In eighteen years serving customers directly at the atelier, I noticed a pattern: the buyers who come back happy do not always understand technique, but they all have criteria. Those who come back disappointed are exactly the ones who bought on impulse, with no filter. I will share the five tests I teach Herreira resellers to apply with customers — and that you can apply alone in any store before paying.
These tests are not theoretical. They are inspection protocol. Each one reveals a specific structural failure, and the presence or absence of the failure decides whether the piece will last three months or three decades. You do not need a microscope or a scale. You need a trained hand, and a hand is trained in a few minutes with clear criteria.
The counterintuitive thesis
Most buyers evaluate semijewelry by shine. Mistake. Shine is the only thing every new piece has, regardless of quality. New costume jewelry shines as much as premium semijewelry in the first thirty days. Shine is not diagnostic — it is packaging.
The five real diagnostics are: weight, magnet behavior, quality of reflected shine, edge and solder finishing, and the warranty offered by the seller. Each of these five reveals structural layers that packaging hides.
Test 1 — Weight in hand
Pick up the piece and feel the weight in your palm. Well-made semijewelry has a base of rolled brass or 925 silver — dense materials, with real weight sensation. Zamac costume jewelry (a zinc alloy) is too light — it feels like metalized plastic. The difference is tactile, immediate, unmistakable once you compare two pieces.
Brass density is about 8.4 to 8.7 g/cm³. Zamac is around 6.6 g/cm³. For a piece of the same volume, brass weighs 25 to 30% more. It is not a difference you need to measure with a scale — you feel it in your hand.
Beware of very light pieces sold as premium semijewelry: either they are hollow (a defendable structure on some pieces, but one that needs declaring), or they are zamac with cosmetic plating. Ask the seller for the base alloy. If the answer hesitates, it is zamac.
Test 2 — Magnet reaction
This test has a subtlety that confuses many. Touch a small magnet (a fridge magnet is enough) to the piece. The correct result depends on the type of piece.
Semijewelry with a pure brass base does NOT attract a magnet — brass is a copper-zinc alloy, both non-magnetic. Semijewelry with a 925 silver base does not attract either. Zamac costume jewelry does not attract (zamac is non-magnetic). But semijewelry with iron base and cosmetic plating — cheap product sold as semijewelry in popular markets — attracts a magnet immediately. It is a clear sign of fraud: a piece advertised as 18k semijewelry that sticks to a magnet is not semijewelry, it is painted metal.
The complication: the intermediate nickel plating — necessary in legitimate semijewelry between the brass and gold — is slightly magnetic. On pieces with thick intermediate plating, there can be small attraction to a strong magnet. It is expected behavior and different from fraud. To distinguish: a fridge magnet does not stick; an industrial magnet can attract slightly. Fraud with iron base sticks to any magnet, with force.
This test requires care, but it catches the most common fraud in the popular market.
Test 3 — Quality of reflected shine
Here the test is visual, but with criteria. Place the piece under natural light or neutral white LED light (5000K). Observe the reflection on the surface. Quality 18k plating has deep, stable shine, with no visible variation between points on the piece. Thin plating (half micron to 1 micron) has surface shine — looks pretty from afar but, when examined closely, shows micro-roughness, iridescent stains, areas with slightly different tone.
Under a jeweler's loupe (10x, costs about US$ 6 at beauty supply stores), the difference is sharp. Thick uniform plating has continuous and smooth surface. Thin plating has granulated texture, sometimes with small spots of oxidation already present even on a new piece.
On pieces with stones, observe the setting. Quality prongs are continuous, symmetrical, without burrs. Poor prongs have crooked tips, visible solders, or looseness felt when slightly moving the stone with a fingernail.
Test 4 — Edge, solder, and clasp finishing
Examine the piece at regions where parts join: edges, chain links, buckles, lobster clasps. On well-made semijewelry, solders disappear into the metal — you see no line, no drop, no color variation between joined parts. On poor pieces, the solder is visible: sometimes gray, sometimes yellow because of wrong material.
Bracelets with open links tend to give it away the most. Each link must look fused, not glued. If you can see where two ends met, the solder was poorly done.
The clasp is where the piece suffers most. A lobster clasp must open and close firmly, with no halfway hitch. A buckle must catch the right hole without excessive force. An earring back must hold — if it is loose from the factory, it will fall at the first party.
Herreira uses European clasps on every piece since 2014, precisely because they are the most critical components in real use. A failing clasp means a lost piece, even if the rest of manufacture is perfect.
Test 5 — The warranty offered
This is the most underestimated and also most revealing test. Ask the seller: "What is the warranty on this piece, and what does it cover?" A serious brand answers without hesitation, with a clear term (1 to 5 years is the common range) and detailed coverage (plating, solders, clasps, stones).
A brand offering less than 6 months admits the piece does not last longer than that — across the market, warranty is the floor of durability the brand accepts to honor. A brand without written warranty is selling the risk to you.
Herreira offers a 2-year warranty on every piece since 2010, with broad coverage (plating, solders, clasps, stones). On manufacturing defect cases, a new piece is delivered at no cost. This commitment is only defensible when the factory operates with a high technical floor and traced processes — it is the counterpart to the engineering in microns described in another article.
The 60-second method
Synthesizing: piece in hand, feel the weight (test 1). Touch a fridge magnet (test 2). Observe shine under neutral light (test 3). Examine edges, solders, clasp (test 4). Ask written warranty (test 5).
In 60 seconds you have a structural diagnosis. It is not a lab diagnosis, but it is enough to decide between buying safely or walking away. Apply these five tests on three pieces you already have at home. You will learn far more about semijewelry in ten minutes than in a week reading catalogs.
The decision every buyer needs to make
Buying semijewelry is choosing between two use models. Model A: cheap pieces, swap each season, recurring spending. Model B: well-made pieces, kept for years, concentrated spending. Both models are legitimate. But only one of them allows that ring your grandmother wore to remain on your hand twenty years from now.
The five tests do not decide the model for you. They decide whether the piece in front of you delivers what it promises within the model under which it was sold. Applied honestly, they are the simplest and most democratic filter that exists — they require no money, no technician, only two minutes of attention.
Next practical step
To understand what the piece is on the inside, read premium semijewelry vs costume jewelry and 18k plating in depth. To understand why these five tests work, read electrodeposition step by step. If a specific question was not covered here, the Academy FAQ answers 25 questions I most often hear in direct service.