How a 520-Person Automotive Supplier Built Quality Management System Software

A Tier 2 automotive supplier in Baden-Württemberg was losing customers. Not because of product quality. Because their quality response process was too slow.

When a customer reported a dimensional deviation, it took 11 days on average to close the 8D report. The problem was not the investigation itself. The problem was finding the data, getting approvals, and connecting events across shifts.

This is how they built a manufacturing quality management system software platform that cut that response time to 48 hours.

The Situation: 520 Employees, 4 Production Lines, 7 Disconnected Systems

The company supplies precision-machined components to three major German OEMs. 520 employees across two shifts. Four CNC production lines. ISO 9001 certified, working toward IATF 16949 full compliance.

Their quality infrastructure before we started:

SAP ERP for production orders, material tracking, customer data
Excel spreadsheets for quality deviation logs (one per line, four total)
Paper NCR forms filled out by operators, scanned, emailed to Quality
SharePoint folder with 2,400+ unstructured quality documents
WhatsApp group where shift leads flagged urgent issues
Outlook for the 8D approval workflow (Quality Manager to Production to Engineering to Customer Quality)

The Quality Manager had been in the role for 8 years. She knew where everything was. The problem: she was the only one who knew.

The Problem: 11 Days to Close an 8D, and Nobody Knew Why

When a customer complaint arrived, here is what happened:

Customer Quality received the complaint via email
Customer Quality forwarded to Quality Manager
Quality Manager searched SharePoint for related inspection records
Quality Manager asked shift leads (via WhatsApp) which shift produced the batch
Shift lead checked paper batch records (if they could find them)
Quality Manager opened the Excel deviation log for that line
Quality Manager drafted the 8D in Word
8D sent to Production Manager for review (email, average 2 days in inbox)
8D sent to Engineering for technical review (email, average 1.5 days)
8D sent back to Quality Manager for customer formatting
8D sent to customer

Average elapsed time: 11 days. Customer expectation: 72 hours.

The Quality Manager told us: "I spend 60% of my time searching for information that exists somewhere in our systems. The actual root cause analysis takes maybe 4 hours. Finding the data takes 4 days."

The Operations Director added: "We lost a EUR 2.1 million annual contract because our quality response times were too slow. The customer switched to a competitor who could respond in 48 hours."

flowchart TD
    A[Customer Complaint Email] --> B[Customer Quality Inbox]
    B -->|Forward, ~4hrs| C[Quality Manager]
    C -->|Search, ~6hrs| D[SharePoint Documents]
    C -->|WhatsApp, ~2hrs| E[Shift Lead]
    E -->|Search, ~1day| F[Paper Batch Records]
    C -->|Manual lookup, ~2hrs| G[Excel Deviation Log]
    C -->|Draft, ~4hrs| H[8D Report in Word]
    H -->|Email, ~2days wait| I[Production Manager Review]
    I -->|Email, ~1.5days wait| J[Engineering Review]
    J -->|Email, ~1day wait| K[Quality Manager Formatting]
    K -->|Send, ~2hrs| L[Customer]
    
    style A fill:#ffcccc
    style H fill:#ffffcc
    style I fill:#ffcccc
    style J fill:#ffcccc
    style L fill:#ccffcc

The diagram above shows why 11 days was actually optimistic. If the Production Manager was traveling, or if the batch records were misfiled, the timeline stretched to 3 weeks.

What Made This Worse: The Excel Disconnect

Each production line had its own Excel file for tracking quality deviations. Four files, four formats, four different levels of detail.

Line 1's Excel had 47 columns including tool wear measurements. Line 3's Excel had 12 columns and no tool data. Line 2's operator had been entering dates in DD.MM.YYYY format while everyone else used MM/DD/YYYY.

When the Quality Manager needed to analyze trends across lines, she had to manually copy data into a fifth "master" Excel file. This happened monthly, took 6 hours, and was always out of date.

Root cause analysis across shifts was nearly impossible. The paper NCR forms referenced batch numbers, but connecting a batch number to the specific machine, operator, and process parameters required cross-referencing three different systems.

For more context on why Excel-based tracking creates these problems, see our comparison of production tracking approaches for IATF 16949 compliance.

What We Built: Manufacturing Quality Management System Software in 8 Weeks

We started with a 7-day proof of concept focused on one problem: can we automatically connect customer complaints to the relevant production data?

The POC pulled data from SAP (batch numbers, production orders), parsed the four Excel deviation logs into a unified schema, and created a search interface. The Quality Manager tested it with 10 historical complaints. Average time to find relevant data dropped from 6 hours to 12 minutes.

Based on that, we built the full platform in 8 weeks.

What the Quality Manager sees now:

When a customer complaint arrives (still via email, we did not change that), she forwards it to a dedicated inbox. The system parses the complaint, extracts the batch number, part number, and complaint category.

Within 30 seconds, she sees a unified view:

All production data for that batch (machine, operator, shift, process parameters from SAP)
All related quality deviations from the past 90 days (parsed from the Excel logs, now unified)
All related documents from SharePoint (inspection reports, process changes, customer specs)
Similar historical complaints and their 8D resolutions

What the approval workflow looks like now:

Instead of email chains, the 8D report lives in the platform. The Quality Manager drafts it with pre-populated data. She clicks "Request Review." Production Manager and Engineering Manager get a notification. They review in the platform, add comments, approve or request changes. The Quality Manager sees status in real-time.

The technical detail that made this work:

The four Excel files used 47 different column headers across 4 files. "Deviation Type" in one file was "Abweichungsart" in another and "Fehlertyp" in a third. We built a mapping layer that normalized 47 field variations into 12 standard fields. This parser also handled the date format inconsistencies (DD.MM.YYYY vs MM/DD/YYYY vs YYYY-MM-DD).

