You installed machine monitoring. You track cycle times. You log setup hours, material usage, and scrap rates. But when was the last time that data changed how you quote a job?
Most digital manufacturers collect shop floor data diligently. Few connect it back to the estimating process in a way that sticks. The result: quotes still rely on tribal knowledge, spreadsheets, or “what we charged last time” — while the floor runs on a different reality.
The data you already have (but aren’t using)
Every job that passes through your shop generates a trail:
- Actual machine hours vs. estimated
- Setup and changeover time by process
- Scrap and rework rates per material, geometry, or machine
- Tool wear and consumable consumption
- Operator idle time vs. machine running time
- Post-processing hours (deburring, heat treat, inspection, finishing)
This isn’t theoretical. It’s recorded every day — often in your MES, ERP, or even paper travelers. The problem isn’t collection. It’s the feedback loop.
What the numbers reveal when you actually look
1. Your setup estimates are consistently low
Shops routinely underestimate setup by 30–50%. The data shows it: first-article inspection, fixture changes, program prove-outs, and material staging all add up. If your quoting template uses a flat setup rate, you’re leaving money on every job.
2. Scrap isn’t random — it clusters
Scrap rates spike on specific geometries, materials, or machines. Thin-wall titanium on a 5-axis? High scrap. Nested sheet metal with tight tolerances? Yield drops. When you map scrap to part features, you can price risk accurately instead of padding every quote “just in case.”
3. Post-processing is where margins evaporate
Deburring, tumbling, anodizing, CMM inspection — these are often quoted as afterthoughts. Floor data shows they can consume 20–40% of total labor hours. If you’re not capturing them per operation, you’re subsidizing the customer’s finish requirements.
4. Machine hour rates drift from reality
Your burden rate assumes 75% utilization. Actual utilization is 42%. That gap means every machine hour quoted at your standard rate under-recovers overhead. Floor data lets you calculate true machine hour rates by process, not by accounting averages.
5. Lead time promises don’t match capacity
Quoting says “2 weeks.” The floor sees 3 weeks of WIP ahead of the job. Real-time capacity data — not gut feel — should drive the lead time on the quote. Otherwise you win the order and lose the customer when you miss the date.
Closing the loop: from data to quoting intelligence
The goal isn’t more dashboards. It’s a quoting engine that learns from every completed job.
Feed actuals back into templates
When a job finishes, its real cycle times, setup hours, scrap yield, and post-processing hours should update the quoting rules for that process-material-machine combination. Next time that geometry appears, the quote reflects reality — not a guess from three years ago.
Tag jobs by complexity drivers
Don’t just track “CNC milling.” Track “5-axis, Inconel 718, <3mm walls, ±0.025mm tolerance." Granular tags let you build pricing models that reflect the actual cost drivers, not broad categories.
Automate the feedback, don’t rely on memory
Estimators shouldn’t need to hunt through job travelers. The MES should push actuals into the quoting system automatically — or at minimum, surface a side-by-side comparison at quote time: “Last 5 similar jobs: avg 4.2 hrs machine, 1.8 hrs setup, 8% scrap.”
Use capacity data to price urgency
Real-time machine availability lets you price rush orders correctly. If the 5-axis is booked solid for 10 days, a 3-day lead time isn’t just “expedited” — it’s displacement cost. Your quote should reflect that.
What changes when you get this right
- Quote accuracy improves — fewer surprises at job close
- Margins stabilize because pricing reflects true cost
- Estimators stop reinventing the wheel on every RFQ
- Sales can defend prices with data, not hope
- Capacity planning shifts from reactive to predictive
The shops that win consistently aren’t the ones with the most data. They’re the ones where the floor talks to the front office — automatically, every day.
Start with one feedback loop
You don’t need a full integration project. Pick one process. Compare the last 10 quoted vs. actual hours. Find the delta. Update the template. Repeat.
Solvi connects quoting, MES, and capacity data so the loop closes itself — actuals flow back into the quote engine, capacity drives lead times, and every job makes the next quote smarter. See how it works.