Case Study Overview: The Problem Wasn’t “Fit” — It Was Unseen Variation

A mid-volume crown-and-bridge lab (about 90–120 units/day across zirconia and hybrid ceramics) started getting a pattern of complaints that sounded familiar: “Needs more adjustment than usual,” “contacts are tight on one side,” “seat is inconsistent.” Internally, they saw a parallel trend: finishing time per unit was creeping up, and remakes were rising—still “manageable,” but eating margin.

They already scanned models at intake and designed cleanly. Milling looked normal. Sintering schedules were unchanged. So the team did what most labs do: they worked harder. More manual adjustments. More QC eyeballing. More “try-in and grind.”

What changed everything was adding one step they had always considered optional: post-processing verification scans—scanning restorations after milling and after finishing to compare against the approved CAD and detect where variation was actually entering the process.

At Repair.Dental, we see this scenario repeatedly: the lab is not “making bad restorations.” The lab is losing money to unmeasured drift.

milled restoration verification on Repair Guru

What the Lab Implemented: Two Verification Scan Gates

The new workflow step

They introduced two scan checkpoints:

Gate A: Post-mill scan (immediately after milling, before sintering/finishing, where applicable)
Gate B: Post-processing scan (after sintering + stain/glaze/polish, ready for dispatch)

They didn’t scan every unit forever. They started with a sampling strategy focused on high-risk cases: multi-unit bridges, implant work, and any case type showing adjustment-heavy behaviour.

Why is this different from a “visual QC”

Visual QC is subjective. A post-processing scan is measurable. It answers questions like:

Did the restoration warp after sintering?
Did finishing remove material in a way that changed occlusion?
Are contacts drifting because of tool wear or CAM offsets?
Is there a recurring bias that points to machine maintenance?

Baseline Metrics Before Verification Scans

The lab pulled 30 days of internal numbers (time-tracked, not guessed). Their baseline looked like this:

Table: Before vs After (Example from the Case Study)

Metric	Before (30-day avg)	After (30-day avg)
Remake rate (all units)	4.8%	2.1%
Avg finishing time per unit	11.5 min	8.2 min
“Adjustment-heavy” clinic notes	Frequent	Occasional
Rush shipping due to rework	Weekly	Rare
Unplanned mill downtime	2–3 events/month	0–1 events/month

These are example figures from the case study. Your numbers may differ—but the pattern is what matters.

Gate A Findings: Post-Mill Scans Exposed Tool and Calibration Drift

What they expected

They thought most variation would appear after sintering and finishing.

What the scans showed

Post-mill scans revealed a consistent pattern:

Contacts were trending 0.05–0.12 mm heavier than the CAD intent on one quadrant more often than random chance would suggest.
Internal surfaces showed intermittent “step” artefacts that were subtle visually but obvious on deviation maps.
Bridge connectors were occasionally under-milled in a repeatable zone.

Interpretation

That pattern pointed away from design and toward manufacturing:

Bur wear or spindle runout producing biased removal
CAM strategy and offsets not matching real tool condition
Potential axis calibration drift

Action taken

Instead of blaming staff technique, they did three targeted interventions:

Replaced the milling burs earlier than their previous schedule
Ran a mill calibration/verification routine and logged results
Introduced a “tool life by case type” rule (bridges consumed tool life faster than single units)

This is a core Repair.Dental principle: post-mill scans tell you when the mill is lying.

Gate B Findings: Post-Processing Scans Proved Where Finishing Was Quietly Changing Geometry

The surprise

The biggest deviation after finishing wasn’t internal fit—it was occlusal and contact drift caused by aggressive adjustment “to make it seat fast.”

In other words, technicians were unintentionally paying for manufacturing variation by removing material in finishing. The crown would seat, but occlusion became unpredictable. Dentists then adjusted chairside, and the lab lost reputation points.

What the post-processing scans made obvious

Stain/glaze and polish weren’t the main issue.
The main issue was manual contact correction happening too early and too broadly because upstream drift wasn’t visible.

