GIA does not grade cut on fancy shapes: what 2.8 million reports reveal about the industry's invisible convention

Summary

GIA assigns an overall cut grade to one shape only. That shape is the round brilliant. Every fancy shape on the GIA grading scale receives polish and symmetry grades but no cut grade. Across 2,866,993 GIA-graded fancy-shape diamonds in the Carat Hunter index, 73% carry a populated cut grade in their retailer-supplied data. That cut grade is not from GIA. It originates with retailers and aggregators who fill the field for cataloguing reasons, then propagates through the listing pipeline. The industry convention is so consistent that a buyer reading three different listings of the same physical stone can see three different cut grades on the same GIA report. We have the data to show it. Data as of 22nd March 2026.

What GIA actually grades on what shapes

GIA's grading service has a specific scope per shape. Round brilliants get the full treatment: color, clarity, an overall cut grade (Excellent, Very Good, Good, Fair, Poor), plus separate polish and symmetry grades. The cut grade on a round is a real GIA assessment based on the stone's proportions, finish, and brightness measurements taken by trained graders.

Fancy shapes do not get an overall cut grade. The GIA report on any of those fancy shapes shows color and clarity and polish and symmetry. The line for cut grade is left blank. This is GIA's longstanding policy, documented on the lab's public information page and reflected on every GIA fancy-shape report issued since the cut grade system was introduced in 2006.

The reason is methodological. GIA's cut grade for round brilliants is derived from a model that maps proportion ratios (table size, depth percentage, crown angle, pavilion depth) to a measured optical performance. That model is calibrated to round geometry and does not generalise. Building an equivalent model for each fancy shape would require a separate calibration for each one, which GIA has chosen not to do. The cut field on a fancy-shape report is therefore deliberately empty. It is not an oversight, and a populated cut grade on a fancy-shape listing is not from GIA.

The cohort tells the story

We checked every GIA-graded diamond in our index for the cut field. The split is sharp.

For round brilliants, 2,517,130 stones (98.8%) have a populated cut grade. The remaining 1.2% are data-quality artifacts where the cut field was not captured during ingestion or comes from a retailer feed that omitted it.

For fancy shapes, the split inverts. Of 2,866,993 GIA-graded fancy-shape stones in the index, 2,096,440 (73.1%) have a populated cut grade. Only 770,553 (26.9%) correctly show an empty cut field that matches the actual GIA report.

Per-shape, the percentage with retailer-derived cut grades varies from 67% on radiant up to 85% on asscher.

• Asscher: 84.8% have retailer-derived cut grade.
• Princess: 77.5%.
• Other fancy: 79.7%.
• Emerald: 74.3%.
• Oval: 73.9%.
• Pear: 72.8%.
• Heart: 72.0%.
• Marquise: 69.8%.
• Cushion: 68.1%.
• Radiant: 66.8%.

If GIA were grading cut on fancy shapes, we would expect 0% of stones to lack a cut grade. We see 27% lacking it. If retailers were uniformly faithful to the actual GIA report on fancy shapes, we would expect 0% to have a populated cut grade. We see 73% populated. The truth is in the gap.

The literal proof: one cert, two answers

Cert GIA 6542641331 is a 2.51ct natural H VVS2 oval at 11.08 by 7.37mm with EX polish and EX symmetry. The actual GIA report leaves the cut field blank because GIA does not grade cut on ovals.

In our index, this cert appears twice. One ingestion of the cert (sourced from one retailer feed) records the cut field as "GD". A second ingestion (sourced from a different retailer feed) records the cut field as empty. Same physical stone. Same public GIA report. Two different cut values in the data because each retailer's feed handled the empty cut field differently.

Anyone can verify this on the GIA report check tool by entering 6542641331. The report that loads will show polish and symmetry but no overall cut grade. The "GD" cut value that appears on some listings is a retailer-applied label. The empty cut value that appears on others is the faithful read.

This is the cleanest possible illustration of the convention. The convention is not malicious. Retailers fill the cut field on fancy shapes because their cataloguing software requires a value, because buyers expect to filter by cut across all shapes, and because the listing pipeline has standardised on populating that field by some means. The means is internal estimation, not a GIA grade.

