MES and ERP integration: the real Siemens Amberg win

September 18, 2026 · 5 min · Free, with full transcript

Siemens Amberg is usually told as a robotics story, with 99.99885% quality and a heavily automated line. The position here is different: the recent gains came from making the Manufacturing Execution System and ERP agree on reality, not from more sensors. The route was treating manufacturing data as a product with owners, contracts and quality checks, rather than chasing one monolithic platform. You get the specific failure mode to look for, where MES counts a unit complete when it leaves the line and ERP counts it when booked to inventory, and what that gap does to supply forecasts. References include MIT Sloan Management Review, O'Reilly Radar and dbt Labs, plus a first project any CDO can start with a meeting.

Listen to the podcast

5 min

Chapters

Key takeaways

  • Agree on shared definitions of a unit, a batch and a defect across MES and ERP before funding any dashboard.
  • Treat manufacturing data as a product with named owners, data contracts and quality checks instead of building one monolithic platform.
  • Measure success as speed of production response, hours instead of weeks, rather than reporting cycle time.
  • Pull MES and ERP side by side and compare yesterday's unit counts; the discrepancy is your first project.
  • Copy the reconciliation logic, not the robot fleet, since the portable part costs a fraction of the automation spend.
Read the full transcript

Host:Welcome to the Leaders Insights Podcast. Today's episode, how Siemens built a unified manufacturing data backbone by integrating MES and ERP. Everyone loves the Siemens-Amberg plant story, 99.99885% quality, robots everywhere, the poster child for the smart factory. So tell me the part the case studies leave out.

Expert:The part they leave out is that the robots were never the hard bit. Amberg has been highly automated for over a decade. What actually moved the needle recently was boring plumbing, getting the MS and the ERP to stop lying to each other. MES is the Manufacturing Execution System, the software that runs the factory floor minute to minute, which machine makes what, when. ERP is enterprise resource planning, the business brain that handles orders, inventory, finance, classic setup, the floor knows one thing, the finance system believes another, and nobody reconciles until month end.

Host:And month end reconciliation is where careers go to die.

Expert:Right.

Host:So the consensus lesson people draw from Amberg is invest in more automation, more sensors.

Expert:Wrong lesson. Their real win was collapsing the gap between those two systems, so a production decision and a financial fact came from the same source.

Host:That's unglamorous. Nobody puts a data pipeline on the factory tour. But isn't Integrate Your Systems the advice every consultant has given since 2005? What's new?

Expert:What's new is they stopped trying to build one giant monolithic system, the fantasy where a single platform does everything. That project always dies. Instead they treated manufacturing data as a product, with owners, contracts, and quality checks feeding both systems. DBT Labs and Alflagit, they sell the data transformation tooling, so crosscheck against independent work, reports customers cutting data pipeline failures sharply after adopting that model. Take the vendor number with salt, but the direction matches what O'Reilly's engineering surveys show independently. Modular beats monolithic.

Host:You're dodging. Give me a number that isn't from someone selling software.

Expert:Fair. MIT Sloan Management Review has tracked this, wad. Firms that tie shop floor data directly to decision making, not just reporting, make production changes in hours instead of the weeks their peers take. That's the actual Amberg effect. Not fewer defects, faster reactions. When a supplier part shifts tolerance, the line adjusts before the bad units pile up, because the MES and ERP agree on reality in near real time.

Host:So what does everyone get wrong when they copy this?

Expert:They buy the dashboard and skip the discipline. A CDO, Chief Data Officer, the person accountable for data as an asset, walks in, commissions a gorgeous real time factory dashboard, and declares victory. Six months later, nobody trusts the numbers, because there's no data contract. No agreement on what a completed unit even means across systems.

Host:Define that failure precisely. Where does it actually break?

Expert:The MES counts a unit complete when it leaves the line. The ERP counts it complete when it's booked to inventory. 10 minute gap, harmless, until someone builds a supply forecast on it and orders parts for phantom output. Multiply across a plant, and you're managing fiction with beautiful graphics.

Host:That's the funny because it's true category. So the fix is definitions, not technology?

Expert:The fix is agreeing what a word means before you spend eight figures visualizing it. Students did the semantic work first. One definition of a unit, a batch, a defect, shared across both systems. Deeply boring. It's also why it worked when the flashier projects at other plants stalled.

Host:Push back on yourself. Isn't Amberg a special case? Massive budget, one product family, German engineering culture? Why should a mid-sized, discreet manufacturer believe any of this transfers?

Expert:It transfers precisely because the expensive part isn't the part you need. You don't need their robot fleet. You need their reconciliation logic, and that's cheap. A mid-sized plant with 200 people can define shared metrics and wire mess to ERP integration for a fraction of the automation spend. The insight is portable, even when the hardware isn't.

Host:Give me the one thing a CDO does Monday morning.

Expert:Pull your MEES and ERP side by side and count how many units each says you made yesterday. If the numbers differ, and they will, you've just found your first project, and it costs nothing but a meeting to start fixing it.

Host:Definitions before dashboards. That'll do it. Sources for today's episode. Towards Data Science, KD Nuggets, The New Stack, DBT Labs, Vendor, Data Tooling, O'Reilly Radar, MIT Sloan Management Review. Done. Want to know where you actually stand? Take the CDO self-assessment at MBA-training.com.

Read and go deeper