Almost every dental lab now says it is “compatible with all major intraoral scanners.” That phrase can mean anything from a one-click connection on your scanner to a web form where someone uploads a file and hopes for the best, and the difference only shows up when a case comes back wrong.
This is what scanner compatibility actually involves, how accurate a scan needs to be, and where digital cases tend to fail between the chair and the lab.
What “compatible” has to cover
A lab is only truly compatible with your scanner if three separate things are true at once.
The first is a connection: the lab appears as a destination inside your scanner’s own software, so a case leaves the chair with the prescription attached. The second is the file: the lab can open the format your scanner produces, at the resolution it was captured, with colour if colour matters. The third is the workflow: the lab’s design software and technicians know what to check in a scan from your system before they start designing.
Plenty of labs have the second without the first, and some have both without the third. A lab that can open an STL is not necessarily a lab that will notice your margin disappears under tissue on the distal of #30.
Direct connections and exported files
Each major scanner family has its own route to a lab.
iTero (Align Technology) uses a lab list maintained by Align. A practice usually adds a lab by calling iTero support and giving the lab’s ID number, then restarting the scanner. 3Shape TRIOS connects through 3Shape Communicate, where the practice searches for the lab’s account and sends a connection request. CEREC and Primescan users connect through Dentsply Sirona’s portal. Medit uses Medit Link, and Carestream uses Carestream Connect.
Once a lab is connected, sending a case takes a few clicks and the prescription travels with the scan. Without a connection, the practice exports a file and uploads or emails it separately. That works, but it adds a manual step where files get mislabelled, the wrong arch gets sent, or the prescription arrives a day after the scan. Most labs publish the connection steps for each scanner, and it is worth setting this up before the first case rather than during it.
Which file format to send
Intraoral scans are exported in three common formats, and they are not interchangeable.
STL (stereolithography) stores surface geometry only, as a mesh of triangles. It carries no colour or texture. It is the format every dental CAD program can read, which is why it remains the default.
PLY (Polygon File Format) and OBJ store the same geometry plus colour and texture. That colour information helps a technician tell tooth from gingiva, see where a margin meets soft tissue, and read stains and existing restorations. The cost is larger files, and some design software needs a plug-in to read them.
For a posterior crown with a clear supragingival margin, an STL is usually enough. For a subgingival margin, an anterior case where shade context matters, or any case where the technician needs to see tissue, send colour.
Check whether your scanner exports these formats freely. On some systems, open export is a licensed or subscription feature, and a practice can find itself able to send cases only to labs inside one ecosystem.
How accurate a scan needs to be
Scanner accuracy is described in two terms defined by ISO 5725-1. Trueness is how close the scan comes to the real shape of the mouth. Precision is how consistently repeated scans of the same mouth agree with each other. A scanner can be precise and still untrue, repeating the same error every time.
The target those numbers serve is the crown margin. The most widely cited figure comes from McLean and von Fraunhofer, who examined more than 1,000 restorations and reported in the British Dental Journal in 1971 that a marginal opening of up to 120 µm was clinically acceptable. Most fit studies still measure against that threshold.
For a single tooth or a short-span bridge of up to about four units, published studies generally find intraoral scans as accurate as conventional impressions. Across a full arch, error accumulates, because the scanner stitches hundreds of small images together and every join adds a little distortion. A 2025 in vitro study in the Journal of Dentistry (Buhl, Ender and Mehl) found full-arch trueness ranging from 30.5 µm for the most accurate scanner tested to 95.0 µm for the least. High-precision conventional impressions came in at 24.5 µm. Every scanner tested stayed under 100 µm and was judged clinically acceptable for full-arch work.
A worked example: the error budget
Treat the 120 µm margin threshold as a budget. Everything between the mouth and the seated crown spends some of it: the scan, the design, the milling, the furnace, and the seating.
Errors do not add up neatly, because some cancel and some do not reach the margin at all. But the comparison is still useful. A full-arch scan at 30.5 µm trueness uses about 25% of the budget (30.5 ÷ 120 = 0.254). A full-arch scan at 95.0 µm uses about 79% of it (95.0 ÷ 120 = 0.792).
That leaves roughly 89.5 µm for every downstream step in the first case, and 25.0 µm in the second. For a single crown, scanned as a quadrant, the difference rarely matters. For a long-span bridge or a full-arch implant case, it is the difference between a lab with room to work and a lab trying to make a good restoration fit a distorted record.