Why retailers populate the cut field

First, buyer expectation. A buyer searching for a diamond expects to compare cut grades across the shapes they are considering. A round-vs-oval comparison where the oval has no cut grade reads to most buyers as missing data, not as accurate reflection of GIA policy. Retailers fill the field to make their listings comparable to round listings.

Second, software constraint. Most retailer inventory management systems treat cut as a first-class field on every diamond. An empty cut field on a fancy shape requires the system either to display an empty cell (which looks like a data error) or to omit the field (which breaks the comparison view). Filling the field is the path of least friction.

Third, internal estimation. Some retailers do compute an internal cut estimate based on the proportions printed on the GIA report (table percentage, depth percentage, length to width ratio). That estimate may be reasonable. It is still not a GIA grade. The visible label "EX" or "VG" on the listing does not distinguish between a GIA-graded round and an internal estimate on a fancy shape, and most listing pages do not annotate the source.

The cohort distribution of cut values on fancy shapes reflects this. On GIA ovals in our index, 40% of populated cut values are EX, 24% are VG, 11% are ID, and 11% are GD. The shape of that distribution is roughly consistent with what an internal-estimation model based on industry-standard proportion thresholds would produce. It does not match what GIA's actual grading would produce, because GIA does not grade cut on ovals.

What buyers should actually do

When comparing fancy-shape diamonds, treat the cut grade as informative but not authoritative. The polish and symmetry grades on a fancy-shape GIA report are real GIA grades and worth reading carefully. The cut grade is a retailer-applied estimate that varies by retailer.

When evaluating an oval or pear specifically, lean on the proportions rather than the cut label. The length-to-width ratio for ovals typically runs 1.35 to 1.50. The depth percentage ideal range is 60% to 62%. Table percentage on ovals runs 53% to 58%. Those proportions tell you more about how the stone will perform than a retailer-derived "EX" tag.

When the cut grade is the deciding factor in a fancy-shape purchase, ask the retailer where the cut grade comes from. A retailer that has computed an internal estimate from the GIA proportions may be transparent about that. A retailer that simply assigned EX to every stone in their fancy-shape inventory is doing something different. The answer to the question changes how you should weight the cut label.

Limitations

The 73% figure represents listings as ingested in our index, not the universe of all GIA-graded fancy-shape stones. Retailers vary in how they handle the cut field, and our cohort overweights retailers whose feeds we have integrated. The directional finding (most fancy-shape listings carry retailer-derived cut grades) holds across reasonable cohort definitions.

GIA's policy on fancy-shape cut grading is stable and well documented. The cohort comparison in this article therefore tests the listing pipeline's adherence to the policy, not the policy itself. A future change in GIA's grading services for fancy shapes would invalidate the analysis.

The cut estimate that retailers produce for fancy shapes is not necessarily wrong. A well-designed internal estimate based on GIA proportions can produce a label that correlates with how the stone performs visually. The article's claim is that the label is retailer-applied and labelled in a way buyers cannot distinguish from a real GIA grade. The claim is about transparency, not accuracy.

Methodology

We applied a simple test to every diamond in our index where the lab is GIA. We checked whether the cut field is populated with a non-empty value. We grouped by shape class (round versus fancy) and by individual fancy shape. We did not normalise the cut values across retailers because the unnormalised distribution is the relevant signal. The inconsistency of vocabulary across populated cut values is itself evidence that the field is filled by independent retailer systems. We see EX and Excellent and ID and Ideal and VG and Very Good and GD and Good all in the same column. That mix would not exist if the cut grade were copied from a single source.

The cert pair illustration uses GIA 6542641331 because the same cert appears twice in our index with different cut values. We selected it as a representative case rather than a unique anomaly. Across our cross-retailer cohort, similar split-cut-value cases are common on fancy-shape certs, and the cohort percentage of fancy-shape stones with retailer-derived cut grades is the relevant aggregate.

Data refreshes daily for active listings. The numbers in this article are accurate as of 22nd March 2026. The analysis will be refreshed quarterly. For the full retailer-inclusion criteria and matching algorithm specification, see the Carat Hunter methodology page.