Real-Time Manufacturing Dashboards: SAP vs MES vs Custom

Your month-end OEE report says 72%. Your plant manager knows the real number is closer to 61%, but proving it requires pulling data from three systems and two WhatsApp groups. By the time anyone acts, the quarter is over.

This is why you're evaluating a production dashboard for real-time manufacturing visibility. The question isn't whether you need one. The question is which approach fits a 200-2000 employee operation that already runs SAP, already has legacy equipment, and has already survived at least one failed MES implementation.

Who Faces This Decision

You're running production at a mid-market manufacturer. Maybe automotive Tier 2, maybe contract manufacturing, maybe medtech with ISO 13485 obligations. You have 15-40 machines across 2-3 shifts. SAP handles your ERP. Excel handles everything SAP doesn't.

Your operators log downtime on paper or in a shared spreadsheet. Your shift leads pass a notebook. Your production planning happens in SAP, but your floor reality lives in WhatsApp groups where someone posted a photo of a jammed conveyor at 2am.

You've been asked to "digitalize the shop floor" at least twice. One initiative died in pilot. Another produced a dashboard nobody uses because updating it takes longer than the old process.

Now leadership wants real-time OEE visibility. They've seen demos. They want what the demo showed. You need to figure out what actually works in your environment.

flowchart LR
    subgraph Current State
        A[SAP ERP] -->|"batch export, daily"| B[Excel Reports]
        C[Operator Paper Logs] -->|"manual entry, end of shift"| B
        D[WhatsApp Groups] -->|"photos, no structure"| E[Shift Lead's Memory]
        F[Legacy PLC/SCADA] -->|"no integration"| G[Standalone HMI]
    end
    
    B -->|"3 weeks later"| H[Monthly OEE Report]
    E -->|"verbal handoff"| I[Next Shift Guessing]
    
    style H fill:#f96,stroke:#333
    style I fill:#f96,stroke:#333

Option 1: Extend SAP with SAP MII or SAP Digital Manufacturing

SAP Manufacturing Integration and Intelligence (MII) or the newer SAP Digital Manufacturing Cloud can connect your ERP to shop floor data. If you're already paying SAP licensing, this seems logical.

What it's good at: Tight ERP integration. Your production orders, BOMs, and cost centers stay in sync. Compliance documentation flows into your existing audit structure. IT likes it because it's one vendor.

Where it breaks: SAP MII requires significant configuration for each machine type. If you have 30 machines from 8 different OEMs with different PLC generations, you're looking at 6-12 months of integration work before your first real dashboard. Licensing adds €50-150K annually for a mid-market operation. Your floor operators will need training on SAP interfaces, which historically hasn't gone well.

Who it's for: Companies with standardized equipment, strong internal SAP expertise, and budget for 12-18 month implementations. If you're already running SAP PP-PI with MES hooks, extending makes sense.

What it costs: €150-400K implementation, €50-150K annual licensing, 12-18 months to first production use.

Option 2: Packaged MES (MPDV, FORCAM, Aegis, or Similar)

A dedicated Manufacturing Execution System sits between your ERP and your machines. MPDV Hydra, FORCAM FORCE, and similar products are common in DACH manufacturing.

What it's good at: Purpose-built for production tracking. OEE calculation is native. Shift scheduling, quality data capture, and machine connectivity are core features. German MES vendors understand IATF 16949 and ISO 9001 requirements. They've seen your plant before.

Where it breaks: MES implementations are notoriously scope-creepy. What starts as "just OEE tracking" becomes a 2-year project covering scheduling, quality, maintenance, and workforce management. Your operators now have another system alongside SAP. Data gets stuck between systems. The MES vendor's connector to your specific Siemens S7-300 generation might not exist. Customization for your process flow adds months.

Who it's for: Companies ready to commit to a multi-year platform investment with dedicated project teams. If you have 50+ machines and complex routing, a full MES eventually pays off. If you have 20 machines and need visibility next quarter, it's overkill.

What it costs: €200-600K implementation, €30-80K annual licensing, 12-24 months to full deployment. Most mid-market companies get partial functionality in 6-9 months.

Option 3: Build Custom on Your Existing Stack

Your IT team or a local development partner builds exactly what you need. A Node-RED instance pulling from your PLCs, a PostgreSQL database, a Grafana dashboard. Or a Python application connecting to your SCADA historian.

What it's good at: You get exactly what you need, nothing more. Your operators see exactly the screens you design. No licensing fees. No vendor lock-in. Your IT team understands it because they built it.

Where it breaks: Your IT team has other priorities. The developer who built it leaves. Documentation is thin. When you need to add a new machine type, it's a development project. When leadership wants a new KPI, it's a development project. Compliance auditors ask about validation, and your custom application wasn't built with 21 CFR Part 11 or ISO audit trails in mind.

Who it's for: Companies with strong internal development capacity and simple, stable requirements. If you need OEE for 10 identical machines and that scope won't change for 3 years, custom works.

What it costs: €40-120K initial development, ongoing maintenance burden, 3-6 months to first version. Hidden cost: technical debt accumulates.

