How to Spot Fake Gold Jewellery: 7 Reliable Tests

Start with the hallmark, not the metal

Before you touch a magnet, read the piece. Most gold sold legitimately carries a fineness mark — 375, 585, 750, 916 or 999 in the metric system, or 9K, 14K, 18K, 22K and 24K where karat marks are used. Those numbers are parts per thousand: 750 means 75% gold by weight, which is what "18 carat" describes.

What matters is not the presence of a mark but its quality. Struck marks on genuine pieces are crisp, consistently deep, and sit where a jeweller would put them — inside a ring shank, on a clasp tongue, on the reverse of a pendant bail. Fakes tend to have marks that are shallow, smeared, oddly spaced, or laser-etched onto a surface rather than punched into it. A mark that looks printed is a warning.

Be equally suspicious of marks that say too much. "18K GP", "18K GF", "18K HGE" and "RGP" are not gold purity marks at all — they read gold plated, gold filled, heavy gold electroplate and rolled gold plate. They are honest descriptions of a base-metal item with a gold layer, and they are frequently misread by buyers who see "18K" and stop.

The magnet test

Real gold is not magnetic. A strong neodymium magnet will reveal most plated or filled fakes immediately, because the cheap cores used in volume counterfeits are usually steel or a ferrous alloy.

It is a screen, not a verdict. Brass, copper, lead and tungsten are all non-magnetic, and tungsten in particular is used deliberately because its density sits very close to gold's. A piece that fails the magnet test is almost certainly fake. A piece that passes has only survived the first round.

The density test

Gold is unusually dense at 19.3 g/cm³ — roughly two and a half times the density of steel. Weigh the piece dry, then suspend it in water to measure the volume it displaces, and divide mass by volume.

This test is far more informative than the magnet, and it is the one most home counterfeits fail. It also has two known blind spots. Tungsten is the notorious one, at 19.25 g/cm³ — close enough that displacement alone will not separate it from gold. And any piece with a hollow section, a stone setting, or a non-gold clasp will read low for entirely innocent reasons. Use density on solid, unmounted items: bars, plain bands, chains without findings.

The acid test

Acid testing kits are inexpensive and verify karat reliably. Each bottle is a nitric-acid solution mixed to dissolve everything below a stated purity, so the piece is tested against a ladder: if the 14K solution attacks it, it is not 14K.

Two practical points. First, it is a destructive test — you are removing metal, and on a plated piece you need to cut or file through the plating to reach the core, which means testing a hidden surface, not the visible one. A drop on the outside of a heavily plated fake reads as gold because it is touching gold. Second, acid kits have a shelf life; solutions that have gone off give false passes. Wear eye protection, work over a ceramic dish, and never test a piece you would mind marking.

The professional XRF test

For anything you believe is worth a meaningful sum, pay for X-ray fluorescence testing at a jeweller or assay office. XRF is non-destructive, fast, and reports the full alloy composition rather than a pass/fail.

Its one genuine limitation is depth: XRF reads the surface, typically a few micrometres in, so a thick plating over a base core can still fool it on a single reading. Reputable testers know this and combine XRF with a specific-gravity check or an ultrasound thickness reading. If someone offers to buy your gold and tests it with XRF alone, that is normal for scrap trade — but for a piece you are buying as solid gold, ask what second method they used.

What each test actually proves

No single test is conclusive. Two independent tests that agree — say, density plus acid on a filed surface — are worth far more than four repeats of the same one.

Where fakes actually come from

The pattern is consistent: fakes reach buyers through channels with no verifiable chain of custody. Marketplace listings with stock photography, holiday-market stalls, private sellers who insist on cash and speed, and "estate clearance" sales with no paperwork. Genuine dealers do not resist testing; a seller who becomes evasive when you mention an independent assay has already answered the question.

Pricing is the other tell. Gold trades on a public spot price, and no legitimate seller is unaware of it. An offer materially below the melt value of the stated weight is not a bargain — it is a statement about what is actually in the piece.

Buying authentic in the first place

The surest defence against fakes is to buy certified bullion from an established dealer rather than chasing bargains from unknown sellers. Silver Gold Bull and Gold Avenue (MKS PAMP) supply LBMA-certified coins and bars with verifiable assay documentation, which removes most authentication risk before it starts. Precious-metals prices are volatile and capital is at risk when you hold or trade gold; nothing here is financial advice.