A classic example of common cause variation is an industrial oven whose temperature drifts slightly up and down during normal operation but stays within its acceptable range. This small, random fluctuation isn’t caused by anything broken or unusual. It’s simply how the equipment behaves under normal conditions. Common cause variation is the natural, built-in variability present in every process, and understanding it is essential for knowing when a system needs attention and when it’s working exactly as expected.
Everyday Examples Across Industries
Common cause variation shows up in virtually every process you can measure. Here are concrete examples across different settings:
- Manufacturing: An oven thermostat that lets the temperature drift slightly above and below the target, while remaining within control limits, is textbook common cause variation. Minor differences in the thickness of parts coming off the same machine, caused by normal tool wear and slight material inconsistencies, also qualify.
- Logistics: A delivery company found that its transit times varied by up to 30 minutes from one shipment to the next. Analysis revealed this was caused by routine factors like traffic patterns and differences in package size, not by any system failure.
- Customer service: Slight differences in how long each call takes at a call center. One call lasts four minutes, the next lasts six. No single cause explains it; it’s the natural rhythm of the work.
- Project management: Minor deviations in how long routine tasks take to complete. A weekly report that usually takes two hours might take 1.5 hours one week and 2.5 the next.
- Your commute: If you drive the same route every day, your travel time will never be identical. Some days you hit a few more red lights, other days traffic flows slightly better. That day-to-day wobble, typically within a predictable range, is common cause variation. A major accident blocking the highway for two hours would be special cause variation.
The thread connecting all these examples: the variation is random, small, and comes from many minor factors baked into the process itself. No single event or malfunction is responsible.
What Makes It “Common Cause”
Common cause variation has a few defining characteristics. First, it’s inherent to the system. It comes from the combined effect of many small, ordinary factors: ambient humidity, slight differences between raw materials, the natural pace of human work, minor equipment vibrations. No single factor dominates. Second, it’s random. If you plot the data points over time, they bounce around the average without a clear trend or pattern. Third, it’s predictable in the aggregate. While you can’t predict any individual data point exactly, you can predict the range the data will fall within.
Statisticians typically define this range as within three standard deviations of the average. As long as measurements stay inside that band, the process is considered stable and “in control.” W. Edwards Deming, the quality management pioneer who popularized this framework, estimated that roughly 94% of all process variation is common cause. He actually raised that estimate later in his career, emphasizing that most problems in a system come from the system’s design, not from individual incidents or errors.
How It Differs From Special Cause Variation
Special cause variation comes from something specific and identifiable: a machine malfunction, a new untrained employee, a power surge, a supplier shipping defective materials. It’s not random. It produces data points that either fall outside the normal range or form recognizable patterns like a steady upward trend.
Think of it this way. If your morning coffee takes between 3 and 5 minutes to brew every day, that range is common cause variation. If one morning it takes 15 minutes because the machine is clogged, that’s special cause variation. The distinction matters because the two types require completely different responses.
How to Spot It on a Control Chart
Control charts, developed by Walter Shewhart and widely used in quality management, are the standard tool for telling the two types of variation apart. A control chart plots data points over time with three reference lines: a center line (the average) and an upper and lower control limit, each set at three standard deviations from the center.
When data points bounce randomly within those limits with no obvious pattern, you’re looking at common cause variation. The process is stable. Several signals suggest something else is going on: a single point outside the control limits, eight or more consecutive points on the same side of the center line, two out of three successive points more than two standard deviations from the center on the same side, or four out of five points beyond one standard deviation on the same side. Any of these patterns flags potential special cause variation that warrants investigation.
Why the Distinction Matters for Fixing Problems
Reacting to common cause variation as if it were a special event is one of the most frequent management mistakes in quality control. If call times at your help desk naturally range from three to seven minutes, investigating every seven-minute call as a “problem” wastes resources and can actually make things worse by introducing unnecessary changes to a stable system. This overreaction is sometimes called “tampering.”
Reducing common cause variation requires changing the system itself. That might mean upgrading equipment, redesigning a workflow, improving training across the board, or selecting more consistent raw materials. These are structural, systemic changes. Special cause variation, by contrast, calls for identifying and removing the specific disruption: fixing the broken machine, retraining the one employee who’s struggling, or switching away from the supplier who sent bad parts.
Getting this distinction right is the foundation of statistical process control. A process running with only common cause variation is predictable, and predictable processes can be improved deliberately. A process plagued by unidentified special causes is unstable, and no systemic improvement effort will stick until those special causes are resolved first.