Somewhere on the spec sheet for every diamond you have been shown, underneath the four grades everyone talks about, sit two numbers with percent signs after them. Depth, then table. Most people glance at them, decide they look technical, and go and find a chart on the internet to check them against.
The chart is the problem. A diamond depth and table percentage are real measurements of a real thing, but they describe the shape of a stone rather than the quality of its cutting, and no laboratory publishes the pass mark those charts imply exists.
The misconception worth naming first is that these figures are a shortcut to the cut grade — a way of checking the laboratory's working, or of judging a stone that has no cut grade at all. They are neither.
The Diamond Depth and Table Percentage Are Ratios, Not Sizes
Both numbers are divided by the same thing. GIA defines total depth percentage as the table-to-culet depth expressed as a percentage of the average girdle diameter, and table size as the average of four table measurements, taken bezel point to bezel point, expressed as a percentage of that same average girdle diameter. The girdle is the narrow band separating the crown from the pavilion, and it is the stone's setting edge.
So the denominator in both cases is the width of the stone. That has a consequence people rarely follow through. A depth percentage can fall because the stone got shallower, or because it got wider. The figure moves either way and tells you nothing about which happened.
It also means neither number says anything about size. A 61.5 per cent depth describes a 6.4mm stone and a 4.1mm stone equally well. If you want to know how large the diamond is, the measurements line gives it in millimetres to two decimal places, and that line is unglamorous and exact and worth more of your attention than either percentage.
Both figures are printed on every GIA report, including reports for shapes that receive no cut grade. Being printed on everything is not the same as meaning the same thing on everything.
Neither Number Means Anything on Its Own
This is not a contrarian position. It is GIA's own finding, stated in the research paper that underpins the cut grading system: it is not any one proportion, but rather the interrelationship of all proportions, that determines whether a diamond will perform well enough to receive a top grade. To arrive at that, GIA ran calculations for almost one million proportion combinations, and concluded that many different proportion sets produce top-grade diamonds.
The mechanism is easier to see once you notice what depth percentage is made of. Table to culet means crown height, plus girdle thickness, plus pavilion depth. Three independent decisions by the cutter, added together and divided by the width. One figure, three inputs, and the figure does not tell you which input supplied what.
A stone can reach 62 per cent with a tall crown over a shallow pavilion. It can reach 62 per cent with a flat crown over a steep pavilion. It can reach 62 per cent with unremarkable angles at both ends and a thick girdle quietly carrying the difference. Those are three different diamonds that behave in three different ways, and a chart comparing depth percentages puts all three in the same column.
Table percentage has the same weakness in a milder form. A table is a window, and what comes back through it depends on the angles beneath. The same 57 per cent sits over good geometry and bad geometry without changing its number. This is the proportions version of an argument we have made about the report as a whole, in our piece on why two diamonds with identical certificates look different.
Every Ideal Proportions Chart Belongs to Someone With Stones to Sell
GIA grades cut from Excellent to Poor for standard round brilliants in the D-to-Z colour range, assessing seven factors: brightness, fire, scintillation, weight ratio, durability, polish and symmetry. What it publishes is grade ranges, not target figures. Its own wording is that each grade range represents a range of proportion sets, so there is plenty of room for personal preference within it, and that a nearly unlimited combination of proportions is possible.
Compare that to the tables in circulation, which hand you a narrow band of acceptable figures and imply a stone outside it has failed. Those bands differ from one website to the next, which on its own should tell you what they are. They are a sorting convenience, built by people with inventory to organise.
The proportion diagram already gives you the inputs. Crown angle is the angle between the bezel facet plane and the table plane, and GIA notes that 32 to 36 degrees tends to produce attractive results — when paired with appropriate table and pavilion proportions. The conditional clause is the entire point. Pavilion angle is the angle between the pavilion main facet plane and the table plane, and GIA is explicit that subtle differences there matter a great deal, with steep angles capable of producing dark areas under the table.
Girdle thickness is described in words rather than a percentage, from extremely thin to extremely thick, averaged across eight positions. Thin has durability consequences; thick can carry weight you paid for and cannot see, an argument made at length in our piece on finger coverage. Culet size runs from none to extremely large, and a large one pulls the grade down.
Read in that order, the depth percentage stops being a test and becomes the sum of decisions already described more usefully further up the page.
When There Is No Cut Grade, the Percentages Are All You Get
GIA does not issue an overall cut grade for fancy shapes — rectangular, oval, square, anything that is not a standard round brilliant. What those reports carry is polish and symmetry, a description of shape and cutting style, the measurements, and a simplified proportion diagram that includes girdle thickness.
Which leaves the two percentages doing more work than they can bear, on exactly the stones where they are hardest to interpret. Crown and pavilion geometry varies by shape and by cutter, so the same figure means something different on an oval, a pear and an emerald cut. Nobody has published the translation, because there is no standard facet arrangement to translate from.
The way through is to stop treating the figure as a verdict and start asking what produced it.
- What is this depth percentage made of? You want it broken into crown, girdle and pavilion. A jeweller who has handled the stone will answer without reaching for the report.
- What are the crown and pavilion angles, and are they working against each other? A good answer talks about the pair. One that quotes a single number and stops has missed the question.
- Where is the girdle thin? The report gives an average; the diagram shows the range. Thin places are a setting decision to make before anything is built.
- For a fancy shape, what am I comparing this against? The honest answer is that there is no published standard, and the person who says so is more use to you than the person who produces a chart.
- Can I see it face up beside a stone with different numbers? The only test that settles anything, and it takes about ten seconds.
The Stones We Work With Have No Version of This
We specialise in Australian sapphires, so read the following as evidence against our own interest. GIA's cut grade covers standard round brilliants in the D-to-Z colour range. A sapphire sits entirely outside that, whatever its cutting quality. No cut grade, no seven factors, no proportion diagram of the kind described above, and no percentage anyone has agreed to argue about.
All of the above is a discipline available to diamond buyers and unavailable to people buying the stones we sell most of. That matters, because the incentive that produces a deep diamond exists for coloured stones too. Stones are priced by weight, and depth is where weight hides. The difference is that for a diamond a laboratory checks the weight against the diameter as routine, and for a sapphire nobody does.
So when we put a sapphire in front of someone, the only check available is the last question in that list. We would rather say so plainly than let this vocabulary imply a rigour that does not exist on our side of the counter. For the diamond version of the same scrutiny, our guide to reading a diamond certification walks the document line by line.
How to Actually Use These Two Numbers
Treat them as a description, not a score. For a round brilliant, the cut grade has already read them alongside the angles, the girdle, the culet and the finish, which is more than any chart will ever do for you. For a fancy shape, there is no grade, so the percentages become a prompt rather than an answer, and the prompt is to ask what they are made of.
The two numbers are worth knowing and not worth deciding on. If a figure on a report ever makes you want to reject a stone you have not seen, that is the moment to go and look at it instead.
Book a consultation at our Surry Hills studio → Learn more about the Laher process →
Laher is a bespoke jewellery studio in Surry Hills, Sydney. We design and handcraft engagement rings and fine jewellery using Australian sapphires and unique coloured diamonds.