Operations
Your process is either capable or it isn't. Measurement error is hiding which.
Process capability analysis tells you whether your operation can actually meet customer specs—or whether measurement noise is masking the real problem. Learn to separate the two and make the redesign-versus-improve decision with certainty.
Process capability analysis quantifies whether your operation can consistently produce output within customer tolerance limits by comparing actual process variation to the allowable range. Before you invest in controls or improvements, you need a statistical assessment that also validates your measurement system—because a measurement error can make a capable process look broken, or mask a process that cannot perform. The analysis reveals whether the problem is fundamental (redesign needed) or marginal (incremental improvement can work), enabling you to direct resources accordingly.
What gets in the way
Organizations that establish clear capability baselines before launching improvement efforts avoid a common trap: pouring resources into processes that lack inherent capability to meet specifications. The gap between leading practitioners and the middle tier is not sophistication—it is sequence. Top performers validate their measurement systems first, establish capability baselines second, and only then invest in control or improvement work. Middle-tier operations often reverse this sequence, investing in controls on data they have not validated and targeting improvements to processes they have not assessed for fundamental capability.
This reversal creates two failure modes. In the first, an organization improves a process that cannot deliver compliant output without redesign, generating diminishing returns and exhausted improvement budgets. In the second, they attribute real problems to measurement error and delay action while the process continues to generate defects. Leaders in this discipline move with speed not because they improve faster, but because they diagnose correctly the first time. They know, before improvement planning begins, whether the constraint is the process itself or the data stream feeding their decisions.
How leaders approach it
Process Capability Analysis: Establish Your Baseline
Process capability analysis measures the relationship between what your process actually produces and what your customer requires. It takes the observed variation in your output—the range of acceptable and out-of-spec results—and compares it mathematically to the tolerance limits your customer has specified. If your process variation is small relative to that tolerance window, the process is capable. If variation consumes most or all of the allowable range, it is not. This is a binary diagnostic, not a ranking. A process either can produce compliant output consistently, or it cannot.
The mechanism works because it separates common-cause variation—the noise inherent to any process—from special causes that drive individual defects. Before you run a capability analysis, the process must be stable: no unusual spikes, no unexplained shifts, no signals of external disruption. Once stability is confirmed, statistical indices (primarily Cpk for short-term capability and Pp for overall performance) quantify exactly how much margin exists between what the process delivers and what the customer accepts. A Cpk of 1.33 or higher is the working threshold; below that, even a stable process will produce out-of-spec output at an unacceptable rate.
What changes when you adopt this practice is the basis for resource allocation. Instead of debating whether a process 'needs' improvement, you have a statistical answer to whether it is capable of meeting specs without redesign. Organizations that run capability analysis before launching improvement initiatives find that 30-50% of their struggling processes lack sufficient inherent capability—meaning incremental improvements will not solve the problem. That insight redirects investment toward processes where improvement can actually work, and toward the processes that need fundamental rethinking before any control effort will succeed.
Leading Practice Report
Full detail: Process Capability Analysis
The full report covers:
- Expected benefits
- Core principles
- Key success factors
- Key metrics
- Risks and mitigations
- Implementation roadmap
Measurement System Analysis: Validate Before You Decide
A measurement system is only as good as its ability to detect true differences in process output. If your gauges drift, your procedures are inconsistent, or your operators are not trained identically, you are not measuring the process—you are measuring the noise in your measurement method. Measurement System Analysis, commonly called Gauge R&R, validates that the tools, procedures, and people involved in taking measurements produce repeatable and reproducible results. Without this foundation, capability analysis becomes guesswork, and improvement decisions rest on unreliable data.
The analysis works by having multiple operators measure the same parts multiple times using the same equipment and procedure. You then calculate what portion of your observed process variation is actually coming from the measurement method itself versus genuine process performance. If measurement variation is high relative to process variation or tolerance limits, your data cannot be trusted—a process may appear incapable when it is actually performing well, or real problems may be hidden by measurement noise. Equipment calibration, operator training, and procedure consistency are non-negotiable elements; neglecting any one of them invalidates the entire measurement system.
Organizations that validate their measurement systems before launching improvement or capability work see dramatic shifts in decision quality. When measurement integrity is confirmed, executives can trust that observed variation reflects true process behavior, not instrument drift or procedural inconsistency. The result is faster problem identification, fewer false starts on improvement initiatives targeting phantom issues, and reduced rework from defects that were masked by poor measurement. Measurement system improvement becomes an ongoing discipline, not a one-time audit.
Leading Practice Report
Full detail: Measurement System Analysis (MSA) / Gauge R&R
Benefits, core principles, success factors, metrics, risks and the implementation roadmap.
Get the full report →First steps
- Audit your current measurement system: identify every gauge, procedure, and operator involved in assessing whether output meets spec. Document calibration schedules and training records. Most organizations discover gaps immediately.
- Run Gauge R&R on your highest-value or highest-variability process. This takes weeks, not months, and reveals whether your measurement data is trustworthy before you invest in capability analysis.
- With validated measurement data in hand, conduct a formal process capability analysis on your three to five most critical processes. Establish which are inherently capable and which require redesign before improvement efforts begin.
Ask us how to sequence measurement validation and capability analysis in your operation, and what your roadmap looks like.
Start free with Ask Kepler →Emerging approaches
Advanced & Emerging Practices
Emerging practices are included with Ask Kepler Pro and Max.
Unlock these practices →Curious how Ask Kepler Research is produced? Read our research methodology.
Performance Playbook
Controls Design & Implementation: How fast should controls move from design to full operation?
Learn how to compress your deployment cycle from months to weeks while catching design defects before controls go live. This playbook shows you how to separate design validation from testing, measure what matters, and bu
Open the playbook →