Vol. I · May 2026
put a ring on it
An editorial on the small, circular things we keep
Journal/Article

How do I choose a sapphire for an engagement ring?

The sapphire that sold me on sapphires sat in a tray on 47th Street for three months because nobody photographed it well. 1.4 carats, unheated Ceylon,...

The sapphire that sold me on sapphires sat in a tray on 47th Street for three months because nobody photographed it well. 1.4 carats, unheated Ceylon, slightly violet in daylight and nearly indigo under a halogen lamp. Choosing a sapphire for an engagement ring starts there, knowing what you're looking at before you pay for a color that only exists under the dealer's light. A good sapphire does not need a jeweler's lamp to justify itself.

Most clients ask about size first. I understand. A one-carat diamond looks like a one-carat diamond. A one-carat sapphire can look like a gray pebble or a tiny violet fire depending on the cut and origin. I'd rather you spend less and get the right color than spend more and get a big, flat window.

Color is not just "blue"

Sapphires are corundum with trace elements, usually iron and titanium, sometimes chromium. The blue ones can be pale steel, violet, or a greenish teal. The most valuable blue is a pure blue with medium to medium-dark tone and strong saturation, no gray, no purple, no black. Dealers used to say "cornflower" for Kashmir and "royal" for Burmese. Those words now get thrown at stones from anywhere. What you want to see in person is a blue you can name: not quite navy, not quite periwinkle. I like stones with a bit of violet because they do better in candlelight and dim restaurants. A pure blue can go black in low light.

Gray deadens it. Purple shifts it toward tanzanite. Black eats the brightness at the edges. Those are the three killers of a sapphire's color. If you're choosing a sapphire for an engagement ring, start with one rule: buy the stone in the worst light, not the best.

Budget reality: a fine unheated Ceylon blue in the one-carat range will often run $1,800 to $3,200 per carat. A heated one of similar color might be $900 to $1,500. A certified Burmese unheated stone with good color can hit $12,000 to $20,000 per carat. Then there is Kashmir, which starts at around $25,000 to $30,000 per carat for anything decent and goes up fast. Those are not fun numbers. They matter, because the market pushes clients toward "blue sapphire" listings where the color is fixed in Photoshop and the origin is a guess.

Origin is priced in, but not destiny

I am going to say something that will annoy some dealers. Origin matters, but mostly as a price floor. Kashmir, Burma, Ceylon (Sri Lanka), Madagascar, Montana. You cannot tell those apart reliably with your eyes, and anyone who says they can is lying or selling. Kashmir stones have a soft, velvety blue that comes from microscopic inclusions giving a slightly sleepy glow. Burma tends toward a pure royal blue, often with rutile silk. Ceylon runs lighter, a bit violet and bright. Madagascar is all over the place. Montana sapphires are found in more varied colors, often teal or green, and they have a legitimate appeal for buyers who want a stone with a shorter supply chain. But a mediocre Kashmir will never beat a great Madagascar at a fair price. I'd set a well-cut heated Burma with strong color before I'd set a sleepy overpriced Kashmir, and I've done exactly that for a client named Daniel last spring because his first stone was a "Kashmir" from an online estate listing that turned out, under the microscope, to be a heavily included Ceylon. That cost him a restocking fee and a lesson.

Treatment is the question you can't skip

About 95% of sapphires on the market are heated. The heat dissolves some of the rutile silk and improves clarity and color. It is permanent and stable and accepted in the trade. That's fine. I set heated sapphires all the time. What is not fine is when a seller omits it and prices the stone like it is unheated. Unheated sapphires command a premium because they are rarer, not because they are always prettier. If a lab report says "no indications of heating" from GIA, AGL, SSEF, or Gübelin, that's real. If the seller says "probably unheated," it's heated. I'd rather you buy a heated sapphire with documentation than an unheated one with a story.

Skip stones described as "diffused" or "beryllium-treated" unless they're dirt cheap and you like the color. The treatment can be shallow, and re-cutting or heavy polishing can remove it. Not a good situation for an engagement ring that might get re-polished over decades.

Cut is where sapphires go wrong

Diamonds are cut to maximize light return. Sapphires are cut to maximize carat weight. That is the shape of the whole problem. You want to see even color face-up, no clear patch in the middle where light passes straight through, because a stone with a large window in the table, a tilted culet, or a belly so deep that the color concentrates at the edges will look dead in a setting. Hold the stone over a printed page. If you can read the letters through the center, it has a window. Put it back. A slightly deep stone can be okay if the color stays saturated, but a shallow one almost never will be.

My usual advice: for an engagement ring, look at oval, round, and cushion cuts in the five to eight millimeter range. That's roughly 0.9 to 1.8 carats depending on depth. Emerald cuts in sapphire are beautiful but show color zoning more than ovals. I do not like pear sapphires in prongs. The tip is fragile unless you use a V-tip, and even then the color can sit oddly.

What setting actually works

You have to choose a sapphire for an engagement ring with the setting in mind, not after. A bright blue stone with a slight gray edge will look different in yellow gold than in white gold. Yellow brings out warmth and can hide a little gray. White gold and platinum can make a violet-blue stone look icy. I usually suggest 18k yellow for blue sapphires with any gray or green undertone, and white gold for stones that are already pure blue or violet. Rose gold is a specific look, but it clashes with some blue shades. Bring the stone to your jeweler and lay it against each alloy, in daylight.

Prong settings are the default for sapphires. Fine. Six prongs on a round, four with V-tips on an oval or cushion. A full bezel is even safer for daily wear and I like it on smaller stones, but it steals a bit of light. If the stone has a vivid color, a bezel is a great choice. If the color is a touch sleepy, don't bezel it.

Lab-grown sapphires are a different conversation

Lab-grown sapphires are real sapphires. Same crystal, same hardness, same look. They cost a fraction of earth-mined, often $150 to $400 for a nice one-carat blue. The only reason not to buy one is if you want an earth-mined stone for the rarity or because it matters to you. I set lab sapphires, and I'll tell you plainly: for an engagement ring on a practical budget, they are the smartest way to get color without compromise. Just don't pay earth-mined prices for one. A lab-grown sapphire should never be sold as an investment, and neither should most earth-mined ones.

Questions to ask before you pay

I keep a short list taped to my bench for clients who are shopping elsewhere:

The stone I'd buy

If I were buying a sapphire for an engagement ring tomorrow, and someone had set a $4,000 budget, I'd buy a heated Madagascar or Ceylon blue around 1.2 carats, well cut, with a GIA report stating heat only, no residue, and set it in an 18k yellow bezel or six-prong basket. That ring would cost maybe $5,200 finished and would look better in ten years than a lot of $15,000 ones. If the budget were $15,000, I'd hunt for a one-carat unheated Ceylon or Burmese with a violet edge and a clean face-up. If the budget were $25,000, I'd start asking about Kashmir, but only if the stone had papers from a lab I trust. Above that, you're buying rarity more than beauty, and you should know that before you write the check.

A sapphire engagement ring does not need to look like a diamond alternative. It should look like the stone you keep looking at when you're supposed to be doing something else. That's the only test I trust.

Written by
Renee Alexander
Continue Reading

Are lab-grown diamonds more ethical than mined ones?

The question assumes a clean binary where none exists. Lab-grown diamonds can be the more ethical choice, but they aren't automatically, and the marketing...