Book Appointment
The CAD approval process is the last point where a custom piece is still a file, not metal — once you sign off, prong height, band profile, metal weight, and engraving placement are locked, and any correction after casting means scrapping and rebuilding the piece from scratch. The practical takeaway: never approve a CAD render on the first look. Check stone seating and prong height, confirm the metal weight against the original quote, verify the band profile matches what the client actually asked for, and lock the engraving placement and depth — every time, at every round. Jewelers who treat CAD sign-off as a formality (“looks good, go ahead”) are the ones eating remake costs and blown timelines. Jewelers who treat it as the actual quality gate rarely see a piece come back wrong.
This matters more for outsourced custom work than for anything made on your own bench, because you’re approving a file for a process you can’t watch happen in real time. There’s no walking over to a bench to eyeball a wax before it goes in the flask. The CAD file is your only checkpoint before the piece becomes physical and irreversible.
The CAD-to-Casting Pipeline, Step by Step
Every custom order moves through the same sequence, whether it’s a single engagement ring or a full custom jewelry build with a matching band. Understanding where CAD approval sits in that sequence is what makes the review step feel urgent instead of routine.
| Stage | What Happens | What You’re Actually Approving |
|---|---|---|
| Order intake | Center stone specs, metal type, band style, sizing, and engraving text are captured in a design brief | That the brief matches the client’s actual order — not a summary of it |
| Initial CAD render | A 3D file is built showing stone placement, prong structure, shank, and any halo or accent stones | First-look proportions, stone count, and overall silhouette |
| Revision round(s) | The file is adjusted based on your markup — geometry, weight, or placement changes | Everything from here on: seating, weight, profile, engraving |
| Final sign-off | You confirm in writing that the current file version matches the order | This locks the exact file that goes to casting — no more free changes |
| Casting | The approved file becomes a wax or resin pattern, then cast in metal in-house | Nothing — this stage executes exactly what you signed off on |
| Setting | Stones are set into the cast piece by hand, in-house | Nothing — setting follows the approved seating geometry |
| Finishing & QC | Polishing, plating/rhodium where applicable, and a final check against the approved CAD | The piece is checked against your sign-off, not against a fresh judgment call |
| Ship | The piece ships unbranded by default, ready for the jeweler to present as their own | — |
Notice where the leverage sits. Every stage after sign-off is execution, not judgment. Casting doesn’t second-guess the file. Setting doesn’t redesign the prong structure. QC checks the piece against what you approved — it doesn’t check whether what you approved was actually right. If an error is baked into the CAD file at sign-off, it ships baked into the piece. Our how it works page walks through the full order flow if you want the end-to-end picture; this post is about the one step inside it that determines whether the rest goes smoothly.
What to Actually Check at Each CAD Revision
A CAD render looks finished long before it is. Rendered lighting, reflections, and a clean turntable view make even a flawed file look production-ready. The four things below are what actually determine whether the physical piece matches expectations — and they’re the four things a rushed approval skips.
Stone seating and prong height
The default render angle most files arrive in is a three-quarter hero shot, optimized to look good, not to show functional detail. Rotate the file to a straight-on profile view and check that prongs clear the girdle enough to hold the stone securely without burying too much of the crown. Low prongs on a fancy-shaped center stone are the single most common seating error that only becomes visible once the piece is physically in hand — by which point it’s a remake, not a tweak.
Metal weight and estimated grams
CAD software estimates gram weight from the file’s volume and the specified karat and alloy. That estimate should match the assumptions baked into your original quote. If a revision round thickened the shank, added a cathedral setting, or changed the band profile, the weight shifts — and so does the actual metal cost. Catching a weight mismatch here is a two-minute fix. Catching it on the invoice after the piece is cast is not.
Band profile and comfort fit
This is the check jewelers skip most often because it feels like a style preference rather than a spec. It isn’t. Comfort-fit (rounded interior), knife-edge, and flat-interior profiles wear completely differently on the hand, and they’re not interchangeable after casting. This matters even more on eternity and wedding bands, where the profile is close to the entire value of the piece — there’s no center stone to distract from a shank that doesn’t match what the client pictured.
Engraving placement
Confirm placement (interior vs. exterior), depth, font, and spelling in the actual revision — not from the original order form, from the file in front of you. Engraving that sits too close to where the band will later need to be sized can get planished shallow or disappear entirely during a resize. This is also the easiest thing to approve without truly looking at, because it’s usually a small callout in the corner of the render rather than the centerpiece of the image.
