The area manager walks up to the operator and asks: "How are we doing on quality today?"
The operator looks at the screen. It shows 97.3%. He nods. "Pretty good, I think."
Two hours later, the area manager gets a call. There are 47 defective parts sitting at the end of the line. The operator who said "pretty good" produced most of them.
He was not lying. He genuinely did not know. The percentage on the screen meant nothing to him — he had no way to connect 97.3% to the 47 parts in the red bin two meters away.
This is not a people problem. It is a measurement problem.
And it is one of the most common — and most expensive — failures in manufacturing performance management.
The Root Cause: One Language for Three Different Jobs
Most manufacturing plants use the same set of indicators at every level of the organization. The operator sees percentages. The team leader sees percentages. The area manager sees percentages. The plant manager sees percentages.
The assumption is that standardization creates alignment. In practice, it creates confusion — because the same number means something completely different depending on where you stand in the organization.
A 2.3% scrap rate means nothing to an operator who produces 200 parts per shift. He cannot visualize 2.3%. He cannot act on 2.3%. He cannot feel pride or urgency about 2.3%.
But he can feel pride about 4 defective parts in a shift instead of 11. He can feel urgency when the red bin is already half full at 10 AM. He can act on a number that is in the same language as his work.
The solution is not more data. It is the right data, in the right language, for the right level.
Three Levels, Three Languages
Every manufacturing organization has three distinct levels of performance visibility. Each level needs a different type of indicator — not because of hierarchy, but because of what each level can actually influence.
PI — Process Indicators
Level: Operator, Team Leader Language: Units, kilograms, meters, pieces, minutes Question answered: Is the process running as it should right now?
Process Indicators are the most granular level of measurement. They track what is happening in the process at the point where work is done.
A PI does not need to be calculated. It is counted, measured, or observed directly.
Examples from automotive manufacturing:
- Scrap per shift: 0.5 kg(not 1.2%)
- Actual Cycle Time: 48 second (not 96% efficiency)
- Parts produced last hour: 187 (not 93.5% of target)
- Unplanned downtime: 23 minutes (not 4.8% downtime)
The operator who knows his target is 0.5 kg of scrap per shift can look at the red bin at any moment and know where he stands. No calculation. No percentage. No interpretation.
That is the standard for a good PI: any operator should be able to tell you whether the process is on track or off track by looking at it directly — without a screen, without a formula, without a manager.
KPI — Key Performance Indicators
Level: Area Manager, Shift Manager Language: Percentages, ratios, rates Question answered: How is the area performing against the standard over time?
Key Performance Indicators aggregate multiple Process Indicators into a single view of area performance. They are useful for comparison, trending, and identifying patterns across shifts, lines, or time periods.
A KPI requires calculation. It is derived from raw data, not observed directly.
Examples:
- OEE: 78.4% (aggregates Availability × Performance × Quality)
- Scrap rate: 2.3% (total defective parts / total parts produced)
- Schedule adherence: 94.7% (orders completed on time / total orders)
- First Pass Yield: 96.1% (parts passing inspection first time)
The area manager who sees a scrap rate of 2.3% can compare it to the 1.8% target, identify which lines are contributing most to the gap, and make decisions about where to focus improvement efforts.
He is not counting parts. He is managing performance across multiple processes simultaneously. Percentages are the right language for that job.
KRI — Key Result Indicators
Level: Plant Manager, Operations Director, Finance Language: Euros, financial impact, cost per unit Question answered: What is the business result of our operational performance?
Key Result Indicators translate operational performance into financial outcomes. They answer the question that every plant manager and every board ultimately asks: what did this cost us, and what did we gain?
A KRI is the financial expression of what happened in the plant.
Examples:
- Cost of poor quality: €47,200 this month (vs. €38,000 target)
- Takt Profit: +€3,400 above target (shift level) or -€1,800 below target
- OEE financial impact: €23,600 lost to unplanned downtime this week
- Scrap cost: €8,900 in material waste this month
The plant manager who sees -€1,800 in Takt Profit for the night shift does not need to understand OEE to know there is a problem. He needs to know where to look — which is why the KRI must be connected to the KPIs and PIs that explain it.