Dentist-facing insight (why clinics noticed)

Clinicians felt it as:

“We’re adjusting more lately,”
“This is close but inconsistent,”
“Contacts are heavy in a weird way.”

Those comments often trace back to unmeasured small deviations that get multiplied across finishing and try-in.

How the Lab Used Deviation Maps Without Slowing Production

They avoided the trap of scanning everything

They built a simple sampling model:

Scan every multi-unit bridge post-processing (Gate B)
Scan 1 out of 10 single units (random)
Scan every remake candidate before remaking (to learn why it failed)

What they compared

Post-mill scan vs CAD
Post-processing scan vs CAD
Post-processing scan vs post-mill scan (this isolates finishing influence)

Table: Deviation Pattern → Likely Cause → Fix

What the scan shows	Likely cause	Best fix
Same contact zone heavy repeatedly	bur wear / spindle bias	change tool schedule, check spindle/runout
Internal surface “steps”	tool path or worn bur	replace bur, adjust CAM strategy
Warpage after sintering	sintering support/placement	tray/support review, schedule verification
Occlusal reduced after finishing	manual adjustment habit	update SOP, delay corrections until verified
One-axis consistent deviation	machine calibration	calibration verification, service if persistent

The “Costs Found” During the Case Study

The lab’s owner wanted a clear answer: what is this costing us per month?

They modelled cost in three buckets:

Lab labour cost (extra finishing minutes + rescans)
Remake cost (materials + machine time + labour)
Clinic friction cost (rush shipping, discounts, relationship strain)

Short cost illustration

If a lab is doing 2,000 units/month and “extra finishing” is 3 minutes above baseline on 40% of units:

2,000 × 40% = 800 units
800 × 3 minutes = 2,400 minutes = 40 labour hours/month
That’s before remakes.

Verification scans didn’t add 40 hours. They replaced chaotic hours with controlled minutes.

What Changed in Their SOP After the Data

1) Finishing rules became measurable

They introduced a simple rule:
If Gate A deviation exceeds threshold, do not “grind it into compliance” without logging the cause.

That forced the lab to fix upstream issues rather than masking them downstream.

2) Milling maintenance became predictive

Instead of “change tools when results look bad,” they used scan deviation trends as a trigger:

tool change
spindle check
axis verification
CAM offset review

3) QC became faster

Because scans highlighted where to look, QC shifted from long visual inspections to targeted checks.

Why This Matters to Repair.Dental: Verification Scans Reduce Guesswork in Repairs

When labs contact us about:

“fit issues,”
“milling inconsistencies,”
“sudden increase in adjustments,”

we ask for one thing that accelerates diagnosis: a deviation-based comparison (post-mill and/or post-processing vs CAD).

It lets us separate:

true scanner issues
milling machine mechanical drift
CAM offset misconfiguration
finishing variability
material/sintering behaviour

That reduces downtime because the lab stops chasing random fixes.

The Outcome: Better Fit, Lower Rework, Cleaner Client Conversations

After implementing post-processing verification scans with a sensible sampling strategy, the lab achieved:

lower remake rate
less finishing labour
fewer “mystery adjustment” patterns
more confident escalation when equipment maintenance was actually needed

Dentists didn’t receive a lecture about deviation maps. They simply experienced fewer surprises—which is what keeps B2B lab relationships sticky.

A Practical QC Checklist They Used for Post-Processing Scans

Quick pass/fail checks

Margins: continuous, no auto-filled gaps
Contacts: within threshold and symmetric to design intent
Occlusion: no broad reductions from finishing
Internal fit zones: no systematic bias
Repeatability: if rescanned, does it align the same way?

The key wasn’t perfection. It was detecting repeatable deviation patterns that indicate process drift.

Medit T310, T510, or T710 scanners often suffer from connection failures, distorted scans, or calibration drift over time. Restore full accuracy and speed with certified repairs using genuine-grade parts. Visit our Medit lab scanner repair page for free estimates, prepaid shipping, and 5–7 business day service.

Please Visit Repair.Dental or Follow us on Facebook & Instagram.

Milled Restoration Verification: Post-Processing Scans