What Rubber-Stamping an Approval Actually Costs You
“Looks good” is not a review. It’s a guess dressed up as one. Here’s the difference between actually checking each round and waving it through:
| Checked Every Round | Rubber-Stamped on Sight | |
|---|---|---|
| Stone seating | Caught in the render, adjusted before cast | Found during or after setting — stone reset or full remake |
| Metal weight | Confirmed against the original quote pre-cast | Discovered on the final invoice, or as a piece that feels too light or too heavy |
| Band profile | Matches what the client actually pictured | Rejected on delivery, back to CAD from scratch |
| Engraving | Placement and depth locked in the file | Wrong once cast — engraving depth can’t be added after the fact |
| Timeline | Adds one short round before casting | Restarts the CAD-to-cast cycle entirely; the finished-piece window resets |
| Who absorbs the cost | Built into the review step — effectively free | The jeweler eats remake metal and labor, or credits it to their own client as goodwill |
That timeline line matters more than it looks. Once a CAD file is properly signed off, casting through finishing on a finished custom piece typically runs 4-6 days. A remake doesn’t extend that window — it resets it, because the piece is going back to a wax pattern, not forward from where it was. A five-minute review round costs you nothing against that clock. A remake costs you the whole cycle again, on top of whatever commitment you already gave your own client.
Building a Review Habit That Actually Works
The jewelers who never get burned by this aren’t reviewing CAD files more slowly — they’re reviewing them more deliberately. A few habits make the difference:
- Assign one person to review every CAD file, and make it the same person every time. Whoever is closest to the original order spec — not whoever is fastest to reply to an email — should be the one signing off.
- Rotate the actual 3D file, not just the static hero-angle screenshot most files arrive with. A flat thumbnail hides prong height and profile issues that a 15-second rotation reveals immediately.
- Redline directly on the render — circle the prong, mark the engraving zone, note the gram estimate — instead of describing changes verbally over a call. Verbal feedback gets summarized, and summaries drop detail.
- Check the file against the original order, not against memory. Revision two should be compared to the order form and to revision one, not just judged on whether it “feels right” at a glance.
- Treat the second round as the real checkpoint. The first render almost always needs adjustment. The habit that actually prevents remakes is scrutinizing round two as hard as round one — that’s usually the version that gets rubber-stamped because it already looks close.
This is also where consolidating diamond sourcing and custom setting under one manufacturing relationship pays off. When the loose stone, the CAD build, the casting, and the setting all move through a single review checkpoint — one vendor, one shipment, one relationship — there’s one file to check instead of specs getting re-interpreted across separate vendors for stones and separate vendors for setting. Our wholesale hub is built around that same one-file, one-approval flow, and access is limited to verified trade partner accounts, which keeps the review process consistent order over order instead of resetting with every new vendor relationship.
If you’re setting up your first custom order and want the CAD review step built into the process from day one, contact us before the file goes to a first render — it’s easier to set expectations on what a revision round should look like before you’re already looking at one.
Frequently Asked Questions
How many CAD revision rounds should I expect before casting?
Most custom orders resolve in one to two revision rounds. The first render establishes the basic geometry and almost always needs adjustment to prong height, metal weight, or band profile. The second round is typically the one that gets approved — which is exactly why it deserves the same level of scrutiny as the first, not less.
What happens if I approve a CAD file without checking prong height or stone seating?
If the error makes it into the cast piece, it’s no longer a file correction — it’s a physical remake. Prongs that are too low or too high can sometimes be adjusted post-cast, but seating errors that affect how a stone sits usually mean pulling the stone, and in some cases restarting the cast entirely. The fix that would have taken two minutes on the file becomes a multi-day rebuild in metal.
Can changes still be made once casting has started?
No. Once the approved file becomes a wax or resin pattern and goes into the flask, there’s no editing the piece — only remaking it. Any change identified after this point means going back to a new CAD revision and running the piece through casting, setting, and finishing again from that point forward.
Who is responsible if the finished piece doesn’t match what I approved?
The piece is built to match the exact file version signed off at final approval — that’s what the sign-off step is for. If the finished piece deviates from the approved file, that’s a manufacturing remake to correct. If the approved file itself contained an error that was missed during review, that’s a review-process gap, not a build error — which is exactly why catching issues before sign-off matters more than catching them after.
Does the 4-6 day custom jewelry turnaround start from CAD sign-off?
Yes. The 4-6 day window for a finished custom piece runs from the point a CAD file is approved through casting, setting, and finishing. It does not apply to loose diamond delivery, and it resets if a piece has to go back through an additional CAD round or a remake after casting.
Should I review the CAD myself, or have someone else on my team check it?
Whoever is closest to the client’s original request and spec sheet should be the one signing off — not necessarily whoever is fastest to respond. A second set of eyes helps, especially on prong height and metal weight, since front-of-house staff reviewing on client excitement alone tend to focus on how the render looks rather than whether it matches the spec.
