Measurement System Analysis (MSA): Why Good Projects Fail Without Good Data

By Tania Noronha

In Lean Six Sigma, data drives every decision — but data is only as good as the system that produces it. Measurement System Analysis (MSA) evaluates the reliability and accuracy of measurement processes to ensure teams are not improving noise instead of the process itself.

Before advancing through the DMAIC Measure phase, MSA answers a critical question: Can we trust the data we are about to analyse?

🚨 HIRING: Tech Talent 💰 $50–$120/hr | 🔥 Multiple Roles

Frontend • Backend • Full Stack • Mobile • AI/ML • DevOps 👉 **Apply Here**

What Is Measurement System Analysis?

MSA is a structured set of statistical techniques used to quantify measurement error and separate it from true process variation. It assesses whether variation comes from the process — or from the way it is being measured.

Without a capable measurement system, even well-executed DMAIC projects risk false conclusions, wasted effort, and failed improvements.

Gage R&R Studies: Repeatability and Reproducibility

The most common form of MSA is the Gage Repeatability & Reproducibility (Gage R&R) study.

For variable data, Gage R&R evaluates:

• Repeatability: variation when the same operator measures the same part multiple times
• Reproducibility: variation between different operators measuring the same part

Results are typically expressed as %GRR, with best practice targets:

• < 10%: acceptable
• 10–30%: conditionally acceptable
• > 30%: unacceptable

A standard study design uses 10 parts, 3 operators, and 3 trials, analyzed via ANOVA or X̄-R methods.

For attribute data (pass/fail, visual inspection), attribute Gage R&R relies on kappa statistics to evaluate agreement between operators and against a known standard.

Bias, Linearity, and Stability

A complete MSA goes beyond Gage R&R by evaluating three additional characteristics:

• Bias measures systematic error by comparing measurements against a known reference value.
• Linearity assesses whether bias remains consistent across the entire measurement range.
• Stability examines whether the measurement system changes over time, typically monitored using control charts on reference standards.

All three must be acceptable before process data can be considered trustworthy. Ignoring these checks allows measurement error to amplify misleading trends and false root causes.

The Cost of a Poor Measurement System

Poor measurement systems hide real process variation and create artificial problems. Teams may end up:

• Chasing “defects” that don’t exist
• Rejecting good product
• Implementing unnecessary process changes

Many Lean Six Sigma projects fail not because solutions are wrong, but because measurement error dominates the data. This is why experienced practitioners emphasize qualifying the measurement system before performing capability analysis, hypothesis testing, or DOE.

As the saying goes: Good projects are often destroyed by bad data.

Final Thoughts

MSA is not a formality — it is a gatekeeper for credibility in Lean Six Sigma. Performing MSA early in the Measure phase ensures that decisions are based on reality, not measurement noise.

A capable measurement system turns data into insight. Without it, even the most advanced analytical tools will lead teams in the wrong direction.

Thank you for being a part of the community

Before you go:

👉 Be sure to clap and follow the writer ️👏️️

👉 Follow us: **Linkedin| [Medium](https://medium.com/codetodeploy)**

👉 CodeToDeploy Tech Community is live on Discord — **Join now!**

Disclosure: This post includes affiliate and partnership links.