How to Authenticate Gold Jewellery Before You Buy

The Short Answer

Authenticate in order, from free to paid: the seller and the paperwork first, then the hallmark, then non-destructive checks — magnet, weight for size, wear at the contact points — then a density test on plain solid pieces, and finally a jeweller’s XRF reading or a professional valuation for anything valuable, antique, signed or set with stones. In the UK, gold items of 1 g or more must be hallmarked before they can be described as gold, so an unmarked piece above that weight sold as gold is a red flag; below it, an unmarked chain may be perfectly genuine.

How far you go depends on where you are buying. The table below sets out the checks that matter in each situation; the sections after it explain each check and its limits. For hands-on testing step by step, see our tutorial on testing gold at home.

Start With the Hallmark

Find the hallmark first — it may be inside the ring band, on the clasp, on the back of a pendant, or inside a watch case back. The fineness mark (375/585/750/916/999) must match the seller's claim. Pre-1973 UK pieces use older hallmark formats; cross-check with the assay office's guides if the marks look unfamiliar.

Physical Tests

Magnet test: gold is not magnetic. A magnet catching the piece reveals gold-plated steel. Weight test: gold is dense (19.3g/cm³). A piece that feels light for its size is suspicious. Visual inspection: worn plating shows at edges and high points; under a 10x loupe, a different metal colour at the wear points is decisive.

Acid Testing

Jewellers' acid test kits are inexpensive and bracket the karat of the surface metal. Scratch a small line on a testing stone, apply a drop of the relevant acid, and the reaction (or lack of it) confirms karat. For high-value pieces, have a jeweller do this rather than risk the piece yourself.

Signed Piece Authentication

For a signed piece, the question is not only the metal but whether the house made it. The authority is the maker or its authorised boutiques: ask the brand what verification or servicing route it offers for the piece, and expect it to inspect the item rather than a photograph. For gold watches, the manufacturer’s service centre can confirm whether a case and movement are genuine. Paperwork from the original sale — receipt, certificate, box — supports the case but can be faked, so it is evidence, not proof.

Professional Route

For a piece where being wrong would be expensive, an independent valuation by a qualified valuer — for example one holding a gemmological qualification — or a jeweller’s XRF reading is worth paying for before the sale closes. Ask the seller whether you can have the piece inspected first, or buy on terms that let you return it; a reputable seller will agree, and one who refuses has told you something.

Authentication-Free Alternative

If you want gold exposure without the authentication burden, buy bullion from an established dealer like SilverGoldBull. Sealed and assayed bars or sovereign coins arrive with certification and need no separate authentication.

Weight and Density: the Test That Catches Most Fakes

Gold is unusually dense, and that is the most useful physical property available to you.

The method needs only a scale accurate to 0.01g and a container of water. Weigh the piece dry. Place a water container on the scale and tare to zero. Suspend the piece fully submerged on a thin thread without touching the sides or bottom, and read the displacement — in grams, that number equals the volume in cubic centimetres. Divide dry weight by volume.

Pure gold is about 19.32 g/cm³. 22k lands near 17.7, 18k near 15.2–15.9, 14k near 12.9–14.6, 9k near 11.3. Brass and bronze sit near 8.5, silver near 10.5. Any of those is unmistakably different.

The limits matter as much as the method. It gives nonsense on hollow pieces, on anything with stones set in it, and on items with clasps or fittings of another metal. And it cannot distinguish gold from tungsten, whose density is almost identical — which is the specific reason bullion bars, rather than jewellery, need more than a density check. Our step-by-step testing tutorial covers all five methods and where each one fails.

Even so, for a plain solid chain or a plain band, density is the single most informative thing you can do at home in ten minutes.

Reading Wear as Evidence

Wear tells you things a chemical test does not, and it is free.

Look at the contact points under a loupe: the underside of a ring shank, the edges of a clasp, the high points of any engraving, the back of a pendant where it rubs a collar. Plating wears through here first. A different colour showing beneath — yellower, pinker, or grey — settles the question immediately.

Look at consistency. Solid gold wears evenly and stays the same colour throughout. Plated items often show subtle colour variation where the layer is thinner.

Look for a seam. Hollow items have one. A chain that feels implausibly light for its size usually is.

Look at the solder joints. On good work they are the same colour as the surrounding metal. Visible silver-coloured or grey solder indicates a repair, a mixed-metal construction, or lower-quality work — not necessarily fake, but not what a signed piece should look like.

Consider the wear pattern against the claimed age. A piece described as decades old with no wear at all on any contact surface is either exceptionally cared for, refinished, or newer than claimed.

None of these is conclusive alone. Together they usually settle it, and they are the checks an experienced buyer runs before touching a chemical.

Common Counterfeit Constructions

Knowing what the fakes actually are makes them easier to spot.

Gold-plated base metal is the commonest. A thin electroplated layer over brass. Detected by wear points, weight and density, and by an acid test on a deep scratch.

Gold-filled items sold as solid. These contain real gold in quantity and are legitimate products in their own right — the fraud is in the description. Density catches it; a surface acid test does not, because the surface genuinely is karat gold. Our gold-filled guide explains the constructions.

Hollow items sold as solid. Common in chains and bangles. Weight relative to size is the tell, and a seam is usually visible.

Lower karat sold as higher. 9k stamped 18k. Acid testing brackets it; XRF measures it exactly.

Genuine gold with a base-metal core. The hardest jewellery case. Density catches most core materials.

Counterfeit hallmarks. Struck fakes exist. A genuine struck mark displaces metal and shows a slight raised lip under magnification; cast reproductions look flat and soft. Our hallmarks guide covers how to read one properly.

Fake branded pieces, where the metal may be genuine gold and the signature is not. Here the metal test passes and the question becomes brand authentication — serial numbers, packaging, documentation and the quality of the finishing.

When to Stop and Pay Someone

Home testing has a ceiling and it is worth knowing where it is.

Stop and use a professional when: the piece is valuable enough that being wrong is expensive; it is antique, where destructive testing damages value; it is signed, where brand authentication rather than metal is the question; it has stones, where the metal may be the smaller part of the value; or you have run the home tests and the results disagree with each other.

What a professional offers: XRF measures elemental composition non-destructively in seconds and most jewellers have one, frequently at no charge. Specific gravity on laboratory equipment does density properly. Ultrasonic and conductivity testing finds cores. And an experienced eye on construction, finishing and marks is often the most valuable part.

For branded pieces, the manufacturer or an authorised dealer is the authority, and for significant stones a gemmological laboratory report is the document that matters.

The cost of all of this is small relative to the value at stake, and the honest framing is that home testing is for triage — deciding what deserves a professional look — rather than for final answers on anything substantial.

Sources

• businesscompanion.info/en/quick-guides/miscellaneous/hallmarking
• gov.uk/online-and-distance-selling-for-businesses
• hallmarkingconvention.org/en/members

Frequently Asked Questions