The TRACE Protocol

TRACE is our systematic approach to diagnostic investigation. It provides structure without being prescriptive, ensuring thorough investigation while remaining adaptable to different problem types.

The Framework

T - Target the Symptom

Before you can solve a problem, you need to define it precisely.

Questions to answer:

• What exactly is happening?
• What should be happening instead?
• What is the gap between current and desired state?
• How do we measure the symptom?

Common mistakes:

• Defining the problem too broadly ("things are broken")
• Defining the symptom as a cause ("we have a training problem")
• Accepting someone else's problem definition without verification

Output: A clear, measurable symptom statement.

R - Record the Evidence

Systematic evidence collection prevents confirmation bias and ensures you don't miss important data.

Evidence types:

• Documentary: Emails, reports, logs, metrics, procedures
• Testimonial: Interviews, observations, firsthand accounts
• Physical: Site observations, artifact inspection
• Temporal: Timelines, sequences, patterns over time

Recording principles:

• Capture everything, filter later
• Note sources and dates
• Distinguish fact from interpretation
• Preserve original context

Output: Organized evidence repository with clear sourcing.

A - Analyze the Chain

Map the causal chain from symptom to source.

Process:

1. Start with the symptom
2. Ask "what directly caused this?"
3. Document the cause
4. Repeat until you reach a point where intervention is possible

Tools:

• Five Whys
• Fishbone Diagrams
• Causal chain mapping
• Timeline analysis

Warning signs you've stopped too early:

• The cause is a person ("John made an error")
• The cause is vague ("communication breakdown")
• You can't explain why this cause occurred

Output: Visual causal chain from symptom to root cause.

C - Challenge Assumptions

Every investigation is built on assumptions. Some are valid. Some aren't.

Questions to ask:

• What are we assuming to be true without evidence?
• What if the opposite were true?
• Whose perspective are we missing?
• What evidence would contradict our current hypothesis?

Common hidden assumptions:

• "The process is being followed as documented"
• "The people involved are telling us everything"
• "The metric we're measuring reflects the real problem"
• "Past performance predicts future behavior"

Output: List of validated and invalidated assumptions.

E - Expose the Root

The root cause is the deepest point in the causal chain where intervention is both possible and effective.

Root cause criteria:

• Removing it prevents recurrence
• It's within the organization's control
• Addressing it is proportional to the problem
• The evidence supports it

Validation tests:

• If this was the root cause, would this symptom have occurred? (Yes)
• If we eliminate this cause, will the symptom recur? (No)
• Can we trace the causal chain from this cause to the symptom? (Yes)
• Is there a deeper cause we haven't explored? (No)

Output: Documented root cause with supporting evidence.

Putting It Together

TRACE is iterative. You'll often cycle back through earlier phases as new information emerges:

1. Start: Target the symptom
2. Gather: Record initial evidence
3. Map: Begin analyzing the chain
4. Test: Challenge emerging hypotheses
5. Refine: Return to evidence gathering with new questions
6. Converge: Expose the root when evidence supports it

When to Use TRACE

• Complex problems with unclear causes
• Recurring problems that resist simple fixes
• High-stakes situations where being wrong is costly
• Situations where multiple stakeholders have different explanations

When TRACE is Overkill

• Simple problems with obvious causes
• Situations where speed matters more than certainty
• Problems where the solution is known and implementation is the challenge

TRACE provides structure, not a script. Adapt it to your context.