumptions.

3. Width: drilled metres are not mined metres

The interval length quoted is measured along the drill hole. If the hole crosses the mineralised zone at a shallow angle, that length overstates the thickness of the zone considerably. The steeper the intersection, the closer the two numbers come; where a hole is perpendicular to the zone, downhole length and true width are effectively the same.

Good reports give both, or at least state the intersection angle. If only the downhole length appears, treat the width as a maximum until proven otherwise. A few other width traps are worth watching:

• Recovery. If only 60% of the core came back through the mineralised zone, the grade describes the rock that survived, not the rock in the ground. Soft, friable cobalt oxides are especially prone to this.
• Internal dilution. Wide intervals often include barren bands. A 20 m result at 0.15% Co that contains 8 m of waste is not the same asset as a clean 20 m.
• "Including" intervals. Companies like to highlight a high-grade sub-interval. That is fine, as long as the full interval is quoted too.

4. The laboratory work: method, detection limits and QA/QC

An assay is a measurement, and measurements carry error. The certificate should tell you how the sample was prepared (crushing and pulverising), how it was digested — a four-acid digest dissolves most elements, while a fusion copes better with resistant minerals — and how it was analysed, commonly by ICP-OES or ICP-MS, sometimes by XRF or AAS.

Then look for the quality control data. A credible programme includes:

1. Certified reference standards inserted at a known rate, with results falling inside the expected range.
2. Blanks, to catch contamination between samples.
3. Duplicates, to show how repeatable the result is.
4. Stated detection limits, because a value sitting at the limit is not a grade — it is the absence of one.
5. Over-limit repeats, where a strong result is re-analysed by a method suited to high grades.

Accreditation to a recognised international laboratory standard is a useful signal, though it confirms the lab's systems rather than the truth of any single number. If none of this appears in the report, you are being asked to take the grade on trust.

5. Red flags worth pausing over

Certain patterns deserve a second look before you get excited. Selective reporting, where only the best intervals from a hole are released. Grab samples or rock chips presented alongside drill results as though the two were equivalent. Grades reported without widths, or widths without grades. A spectacular single sample with no repeat analysis. Units that quietly change between paragraphs. And results described as "encouraging" with no cut-off, no width and no comparison to nearby drilling.

6. A five-minute routine you can repeat

Given a new assay report, work through it in the same order every time. Check the header and the sample numbering. Convert every grade to one unit so the numbers can be compared. Note whether cobalt is the main metal or a credit. Write down the width, and whether it is a true width. Scan the QA/QC section. Then ask the only question that really matters: does this result change the size, grade or economics of the deposit as previously understood?

One last thing. An assay report is a technical document, not investment advice, and a single intercept is rarely enough to judge a project. If you are making a financial decision based on drilling results, take proper advice from someone qualified to give it.

Photo: Kindel Media / Pexels