← Back to Field Notes

July 21, 2025 · The Bloomfield Team

12 Manufacturing KPIs That Actually Matter

Most manufacturing dashboards display 30 to 50 metrics. The production team checks three of them. The rest exist because someone asked for them during a software implementation two years ago and nobody has removed them since. A metric that does not change a decision is decoration.

These 12 KPIs are the ones that drive action in well-run manufacturing operations. Each one connects to a specific decision, is measurable with data most shops already collect, and has a benchmark that tells you whether your number is competitive. For a broader view of how to build a dashboard your team will use, see our guide to production visibility in manufacturing.

Production

Percentage of orders shipped on or before the promised date. The single most important customer-facing metric. Shops below 90% are losing customers whether they know it or not. Shops above 95% are building preferred vendor status that compounds over years.

Availability multiplied by performance multiplied by quality. Most job shops run between 45% and 65% OEE. World-class operations hit 85%. The gap between your number and the benchmark represents recoverable capacity from your existing equipment.

Percentage of parts that pass inspection on the first run without rework or scrap. Every point below 98% has a dollar value attached to it: material cost, labor cost, and the opportunity cost of the machine time consumed by rework.

Percentage of planned production that actually runs as scheduled. Low schedule adherence means frequent changes, which means wasted setups, disrupted flow, and unpredictable delivery dates. This metric tells you how much your plan reflects reality.

Financial

Revenue minus direct costs (material, labor, outside processing) divided by revenue, calculated per job. The aggregate number matters less than the distribution. If 20% of your jobs run below 20% margin, those jobs are candidates for repricing or declining on the next order.

The difference between quoted hours and actual hours on completed jobs. Consistent positive variance means you are underquoting. Consistent negative variance means you are leaving money on the table. Variance above 15% in either direction signals a quoting process that needs calibration.

Total revenue divided by headcount. This is a rough efficiency metric, but it reveals structural issues over time. A declining number with a stable team usually indicates underutilized capacity or a shift toward lower-margin work.

Quoting and Sales

Percentage of submitted quotes that convert to purchase orders. Below 20% usually means pricing is too high or turnaround is too slow. Above 40% often means pricing is too low. The target is a win rate that maximizes total margin, which is volume multiplied by margin per job.

Calendar days from RFQ receipt to quote submission. The correlation between turnaround speed and win rate is strong and consistent. Shops that respond within 48 hours win at materially higher rates than shops that take four or five days.

Quality and Knowledge

Percentage of shipped parts returned for quality issues. Internal scrap is a cost. Customer returns are a relationship and revenue risk. Track this by customer and by part family to identify patterns before they become account losses.

Total setup hours divided by total production hours. In high-mix job shops, setup time is a structural cost. The goal is to reduce it through better documentation, fixture standardization, and scheduling that groups similar setups. Every percentage point reduction adds capacity without adding machines.

Percentage of critical processes that depend on one person. This is the metric most shops never track, and it represents the largest unquantified risk in the operation. If one person leaving would shut down quoting, first-article inspection, or five-axis programming for more than a week, that process carries unacceptable concentration risk.

The Rule of Three

Twelve KPIs is the full set. In practice, the most effective manufacturing leaders focus on three at any given time: one they are actively trying to improve, one they are monitoring for regression, and one that serves as a leading indicator of the other two. The rest run in the background. Trying to improve all twelve simultaneously means improving none of them.

Pick the three that connect most directly to your current constraint. Build the visibility to track them weekly. Act on what the data shows. That is how KPIs become operational advantages instead of dashboard decorations.