map-workflow

name	map-workflow
description	Extract the daily/weekly workflow, pipeline stages, time allocation, and healthy metrics from the domain expert.

/map-workflow - Workflow and Pipeline Mapping

When to Use

• After /discover-domain has captured initial domain understanding
• When domain-input/domain-workflow.md needs completion
• When the designer wants to detail the pipeline stages and daily cadence

Inputs

• Auto-loaded: domain-input/domain-workflow.md (partial, from /discover-domain)
• Auto-loaded: domain-input/identity-model.md (for user type context)

If domain-workflow.md is empty and /discover-domain hasn't been run, suggest running /discover-domain first. But don't block — the designer may want to fill this in directly.

Process

Step 1: Daily/Weekly Rhythm

Ask the designer to walk through the user's ideal operating rhythm:

1. What does the first 30 minutes look like? What should the user do first every day/week? What information do they need to see?
2. What are the 3-5 recurring actions? Things the user does every cycle (daily or weekly). For each: what triggers it, how long does it take, what's the output?
3. What's event-driven? Actions that happen in response to something (a new lead, a deadline, a request). List each event and the expected response.
4. What's the time budget? How much time per day/week should the user spend on this domain? What's the breakdown across recurring vs. event-driven?

Structure the answers into this format:

## Daily/Weekly Rhythm

### Morning Block ([X] minutes)
1. [Action]: [description] → [output]
2. [Action]: [description] → [output]
3. [Action]: [description] → [output]

### Recurring Actions ([cadence])
| Action | Trigger | Duration | Output |
|--------|---------|----------|--------|
| [X] | [when] | [how long] | [what it produces] |

### Event-Driven Actions
| Event | Response | Time Sensitivity |
|-------|----------|-----------------|
| [X] | [what to do] | [immediate / same day / within week] |

### Time Budget
- Total: [X] min/day or hours/week
- Breakdown: [X]% scanning, [Y]% creating, [Z]% reviewing

Step 2: Pipeline Deep Dive

For each pipeline stage identified in /discover-domain:

1. Entry criteria: What qualifies something to enter this stage?
2. Actions at this stage: What does the user do here?
3. Exit criteria: What moves it to the next stage?
4. Death criteria: What kills it at this stage? How common is this?
5. Healthy conversion rate: What % should move forward?
6. Time at stage: How long should something spend here before it's stale?

Structure into:

## Pipeline: [Name]

| Stage | Entry Criteria | User Actions | Exit Criteria | Death Rate | Healthy Conversion | Max Time |
|-------|---------------|-------------|--------------|-----------|-------------------|----------|
| [1] | [X] | [X] | [X] | [X]% | [X]% | [X days] |
| [2] | [X] | [X] | [X] | [X]% | [X]% | [X days] |

Step 3: Bottleneck Patterns

Ask the designer:

1. What's the most common bottleneck? Where do things pile up? What causes the jam?
2. What are the 3-5 bottleneck patterns? (e.g., "too many leads, not enough qualification" or "proposals take too long because of research time")
3. For each pattern, what's the diagnostic? How do you know this bottleneck is happening? What data would show it?
4. For each pattern, what's the fix? What should the user do differently?

Structure into:

## Bottleneck Patterns

### Pattern 1: [Name]
- **Symptom:** [what the user sees]
- **Diagnostic:** [what data confirms it]
- **Cause:** [why it happens]
- **Fix:** [what to change]

### Pattern 2: [Name]
...

Step 4: Automation Candidates

Based on everything above, identify:

1. What should run automatically on a schedule? (equivalent to the morning briefing)
2. What should auto-trigger after another action? (equivalent to "after resume-tailor, run ATS check")
3. What should the system track passively? (equivalent to pipeline health monitoring)

Output

Write the complete workflow map to domain-input/domain-workflow.md, structured with all sections above.

Tell the designer:

Workflow mapped. Key metrics:
- Daily time budget: [X] minutes
- Pipeline stages: [N] stages, key death point at [stage] ([X]% drop-off)
- Bottleneck patterns: [N] identified
- Automation candidates: [N] scheduled, [N] auto-triggered

Next: Run /define-audiences to detail who evaluates the user's work.

Quality Checks

Good workflow mapping:

• Every pipeline stage has a conversion rate, not just a name
• Bottleneck patterns have diagnostics (how to detect), not just descriptions
• Time budget is realistic (not "spend 8 hours a day on this")
• Automation candidates map to specific recurring actions from Step 1

Bad workflow mapping:

• Pipeline stages are just names without entry/exit criteria
• No conversion rates ("I don't know" is acceptable — "not included" is not)
• Bottleneck patterns are vague ("things slow down sometimes")
• No automation candidates identified