What kind of gold is it?
A recognized bullion bar, retail coin, signed jewel and mixed scrap lot can contain the same metal but trade in different markets.
Gold guides and tools
Understand what gold is worth.
Move from a quoted price per troy ounce to the value of a bar, coin or piece of jewelry—without confusing purity, gross weight, melt value and a real buyer's offer.
Start with the object
Four questions before one answer
A gold quote becomes useful only after the object, unit, purity and transaction have been identified.
What is the net gold-bearing weight?
Gemstones, steel watch parts, springs, dirt and other non-gold components should not be valued as gold.
What is the tested fineness?
Karat and millesimal marks state a purity claim. A stamp helps classification, but it is not infallible proof.
Which price is being compared?
Wholesale spot, a benchmark auction, futures, retail asking prices and cash buyback bids are related but not interchangeable.
Gold value in plain language
The basic calculation
Contained-gold value starts with three inputs: net gold-bearing weight, fineness and a comparable reference price. A real sale then adds or subtracts product premiums, testing, refining, risk, service and buyer margin.
Precious metals are commonly quoted by the troy ounce. One troy ounce is exactly 31.1034768 grams. It is not the ordinary avoirdupois ounce used for many household goods. Converting the unit correctly is the first step toward a defensible estimate.
Fineness is the share of an alloy that is gold. Karat expresses that share in twenty-four parts: 18K is 18/24, or 75% gold; 14K is 14/24, or about 58.33%. Millesimal fineness states parts per thousand, so 750 and 18K describe the same nominal gold share.
fine-gold weight = net gold-bearing weight × fineness reference melt value = fine-gold troy ounces × reference price per troy ounceMelt value is a benchmark calculation, not a promise that a buyer will pay that amount. A buyer may need to verify purity, remove non-gold material, aggregate a small lot, fund the transaction and send the material to a refiner. Conversely, a rare coin, important maker, collectible design or marketable gemstone can make an item worth more than its metal.
Common gold karats and fineness
| Description | Fraction | Nominal gold | Common fineness expression |
|---|---|---|---|
| 24K | 24/24 | Trade description for nearly pure gold | 999 or 999.9, depending on the product |
| 22K | 22/24 | 91.67% | 916 or 917 |
| 18K | 18/24 | 75.00% | 750 |
| 14K | 14/24 | 58.33% | 583 or 585 |
| 10K | 10/24 | 41.67% | 417 |
| 9K | 9/24 | 37.50% | 375 |
Actual marking rules, permitted tolerances and customary standards vary by jurisdiction and period. Use the table to understand the arithmetic, then treat the mark as a clue to be verified when money depends on it.
Why identical weights can sell differently
A recognized bullion product may have an active resale market and a product premium. A broken chain may be bought primarily for recovery. A signed period jewel may be evaluated for maker, design, condition and provenance. Selling the last two items through the wrong channel can erase value that no melt calculator can see.
- Bullion: brand, authenticity, bar or coin condition and dealer demand can affect the spread around a reference price.
- Jewelry: craftsmanship, maker, gemstones and resale demand may matter in addition to recoverable metal.
- Coins: rarity, grade, legal-tender status and collector demand can outweigh metal value.
- Mixed scrap: uncertainty, small lot size, solder and non-gold parts usually require testing and processing.
The practical rule is simple: classify first, calculate second, compare net offers third.
Marks and tests are evidence
Visual inspection and hallmarks are useful screening tools. Touchstone and acid testing can estimate surface karat. Electronic conductivity instruments and X-ray fluorescence can provide non-destructive readings, but plating, shape, calibration and surface composition can mislead. Fire assay is highly accurate and destructive, making it more appropriate for representative samples and refinery lots than for an intact retail object.
No method makes sampling irrelevant. A precise test of the wrong part of a mixed or plated item can still give the wrong commercial answer.