This is why a good lab asks what scanner you use and how the scan was taken before it designs a long-span case, not after.
What a compatible lab checks when a scan arrives
When a scan lands, an experienced digital technician looks at several things before any design work starts:
- Whether the margin is continuous all the way around, without gaps where tissue, blood or saliva hid it.
- Whether the bite scan aligns the arches correctly, ideally confirmed from both sides.
- Whether the adjacent teeth and opposing arch were captured fully enough to design contacts and occlusion.
- For implants, whether the scan body matches the implant system and library named on the prescription.
- Whether shade notes or photos came with an anterior case.
A lab that sends a scan back with a specific question is doing its job. A lab that never asks one is either receiving perfect scans or not looking.
Where this goes wrong
The lab is “compatible” but not connected on your scanner
The lab supports your scanner brand, but nobody added it to your scanner’s lab list. The case gets exported by hand, the prescription travels separately, and a day is lost matching them up. Worse, the case goes to the previous lab that is still the default destination.
An STL is sent when colour mattered
A subgingival margin on a colourless mesh is hard to read with confidence. The technician has to guess where tooth ends and tissue begins, and the crown is designed to the guess. Sending PLY or OBJ for those cases costs nothing but file size.
A full-arch scan is used for a long-span case without checking it
Stitching error is largest across the arch, which is exactly where a long-span bridge or full-arch implant framework needs accuracy most. A passive fit on a framework depends on the distance between abutments being right, and that is the measurement intraoral scans are weakest at. For these cases, a verification jig or a check model is cheap insurance.
The bite scan is misaligned
A buccal bite captured on one side only, or with the patient not fully closed, can tip the whole model. The crown is then designed to a bite that does not exist, and it comes back high or out of contact. Scanning the bite on both sides lets the lab confirm the alignment.
The wrong scan body or library is used on an implant case
The scan body tells the lab where the implant sits. If the scan body on the patient does not match the system or library named on the prescription, the design software places the implant connection in the wrong position. The restoration will not seat, and the error may not be obvious until try-in.
The export is down-sampled
Some scanners and export settings reduce mesh resolution to make files smaller. The geometry looks complete on screen, but fine detail at the margin has been smoothed away. If your scanner offers an export resolution setting, send the highest.
How to check a lab before the first case
A few questions sort a genuinely digital lab from one that merely accepts files. Ask which scanners connect to the lab directly and which require an export. Ask what file format the lab prefers for anterior and subgingival cases. Ask what the lab does when a margin is unclear: does it call, send screenshots, or design to its best guess? And ask how it handles full-arch and implant cases scanned intraorally.
The answers matter more than the list of logos on the website. If you are looking for a digital dental lab in New York City that takes cases from iTero, TRIOS, CEREC and Carestream scanners, those are the questions to ask first. Simple single crowns, such as full-contour zirconia from a clean quadrant scan, are the best place to test the relationship before sending anything complex.
Frequently asked questions
Do I need a lab that supports my specific scanner?
Almost any lab can open an exported STL, so in the narrowest sense, no. In practice, a lab connected directly to your scanner receives the prescription with the scan, sees colour data when you send it, and is familiar with what scans from your system look like. That saves time and prevents mix-ups.
What is the difference between STL and PLY files?
STL files contain shape only, with no colour. PLY files contain shape plus colour and texture. PLY files are larger but help a technician read margins near soft tissue and anterior shade context.
Are digital impressions as accurate as traditional ones?
For single crowns and short bridges, published studies generally find them comparable. Across a full arch, high-precision conventional impressions still record the arch slightly more accurately, although current scanners are within clinically acceptable limits.
Can I switch labs if my scanner is tied to one ecosystem?
Usually yes, through file export. Check whether your scanner’s open export is included or requires a separate licence, and ask the new lab whether it can connect directly to your system.
Why did my lab send my scan back?
Most often because part of the margin was not captured, the bite did not align, or an implant scan body did not match the prescription. A rescan costs the patient a short visit. A crown designed to an unclear scan can cost a remake.
This article is general educational information for dental professionals. It is not a substitute for clinical judgment or for the manufacturer’s instructions for any scanner, software or material.
