How to Evaluate a Pakistani Mining Claim: A Due Diligence Framework I Wish I'd Had

I bought my first mining claim in Gilgit Baltistan in 2019. I overpaid. By about 40%.

The seller showed me hand samples that sparkled, a license that looked official, and a road that supposedly reached the site. Two of those three things turned out to be true. The samples weren't from the claim — they were from a friend's claim 18 km away. I figured this out three months later when I actually visited.

So when people ask me how to evaluate a Pakistani mining claim, I don't give them the textbook answer. I give them the version I wish someone had handed me before I wrote that check.

Start with the paperwork, not the rock

Everyone wants to talk about the mineral first. Wrong order.

The single biggest risk in Pakistani mining isn't geology. It's title. I've seen claims sold three times to three different buyers, all with "original" documents. I've seen leases expire two months before the sale closed (seller knew, buyer didn't). I've seen claims where the surface rights and mineral rights belonged to different parties who actively hated each other.

Here's what you check, in this order:

1. The lease itself — issued by which provincial department, valid until when, for which specific minerals. A chromite lease doesn't let you mine the copper underneath. This catches roughly 1 in 4 first-time buyers.
2. The royalty status. Unpaid royalties travel with the claim. You buy the mine, you inherit the debt.
3. The coordinates. And I mean actually plot them on a map. I've seen lease documents where the GPS points formed a polygon that didn't close — physically impossible boundaries that nobody noticed for years.
4. Overlap with protected areas, military zones, or contested tribal land. Especially in Balochistan and KP.

If any of this is missing or fuzzy, walk. Don't negotiate harder. Walk. There are 200 more claims behind it.

Then verify the geology — before you visit

This is where GeoMine AI started, honestly, because I got tired of flying to sites that turned out to be nothing. A round trip to Skardu costs me time I don't have. So we built a system that pulls Sentinel-2, ASTER, and SRTM DEM data for any coordinate set in Pakistan and tells you what's actually there spectrally.

But you don't need our platform to do the basics. You need to answer four questions before you book the flight:

What does the host rock look like from orbit? ASTER bands 4 through 9 will tell you if the alteration mineralogy matches what the seller claims. If someone's selling you a porphyry copper prospect and there's no argillic or phyllic alteration signature anywhere near the claim boundary — that's a problem.

Is there structural control? DEM data shows you faults, lineaments, intersections. Most economic deposits in Pakistan sit on or near major structures. Reko Diq isn't where it is by accident. Neither is the Saindak system. If the claim shows zero structural complexity, the seller's optimism is doing a lot of heavy lifting.

Has anyone else worked the area? GSP reports, old BHP and Tethyan filings, university theses from Peshawar and Quetta. Half of Pakistan's mineral knowledge sits in PDFs nobody's indexed properly. Spend a week reading before you spend a day driving.

What's the access reality? I mean actually — can a truck get there in winter? Can you run a drill rig in? One of my Gilgit claims looked fantastic on every metric except the last 4 km were a goat path. That changes economics fast.

The site visit (what to actually do when you're there)

Okay, paperwork's clean, satellite signals look real, reports support the story. Now you go.

Bring a GPS. Bring it for you, not the seller. Walk the corners of the claim yourself. I cannot tell you how many times the "claim boundary" the seller walks you along is conveniently 200 meters inside the actual lease — keeping you away from the part that's already been high-graded or doesn't outcrop the way they implied.

Take your own samples. From locations you choose. Sealed in front of you. Sent to a lab you picked (SGS Pakistan or Bureau Veritas, not the seller's cousin's lab in Lahore). And take twice as many as you think you need, because the second round of assays — the one that confirms or kills the deal — is going to want context samples you didn't think to grab.

Talk to the locals. Not through the seller's translator. The person running the tea shop nearest the site knows more about who's been digging where than any document will tell you. I learned this the hard way on a marble claim near Mardan where the "untouched" deposit had been quietly quarried for six years by a neighboring family.

What I'd add now that I didn't know then

Look, mining claim due diligence in Pakistan isn't complicated. It's just unforgiving. The mistakes are expensive and most of them aren't recoverable because the legal system moves slowly and sellers disappear quickly.

The framework I use now on every claim — mine or a client's — is roughly: 40% of the work is title and legal, 35% is remote sensing and historical data, 20% is the physical visit, and 5% is gut feel about the people involved. That last 5% has saved me more money than the other 95% combined.

If the seller won't let you take your own samples, won't share the original lease document, won't introduce you to the patwari, or keeps pushing you to close before the next monsoon — you already have your answer.

And if you're staring at a claim right now wondering whether the satellite signatures actually match what's in the offer document, that's literally what we built geomines.org to answer. But even if you never use our platform, run the framework above. The 2019 version of me would've saved a lot if someone had just written this down.