The Cascade: How KRI, KPI, and PI Connect
The power of this framework is not in the three levels individually — it is in how they connect.
A single KRI can be influenced by multiple KPIs. Each KPI is driven by multiple PIs. When you know this cascade, you know exactly where to act to change a financial result.
Example cascade:
KRI: Cost of poor quality this month — €47,200 (target: €38,000) ↓ KPI: Scrap rate Line 2 — 4.1% (target: 2.0%) KPI: Rework rate Line 2 — 3.2% (target: 1.5%) ↓ PI: Defective crimps per shift — 23 pieces (target: 4) PI: Cable cut length deviation — 3mm average (target: ±0.5mm) PI: Tool wear cycle — 847 cycles since last change (standard: 600)
The plant manager sees a KRI that is €9,200 over target. He does not count parts. He asks which KPI is driving the gap. The area manager sees the scrap rate on Line 2 is double the target. He does not check tool wear cycles. He asks which PI is driving the scrap. The team leader sees the cable cut length deviation and the overdue tool change. He acts.
Each level sees the indicator in their language. Each level acts on what they can influence. The information flows in both directions — problems escalate up through the KRI, actions flow down through the PI.
This is not a reporting structure. It is a decision structure.
Why Most Plants Get This Wrong
The most common failure is using KPIs at every level — including at the operator level where PIs are needed.
The second most common failure is using KRIs only at board level — disconnected from the operational indicators that explain them.
Both failures produce the same outcome: people at every level looking at numbers they cannot act on, making decisions based on incomplete information, and eventually losing trust in the measurement system entirely.
The test for any indicator:
Before putting a number on a board or a screen, ask one question: can the person looking at this number directly influence it with their own actions?
If the answer is no — it is the wrong indicator for that level.
The operator cannot influence the monthly cost of poor quality. He can influence the number of defective parts he produces in the next hour. Put the right number in front of him.
A Practical Framework for Your Plant
Use this matrix to audit your current indicators. For each level, ask: are we measuring in the right language?
If your operators are looking at percentages — your indicators are misaligned. If your plant manager is counting individual parts — your indicators are misaligned. If nobody can connect a financial result to a specific process action — your cascade is broken.
The Connection to True North
The True North for a plant is always expressed as a KRI — a financial or high-level operational outcome that matters to the business.
"Reduce cost of poor quality from €47,200 to €30,000 by December 31st."
But the daily actions that achieve that True North happen at the PI level — in kilograms, pieces, and minutes, on the shop floor, in every shift.
The KRI/KPI/PI framework is what connects the strategic goal to the daily work. Without it, the True North is a number on a board. With it, every operator in the plant knows — in their own language — whether their actions today are moving the plant toward its goal or away from it.
That connection — from the plant manager's euro to the operator's kilogram — is what turns a measurement system into an improvement system.
Where to Start
If you are building this framework from scratch, start at the PI level.
Walk the shop floor. Ask operators: what do you measure today? What tells you whether your process is running well or not? What number would make you stop and call your team leader?
The answers to those questions are your PIs. Build the KPIs that aggregate them. Build the KRIs that translate them into financial impact.
Then connect the cascade — so that every financial result can be traced back to a specific process action, and every process action can be connected to a financial consequence.
That connection is what makes performance management real — instead of a reporting exercise that everyone participates in and nobody acts on.
At AdaptiveOps, the KRI/KPI/PI framework is the foundation of how we design performance management systems in manufacturing plants. If you want to discuss how this applies to your organization, book a free 30-minute diagnostic call at /contact.
Related: Why Having 7 Priorities Means Having None — how the True North is expressed as a KRI and cascaded to every level of the organization.
Related: How to Build a Daily Management System That Actually Works — how KRI, KPI, and PI connect to the daily management routine.