Analyze Phase in DMAIC: Uncovering the Root Cause

When something goes wrong in a process—defects, delays, rework—the natural tendency is to jump to solutions. But in Lean Six Sigma, the Analyze phase of DMAIC reminds us to pause, dig deeper, and understand why the problem exists before attempting to fix it.

What Is the Analyze Phase?

The Analyze phase is the third step in the DMAIC (Define-Measure-Analyze-Improve-Control) framework. Its purpose is to:

• Identify the root causes of problems

• Validate them with data

• Prioritize which causes to address

In short, it’s where you turn data into insight.

Key Objectives of the Analyze Phase

1. Explore the data collected in the Measure phase

2. Identify patterns, trends, and deviations from the norm

3. Test hypotheses about possible causes

4. Validate the true root causes that significantly impact the problem

5. Build a clear cause-and-effect story

Tools and Techniques Used

Let’s break down the most common and effective tools used in the Analyze phase:

1. Process Mapping / Value Stream Mapping

Start with mapping the end-to-end process to see where delays, handoffs, and rework occur. This gives you a visual clue to potential problem areas.

📌 Example: In a hospital’s patient admission process, a value stream map may reveal that insurance verification is causing delays before treatment can begin.

2. Fishbone Diagram (Cause-and-Effect)

Also called the Ishikawa Diagram, this tool helps brainstorm possible causes under categories like:

• People

• Methods

• Machines

• Materials

• Environment

• Measurement

📌 Example: For a call center with increasing customer complaints, the team might use a Fishbone to explore whether the issue stems from agent training (People) or outdated scripts (Methods).

3. 5 Whys

An incredibly simple but powerful technique to drill down to the root cause by repeatedly asking “Why?” until you reach the systemic issue.

📌 Example:

• Why are claims being rejected? → Because of incomplete documentation.

• Why is documentation incomplete? → Because the form has confusing fields.

• Why is the form confusing? → Because it's outdated and hasn’t been reviewed.

By the 5th "why," you’re much closer to the root cause than the symptoms.

4. Pareto Chart

Use a Pareto Chart to apply the 80/20 rule—identify the “vital few” causes that contribute to the majority of the problem.

📌 Example: If 80% of delays in a delivery system come from just 3 out of 10 steps, focus your analysis and improvement efforts there.

5. Scatter Plot & Correlation Analysis

These statistical tools help uncover relationships between variables. For example, does longer training correlate with higher quality output?

📌 Example: In a loan processing operation, a scatter plot might reveal that more experienced processors have fewer errors.

6. Hypothesis Testing (Optional for Complex Projects)

If you're dealing with high-stakes processes or large data sets, use statistical tests like:

• t-tests (comparing two means)

• ANOVA (comparing multiple groups)

• Chi-square (categorical data)

• Regression analysis (predictive models)

These confirm whether the differences or patterns in your data are statistically significant, not just due to chance.

🚩 Warning Signs of Poor Analysis

• Jumping to conclusions without root cause validation

• Solving symptoms rather than the problem

• Data not linked to the problem definition or project Y

• Using too many tools without connecting insights

Always ask: Is this cause proven by data? Or is it a guess?

🧠 Real-World Example: Healthcare

A clinic experiences long wait times for walk-in patients. During the Analyze phase, the team uses:

• Time tracking data (from Measure)

• A value stream map showing steps in intake and triage

• 5 Whys revealing that nurses spend 15+ minutes finding patient files

The root cause? Poor filing system and inconsistent naming conventions.

No assumptions—just data-backed insights.

🛠️ Deliverables from the Analyze Phase

By the end of this phase, you should have:

✅ A refined process map✅ A validated list of root causes✅ A narrowed focus for the Improve phase✅ Data analysis summaries (charts, graphs, findings)✅ Team consensus on what to improve and why

✍️ Template: Root Cause Summary

Potential Cause	Tool Used	Validated?	Next Step
Outdated SOP	Fishbone Diagram	✅ Yes	Redesign SOP
Long walk-in times	VSM + 5 Whys	✅ Yes	Pilot solution
Printer issues	Pareto Chart	❌ No	Deprioritize

🧭 What's Next?

Now that you’ve identified and validated your root causes, you’re ready to move to the Improve phase—where creativity, experimentation, and breakthrough solutions come into play.

Remember: A strong Analyze phase makes the Improve phase easier and more effective.

📌 Final Thought

The Analyze phase is the most critical checkpoint in DMAIC. It helps you avoid solving the wrong problem, wasting time and resources. Be skeptical. Let the data speak. And use this phase to build the story behind your process problems.