Reference

How to Tell if Gold Is Real: 7 At-Home Tests That Actually Work

Published July 2026 · 7 min read

You've found a ring at the back of a drawer, or inherited a chain, and the question is nagging at you: is this actually gold, or a good-looking fake? There are plenty of tests you can do at home before you spend anything on a professional opinion. Some are genuinely useful. Some only rule things out rather than prove them. And one is downright dangerous if you're careless. Here's how to work through them properly, and when to stop tinkering and pay for certainty.

Set your expectations first. No home test proves purity — the difference between 9ct and 18ct, say. The best they do is tell you whether you're probably holding real gold or definitely holding something else. For the actual carat and value, you need the hallmark and, ideally, a professional scan.

1. Look for a hallmark first

Before any clever trick, get a magnifier and hunt for a stamp. In the UK a genuine gold item usually carries a hallmark: a fineness number (375, 585, 750, 916 or 999), a sponsor's mark and an assay office symbol. That mark is a legal guarantee struck at an assay office, and it's worth more than every home test combined. If you find a clear British hallmark, you're most of the way there. Our hallmarks guide shows you what each mark looks like.

A stamp isn't always the whole story Small or foreign items may be unmarked yet real, and, rarely, fakes carry forged stamps. So treat a hallmark as strong evidence, then confirm with a test if the value is significant.

2. The magnet test

Gold is not magnetic. Grab a strong magnet — a fridge magnet is too weak, so use a neodymium one if you can — and bring it near the piece. If the item jumps to the magnet, it contains a magnetic base metal like iron or steel and it isn't solid gold. This is a great quick disqualifier. The catch: plenty of fakes use non-magnetic metals like brass or copper, so passing the magnet test doesn't prove anything on its own. It only catches the crude fakes.

3. Skin discolouration and green marks

Real gold doesn't react with your skin. If wearing a ring leaves a green or black mark on your finger, that's usually copper or nickel in a base-metal alloy reacting with sweat. Solid gold jewellery won't do this. Be fair to the test, though: even real low-carat gold is alloyed with other metals, and heavy plating can hide a base core for a while. A green finger is a red flag, not a verdict.

4. The ceramic streak test

Find an unglazed ceramic tile or the rough underside of a bathroom tile. Drag the piece gently across it. Real gold leaves a gold-coloured streak. Fake, or heavily plated base metal, tends to leave a black or grey streak. It's a mild test and it will scratch the item slightly, so only do it somewhere that won't show. Like the others, it's better at spotting fakes than confirming purity.

5. The density (water displacement) test

This is the most scientific check you can do at home, and it's surprisingly telling because gold is dense — about 19.3 grams per cubic centimetre, far heavier than most metals used to fake it. Here's the method:

  1. Weigh the item in grams on an accurate scale.
  2. Half-fill a measuring container with water and note the level in millilitres.
  3. Drop the item in and note the new level. The difference in millilitres equals the volume in cubic centimetres.
  4. Divide the weight by the volume. That's the density.

If the result lands near 19.3 g/cm³, you're likely looking at high-purity gold. Lower-carat gold reads lower because of the alloy — 9ct sits closer to 11–13 g/cm³ — so this test hints at carat as well. It's not lab-precise; small items and clinging air bubbles throw the reading off. But a chain that comes out at 8 g/cm³ is not gold, whatever it looks like. Once you've confirmed it's real and know the rough carat, you can put the weight into our gold calculator for a value estimate.

6. The acid test — handle with real care

Acids are corrosive and dangerous Testing acids can burn skin, damage eyes and give off harmful fumes, and they will mark or damage the piece. If you go ahead, wear gloves and eye protection, work in a ventilated space, and never rush. Honestly, for most people this is the point to hand over to a professional instead.

Acid test kits work because different acid strengths dissolve lower-carat golds but leave higher-carat golds untouched. You make a tiny scratch on a testing stone, apply the acid for a given carat, and watch whether the mark dissolves. It can indicate carat, but it's fiddly, permanently marks the item, and gives plated fakes a chance to fool you if the plating is thick. Given the risks and the mess, most people are better off skipping this one.

7. The "ping", the weight in the hand, and when to pay for XRF

Experienced dealers rely partly on feel. Real gold has a heft to it and, when tapped, gives a soft ring rather than a dull clank. These are instincts, not proof, and they take years to trust. Which brings us to the honest bottom line.

Every home test above either rules fakes out or gives a hint. None of them tells you, reliably, that you're holding 18ct rather than 14ct — and that distinction is the difference between a good payout and a poor one. The tool that does is XRF (X-ray fluorescence), a scanner most gold buyers and jewellers own. It reads the exact metal composition in seconds, without damaging anything, and it's usually free when you're getting a quote.

So use the home tests to satisfy your curiosity and weed out obvious fakes. But when there's real money at stake, let a genuine hallmark and a professional XRF scan settle it. If the colour of your piece is throwing you — white or rose rather than classic yellow — read white, yellow and rose gold explained, because colour has nothing to do with whether it's real and everything to do with the alloy. The safest path is always the same: check the hallmark, confirm with a professional, then work out what it's worth.


All figures are estimates of melt value, never offers. Not financial or tax advice.

Value your gold now →