Option 4: Operations Platform Approach (TIMPIA's Model)

A configurable platform that connects your existing systems without replacing them. Pull machine data from SCADA, orders from SAP, quality notes from wherever they live today. Build the dashboards and workflows you need without starting from zero or committing to a 2-year MES rollout.

What it's good at: Fast deployment on your actual systems. Your SAP stays your ERP. Your SCADA stays your SCADA. The platform handles the integration layer and presents unified dashboards. Operators can capture data on tablets without learning SAP transactions. Changes don't require development projects.

Where it breaks: If you need the full depth of MES functionality (detailed finite scheduling, complex batch genealogy, full SPC), a purpose-built MES has more features. If your IT policy requires everything in SAP, this adds another system. If your equipment is completely offline (no PLC access), the platform can't magically create data that doesn't exist.

Who it's for: Mid-market manufacturers who need production dashboard real-time visibility in weeks, not years. Companies with heterogeneous equipment. Operations that want to start with OEE and shift handoff visibility, then expand.

What it costs: €30-80K implementation, €15-40K annual licensing, 4-8 weeks to first production dashboard.

flowchart TB
    subgraph Platform Approach
        A[SAP ERP] -->|"API, real-time"| P[Operations Platform]
        B[SCADA/Historians] -->|"OPC-UA or polling"| P
        C[Legacy PLCs] -->|"edge gateway"| P
        D[Operator Tablets] -->|"mobile capture"| P
    end
    
    P --> E[Real-Time OEE Dashboard]
    P --> F[Shift Handoff Log]
    P --> G[Quality Deviation Alerts]
    P --> H[SAP Order Updates]
    
    E --> I[Plant Manager View]
    F --> J[Shift Lead View]
    G --> K[Quality Manager View]
    
    style P fill:#9cf,stroke:#333

Decision Framework: Which Path Fits Your Plant

The right choice depends on three factors: your current systems, your timeline, and your scope.

Choose SAP extension if:

You have strong SAP PP expertise in-house
Your equipment is relatively standardized (same PLC family)
You have 12+ months before leadership expects results
Your IT governance requires single-vendor solutions

Choose packaged MES if:

You have 50+ machines with complex routing
You need deep scheduling integration, not just visibility
You're willing to commit to a 2-year platform transformation
You have dedicated project resources (not borrowed from operations)

Choose custom build if:

Your requirements are narrow and stable
You have development capacity that won't get pulled to other projects
Compliance validation requirements are minimal
You're comfortable owning technical debt

Choose a platform approach if:

You need a production dashboard with real-time manufacturing data in 4-8 weeks
Your equipment is heterogeneous (multiple OEMs, PLC generations)
You want to prove value on one line before committing plant-wide
Your team has been burned by 18-month implementations that never delivered

flowchart TD
    A[Need real-time OEE visibility] --> B{Timeline?}
    B -->|"12+ months OK"| C{SAP expertise in-house?}
    B -->|"Need it this quarter"| D{50+ machines?}
    
    C -->|"Yes"| E[SAP MII/DMC Extension]
    C -->|"No"| F{Ready for 2-year commitment?}
    
    F -->|"Yes"| G[Packaged MES]
    F -->|"No"| H[Operations Platform]
    
    D -->|"Yes, complex routing"| G
    D -->|"No, need speed"| I{Internal dev capacity?}
    
    I -->|"Strong and stable"| J[Custom Build]
    I -->|"Limited or variable"| H
    
    style E fill:#ffd,stroke:#333
    style G fill:#dfd,stroke:#333
    style J fill:#fdd,stroke:#333
    style H fill:#9cf,stroke:#333

The Real Calculation

Here's the math on a 25-machine operation running 3 shifts:

Current state (manual tracking):

Operators log downtime: 5 min/event × 8 events/shift × 3 shifts = 120 min/day
Shift handoff verbal: 15 min × 3 handoffs = 45 min/day
Weekly OEE compilation: 4 hours
Monthly reconciliation with SAP: 8 hours
Annual labor on tracking alone: ~€35-50K

Time to first value:

SAP extension: 12-18 months
Packaged MES: 6-12 months (partial), 18-24 months (full)
Custom build: 3-6 months
Platform approach: 4-8 weeks

If you're losing 2-3% OEE to invisible problems (and most plants are), the cost of waiting 12 months for visibility is significant. On a €20M annual production value, 2% is €400K.

What Actually Happens Next

Whatever you choose, start smaller than the vendor wants you to. One production line. One shift. Prove it works before expanding.

If you're leaning toward MES, ask for references from companies your size, not their largest customers. Ask how long until their operators could use it without support.

If you're considering custom, document everything as if the developer will leave tomorrow. Because statistically, they might.

If the platform approach sounds right, the test is simple: can you see real-time OEE on one line within 8 weeks using your actual data? If a vendor can't commit to that, their "agile implementation" isn't.

Not sure which path fits your operation? Book a 20-minute walkthrough and we'll map it based on your actual systems.