The SharePoint integration required OCR for scanned inspection reports. We trained a document classifier on 340 sample documents to categorize incoming files (inspection report, process change, customer spec, 8D report, other).

Timeline:

Week 1: POC (SAP connection, Excel parsing, search interface)
Weeks 2-3: Unified deviation database, SharePoint integration
Weeks 4-5: 8D workflow builder, approval routing
Weeks 6-7: Reporting dashboard, trend analysis
Week 8: User training, parallel run with old process

sequenceDiagram
    participant CQ as Customer Quality
    participant SYS as Quality Platform
    participant SAP as SAP ERP
    participant SP as SharePoint
    participant QM as Quality Manager
    participant PM as Production Manager
    participant ENG as Engineering
    participant CUST as Customer

    CQ->>SYS: Forward complaint email
    SYS->>SAP: Query batch/production data
    SAP-->>SYS: Production parameters (instant)
    SYS->>SP: Query related documents
    SP-->>SYS: Inspection reports, specs (instant)
    SYS->>SYS: Match historical deviations
    SYS-->>QM: Unified view ready (~30 sec)
    QM->>SYS: Draft 8D with pre-populated data
    QM->>SYS: Request review
    SYS->>PM: Notification
    SYS->>ENG: Notification
    PM->>SYS: Approve (avg 4 hrs)
    ENG->>SYS: Approve (avg 6 hrs)
    SYS-->>QM: All approvals complete
    QM->>SYS: Generate customer PDF
    SYS->>CUST: 8D Report delivered

The Results: 11 Days to 48 Hours

After 90 days of operation, the numbers:

8D Response Time

Before: 11 days average
After: 48 hours average
Calculation: Data gathering dropped from 4+ days to 30 minutes. Approval cycle dropped from 5+ days to 10 hours average. Formatting and delivery dropped from 1 day to 2 hours.

Quality Manager Time on Data Hunting

Before: 60% of working hours (approximately 24 hours/week)
After: 10% of working hours (approximately 4 hours/week)
Time freed: 20 hours/week, now spent on actual root cause analysis and supplier quality

Cross-Line Trend Analysis

Before: Monthly, 6 hours manual Excel consolidation, always 2-4 weeks out of date
After: Real-time, available in dashboard, updated as deviations are logged

8D First-Time Approval Rate

Before: 34% (most 8Ds bounced back for missing data or formatting)
After: 78% (pre-populated data reduces errors, standardized template)

Customer Quality Scorecard

Before: "Yellow" status with two OEMs due to response time
After: "Green" status with all three OEMs

Estimated Annual Savings:

Category	Calculation	Annual Value
Quality Manager time recovered	20 hrs/week × 48 weeks × EUR 45/hr	EUR 43,200
Reduced customer complaints escalation	12 escalations/year avoided × EUR 8,000 avg cost	EUR 96,000
Contract retention (conservative)	1 contract retained × EUR 2.1M × 5% margin contribution	EUR 105,000
Total		EUR 244,200

The platform cost EUR 67,000 to build (including the EUR 3,500 POC). ROI in the first year: 264%.

For comparison with other real-time manufacturing approaches, see our analysis of production dashboard options.

flowchart LR
    subgraph BEFORE["Before: 11 Days"]
        B1[4 Excel Files] --> B2[Manual Search]
        B3[Paper NCRs] --> B2
        B4[SharePoint Chaos] --> B2
        B2 --> B5[Email Approval Chain]
        B5 --> B6[Word Document]
    end
    
    subgraph AFTER["After: 48 Hours"]
        A1[Unified Platform] --> A2[Auto-populated 8D]
        A2 --> A3[Parallel Approval]
        A3 --> A4[PDF Generation]
    end
    
    BEFORE -.->|"Replaced by"| AFTER
    
    style B1 fill:#ffcccc
    style B3 fill:#ffcccc
    style B4 fill:#ffcccc
    style B5 fill:#ffcccc
    style A1 fill:#ccffcc
    style A3 fill:#ccffcc

What They Did Not Change

This is important: we did not replace SAP. We did not replace SharePoint. We did not ask operators to learn a new system for logging deviations (they still use Excel, but we pull from it automatically).

The manufacturing quality management system software sits on top of existing systems. It reads from SAP, parses Excel, indexes SharePoint, and provides a unified interface for quality workflows.

The Operations Director's requirement was clear: "Our operators have been using these tools for 10 years. I am not retraining 80 people on a new system. Whatever you build needs to work with what we have."

That constraint shaped everything. The platform is a coordination layer, not a replacement layer.

What Is Next for Them

Phase 2 (starting Q2) will add:

Supplier quality integration (connecting incoming inspection data to supplier scorecards)
Predictive deviation alerts (flagging process parameter drift before it becomes a quality event)
Direct 8D submission to customer portals (two of their OEMs have supplier portals that accept automated uploads)

They are also planning to consolidate the four Excel files into direct platform entry. But they are doing it gradually, one line at a time, with operator input on the interface design.

If Your Quality Process Looks Like the Before Picture

The manufacturing quality management system software we built for this Tier 2 supplier is not a product. It was built for their specific systems, their specific workflows, their specific Excel formats.

But the pattern is replicable. If you have:

Quality deviations tracked in multiple Excel files
8D reports that bounce between email inboxes for days
A Quality Manager who spends more time searching than analyzing
Customer scorecards turning yellow because of response times

We can show you what the "after" looks like in 7 days.

The POC costs EUR 3,500. It connects to your SAP, parses your Excel files, and shows you a unified view of your quality data. You keep the working prototype regardless of whether you continue.

Book a call to discuss your quality workflow.