Iterative Improvement Frameworks

Process improvement isn't a one-time event—it's a cycle. Almost nothing worthwhile is accomplished on the first try. This chapter covers frameworks for structured, repeated improvement efforts.

The Nature of Iteration

Expertise looks effortless because practitioners have iterated thousands of times invisibly. The same applies to processes: good processes are the result of many cycles of trying, learning, and adjusting.

"Almost nothing worthwhile is accomplished on the first try."

Even understanding fundamental concepts like thermodynamics required over 150 years of iterative refinement. Your processes won't be perfected overnight either—and that's okay.

DMAIC: Define, Measure, Analyze, Improve, Control

DMAIC is the signature framework of Six Sigma, providing structure for solving existing problems.

The Five Phases

Phase Details

Define

Purpose: Clarify the problem and set direction

Key Activities:

  • Write problem statement
  • Define scope and boundaries
  • Identify stakeholders
  • Set measurable goals
  • Create project charter

Deliverable: Project charter with clear problem statement, goals, scope, and team

Measure

Purpose: Understand current performance

Key Activities:

  • Determine what to measure
  • Validate measurement system
  • Collect baseline data
  • Document current process
  • Calculate current performance level

Deliverable: Baseline metrics and validated measurement approach

Analyze

Purpose: Find root causes

Key Activities:

  • Brainstorm potential causes
  • Analyze data for patterns
  • Test hypotheses
  • Verify root causes
  • Prioritize causes to address

Tools:

  • Fishbone diagrams (cause and effect)
  • Pareto analysis (80/20 rule)
  • Five Whys
  • Statistical analysis
  • Process mapping

Deliverable: Verified root causes with supporting evidence

Improve

Purpose: Develop and implement solutions

Key Activities:

  • Generate solution ideas
  • Evaluate and select solutions
  • Pilot solutions
  • Measure results
  • Refine and implement

Deliverable: Implemented solution with measured improvement

Control

Purpose: Sustain the gains

Key Activities:

  • Document new process
  • Establish monitoring
  • Create response plans
  • Train personnel
  • Hand off to process owner

Deliverable: Control plan, documentation, and trained team

When to Use DMAIC

  • Existing process needs improvement
  • Problem is recurring
  • Root cause isn't obvious
  • Data is available or collectible
  • Structured approach needed

PDCA: Plan, Do, Check, Act

PDCA (also called the Deming Cycle or Shewhart Cycle) is simpler than DMAIC but equally powerful for continuous improvement.

The Four Steps

Using PDCA

Step Activities Questions to Ask
Plan Analyze current state, identify changes, predict results What are we trying to improve? What change might work? What do we predict will happen?
Do Implement change on small scale, collect data Did we implement as planned? What happened?
Check Compare results to predictions, analyze what worked Did we get expected results? What did we learn?
Act Standardize if successful, adjust if not, choose next cycle Should we adopt, adapt, or abandon? What's next?

PDCA Example

Situation: Call center handle time averaging 8 minutes; goal is 6 minutes.

Cycle 1:

  • Plan: Provide quick-reference guides for common issues
  • Do: Pilot with 5 agents for 2 weeks
  • Check: Handle time dropped to 7.2 minutes for pilot group
  • Act: Adopt guides, expand to all agents

Cycle 2:

  • Plan: Add screen shortcuts for frequent transactions
  • Do: Implement for all agents, track for 2 weeks
  • Check: Handle time now 6.5 minutes
  • Act: Standardize shortcuts, look for next improvement

Cycle 3: Continue toward goal...

PDCA vs. DMAIC

Aspect PDCA DMAIC
Complexity Simple More structured
Duration Can be very quick Typically longer
Data requirements Flexible More rigorous
Scale Small to large Typically larger
Best for Continuous improvement Specific problems

The OODA Loop

Originally developed for military decision-making, OODA (Observe-Orient-Decide-Act) is valuable in fast-moving situations.

The Four Elements

Element Purpose Activities
Observe Gather information Monitor environment, collect data, notice changes
Orient Make sense of observations Analyze, consider context, update mental models
Decide Choose action Evaluate options, select response
Act Execute Implement decision, then observe results

When to Use OODA

  • Rapidly changing situations
  • Competitive environments
  • Crisis response
  • When speed of decision matters
  • Adapting to unexpected events

The Five Whys

A simple but powerful technique for getting to root causes.

How It Works

When you encounter a problem, ask "Why?" five times (or until you reach a root cause).

Example

Problem: Customer received wrong product

  1. Why? The warehouse shipped the wrong item
  2. Why? The picker grabbed from the wrong bin
  3. Why? The bins weren't clearly labeled
  4. Why? Labels faded and weren't replaced
  5. Why? No maintenance schedule for bin labels

Root cause: Lack of maintenance process for warehouse labels

Solution: Implement regular label inspection and replacement schedule

Five Whys Tips

  • Don't stop at surface answers
  • Follow the chain of causation
  • Verify each "because" is factually true
  • May need more or fewer than five questions
  • Can branch when multiple causes exist

Agile Improvement Approaches

Agile methodologies emphasize frequent iteration and continuous adjustment.

Scrum

Scrum organizes work into fixed-length sprints (typically 2 weeks):

For process improvement:

  • Sprint planning: Select improvements to tackle
  • Daily standups: Track progress, remove blockers
  • Sprint review: Demonstrate improvements
  • Retrospective: What worked? What to adjust?

Kanban

Kanban visualizes work and limits work-in-progress:

| To Do | In Progress (limit: 3) | Done |
|-------|------------------------|------|
| Task A| Task D                 | Task G |
| Task B| Task E                 | Task H |
| Task C| Task F                 |        |

Key principles:

  • Visualize the workflow
  • Limit work in progress
  • Manage flow
  • Make policies explicit
  • Improve collaboratively

Benefits of Agile for Improvement

  • Faster feedback loops
  • Smaller, safer changes
  • Regular reflection built in
  • Flexibility to adjust direction
  • Visible progress

Choosing the Right Framework

Decision Guide

If you need... Consider...
Rigorous problem-solving DMAIC
Quick, simple cycles PDCA
Rapid response OODA
Root cause analysis Five Whys
Ongoing team improvement Scrum/Kanban
Complex problem with data DMAIC
Simple problem, fast fix PDCA

Combining Frameworks

Frameworks can work together:

  • Use Five Whys within DMAIC's Analyze phase
  • Apply PDCA for small improvements between major DMAIC projects
  • Use OODA for daily decisions while running Scrum sprints
  • Apply Kanban to manage improvement project workflow

Common Iteration Mistakes

Skipping the Learning Step

It's tempting to act, then act again, then act again—without pausing to check what's working. This leads to change without improvement.

Making Changes Too Big

Large changes are harder to evaluate. When something doesn't work, you don't know which part failed. Smaller iterations provide clearer feedback.

Giving Up Too Soon

Real improvement takes multiple cycles. If the first attempt doesn't work, that's information, not failure.

Not Standardizing Successes

When something works, document it. Make it the new baseline. Otherwise, improvements slip away.

Key Takeaways

  • Improvement is iterative—expect multiple cycles
  • DMAIC provides rigorous structure for significant problems
  • PDCA enables quick, continuous improvement
  • OODA works for fast-moving situations
  • Five Whys digs to root causes
  • Agile methods build iteration into team workflow
  • Combine frameworks based on your needs
  • Always pause to learn before the next cycle