Building a Manufacturing Dashboard: What Metrics Actually Matter

The Dashboard Problem

Every manufacturing company wants a dashboard. The problem is that most dashboards end up as digital wallpaper — screens full of charts and numbers that look impressive but do not change anyone's behaviour. They show everything and highlight nothing.

We have built manufacturing dashboards for companies across India — automotive, pharma, food processing, textiles, and engineering. The ones that actually get used every day have one thing in common: they show a small number of metrics that directly connect to decisions people need to make. Everything else is noise.

The Metrics That Matter: A Practical Framework

Rather than listing 50 possible KPIs, let us focus on the metrics that manufacturing managers and business owners actually need, organised by the decisions they inform.

Production Efficiency

Overall Equipment Effectiveness (OEE) is the single most important production metric, and it should be the centrepiece of your dashboard. OEE combines three factors:

• Availability: What percentage of planned production time was the machine actually running? (Accounts for breakdowns and changeovers)
• Performance: When running, was it running at its designed speed? (Accounts for slow cycles and minor stops)
• Quality: Of the units produced, what percentage was good on the first pass? (Accounts for defects and rework)

OEE = Availability x Performance x Quality. World-class OEE is 85%. Most Indian manufacturers operate between 40-65%. The gap represents enormous untapped capacity — the equivalent of adding a second shift without buying a single new machine.

But OEE as a single number is not enough. You need to see the breakdown. A machine running at 70% OEE because of frequent changeovers needs a different solution than one running at 70% OEE because of a high rejection rate.

Quality Metrics

Beyond the quality component of OEE, your dashboard should track:

• First Pass Yield (FPY): Percentage of units that pass quality inspection without rework. This is a more actionable metric than overall rejection rate because it captures the cost of rework — units that eventually pass but consumed extra resources.
• Defect Pareto: The top 5 defect types by frequency. This should update in real time. If a new defect type suddenly appears in the top 5, something has changed on the line and needs immediate attention.
• Supplier quality: Incoming material rejection rate by supplier. If Supplier A's rejection rate has doubled this month, you need to know before it cascades into production problems.

Delivery Performance

Your customers care about one thing: did you deliver what you promised, when you promised it? Your dashboard should show:

• On-Time In-Full (OTIF): Percentage of orders delivered on the promised date with the full quantity. This is the metric your customers are tracking about you, whether you track it or not.
• Order backlog: Total value and quantity of orders pending fulfilment, broken down by due date. A healthy backlog is normal; a growing backlog is a warning sign.
• Production schedule adherence: Planned production versus actual production by product line. If you consistently produce less than planned, the root cause might be unrealistic planning, material shortages, or capacity constraints — each requiring a different response.

Cost Metrics

• Cost per unit: The actual cost of producing each unit, including materials, labour, overheads, and rework costs. Compare this against your standard cost to identify variance trends.
• Material yield: For process manufacturers, how much finished product you get from a given quantity of raw material. A 1% improvement in material yield can translate to crores in savings for high-volume manufacturers.
• Energy cost per unit: With Indian electricity costs rising, energy consumption per unit of production is becoming an increasingly important metric, especially for energy-intensive processes like heat treatment, melting, or drying.

Inventory Health

• Days of inventory on hand: Separately for raw materials, WIP, and finished goods. Too little means stockout risk; too much means capital locked up.
• Inventory accuracy: System stock versus physical stock, tracked through cycle counts. If this is below 95%, you cannot trust your MRP and your dashboard numbers are unreliable.
• Dead stock value: Material that has not moved in 90+ days. This number should be trending down. If it is not, your purchasing or planning process needs attention.

Dashboard Design Principles

Layer the Information

The most effective dashboards have three layers:

• Executive view: 4-6 headline numbers on a single screen. OEE, OTIF, order backlog, cash position. A managing director should be able to glance at this and know if the business is on track.
• Manager view: Drill-down into each headline number. Click on OEE and see the breakdown by machine. Click on OTIF and see which orders are late and why. This is where root cause analysis happens.
• Operator view: Real-time, shift-level data at each workstation. Current production count versus target. Rejection count. Machine status. This is the view that drives immediate action on the shop floor.

Show Trends, Not Just Snapshots

A number without context is useless. OEE at 62% — is that good or bad? It depends on whether it was 55% last month (good trend) or 70% last month (bad trend). Every key metric should show at least 12 weeks of trend data.

Use Colour Sparingly but Meaningfully

Red, amber, green — but only when thresholds are clearly defined and agreed upon. If everything is green, the thresholds are too loose. If everything is red, the thresholds are too tight. Aim for 60-70% green, 20-30% amber, and 5-10% red at any given time.

Make It Actionable

Every metric on the dashboard should answer the question: "If this number is outside its threshold, what do I do?" If there is no clear action associated with a metric, it does not belong on the dashboard.

Connecting Your Dashboard to Your ERP

A dashboard is only as good as its data source. If you are running SAP Business One, you already have most of the data you need — production orders, inventory transactions, sales orders, quality inspection results, and financial postings. The dashboard layer reads from this data in real time and presents it in the format described above.

For shop floor data that lives outside the ERP — machine cycle times, downtime reasons, energy consumption — you need a data collection layer (IoT sensors, PLC integration, or manual entry at terminals) that feeds into either SAP B1 directly or a separate data warehouse that the dashboard reads from.

Start Simple, Iterate Fast

The biggest mistake companies make with dashboards is trying to build the perfect one in a single project. Start with 5-6 metrics that you know matter, get them accurate and real-time, and put the dashboard in front of the people who need it. Within two weeks, you will learn what is missing and what is unnecessary. Iterate from there.

Our team at Indivar has built custom manufacturing dashboards integrated with SAP Business One for clients across multiple industries. If you are looking for visibility into your operations — real visibility, not just more charts — let us talk about what your dashboard should look like.