Retrospective

Retrospective Skill

Facilitate and document sprint retrospectives, release retrospectives, and post-project reviews for Talosix teams. Generate structured retrospective documents that capture team insights and drive continuous improvement.

When to Use

• At the end of a sprint
• After a release deployment
• After a project milestone or completion
• After a significant incident (post-incident review)
• When a team wants to reflect on a specific period or event

Retrospective Formats

Format 1: Standard Sprint Retrospective

# Sprint Retrospective: Sprint <number>
Date: <date>
Team: <team name>
Facilitator: <name>
Attendees: <list>

## Sprint Summary
- Sprint Goal: <goal>
- Committed: <X> story points / <Y> tickets
- Completed: <X> story points / <Y> tickets
- Carry-over: <list of incomplete items>

## What Went Well
1. <item>
2. <item>
3. <item>

## What Didn't Go Well
1. <item>
2. <item>
3. <item>

## What Can We Improve
1. <item>
2. <item>
3. <item>

## Action Items
| # | Action | Owner | Due Date | Status |
|---|--------|-------|----------|--------|
| 1 | <action> | <name> | <date> | Open |
| 2 | <action> | <name> | <date> | Open |

## Follow-Up from Previous Retro
| # | Action | Owner | Status | Notes |
|---|--------|-------|--------|-------|
| 1 | <previous action> | <name> | <status> | <notes> |

Format 2: Release Retrospective

# Release Retrospective: v<version>
Date: <date>
Release Date: <release date>
Team: <team name>
Facilitator: <name>

## Release Summary
- Version: <version>
- Features delivered: <count>
- Bugs fixed: <count>
- Planned release date: <date>
- Actual release date: <date>
- Deployment duration: <time>
- Rollback required: Yes/No

## Planning & Scope
### What went well
- <items>

### What didn't go well
- <items>

## Development & Quality
### What went well
- <items>

### What didn't go well
- <items>

## Validation & Compliance
### What went well
- <items>

### What didn't go well
- <items>

## Deployment & Release
### What went well
- <items>

### What didn't go well
- <items>

## Metrics
| Metric | Target | Actual |
|--------|--------|--------|
| Defects found in validation | <target> | <actual> |
| Defects escaped to production | 0 | <actual> |
| Validation cycle time | <target> | <actual> |
| Deployment downtime | <target> | <actual> |
| Post-release incidents | 0 | <actual> |

## Key Learnings
1. <learning>
2. <learning>

## Action Items
| # | Action | Owner | Due Date | Category |
|---|--------|-------|----------|----------|
| 1 | <action> | <name> | <date> | Process |
| 2 | <action> | <name> | <date> | Technical |

Format 3: Post-Incident Review

# Post-Incident Review: INC-<number>
Date: <date>
Incident Date: <incident date>
Severity: SEV-<level>
Facilitator: <name>
Attendees: <list>

## Incident Summary
- Description: <what happened>
- Duration: <time to resolve>
- Impact: <users/studies affected>
- Clinical Impact: <assessment>

## Timeline Review
| Time | Event |
|------|-------|
| <time> | <event> |

## What Went Well in Response
1. <item>

## What Could Be Improved in Response
1. <item>

## Root Cause
<root cause summary>

## Prevention
| # | Preventive Action | Owner | Due Date | Type |
|---|-------------------|-------|----------|------|
| 1 | <action> | <name> | <date> | Technical/Process |

## Detection Improvement
How can we detect this faster?
1. <item>

## CAPA Required: Yes/No
If yes: CAPA-<number> initiated

Facilitation Guidelines

Before the Retrospective

1. Review previous retrospective action items and their status
2. Gather data: sprint metrics, deployment logs, incident reports
3. Send a brief pre-retro survey to collect initial thoughts
4. Prepare the retrospective template

During the Retrospective

1. Set the stage (5 min): Review the purpose, establish psychological safety
2. Gather data (15 min): Have each participant share observations
3. Generate insights (15 min): Group similar items, identify patterns
4. Decide what to do (15 min): Prioritize and assign action items
5. Close (5 min): Summarize, confirm action items, gather feedback

Facilitation Principles

• Blameless: Focus on systems and processes, not individuals
• Specific: "We need better testing" becomes "Add integration tests for the query workflow"
• Actionable: Every improvement becomes an action item with an owner
• Limited: Pick 2-3 action items maximum to avoid overload
• Tracked: Follow up on previous retro actions at the start of each retro

Clinical Trial EDC Context

When facilitating retrospectives for Talosix teams, consider these domain-specific areas:

• Validation efficiency: Are validation cycles too long? Can we improve IQ/OQ automation?
• Regulatory compliance: Were there any compliance gaps or near-misses?
• Audit readiness: Would we be comfortable with a regulatory audit of this sprint/release?
• Data integrity: Were there any data integrity concerns raised?
• Clinical workflow impact: Did changes improve or complicate site user workflows?
• Cross-functional collaboration: How well did Engineering, QA, Clinical, and Regulatory work together?

Action Item Best Practices

Good action items are:

• Specific: Clear description of what needs to happen
• Measurable: You can tell when it's done
• Owned: One person is accountable (not "the team")
• Time-bound: Has a due date (ideally within the next sprint)
• Realistic: Can be completed alongside sprint work

Examples:

• "Add automated smoke test for e-signature flow" (Owner: Jane, Due: Sprint 24)
• "Update deployment runbook with new database migration steps" (Owner: Bob, Due: Feb 28)
• "Schedule cross-team sync between Engineering and Clinical Ops for study launches" (Owner: Alice, Due: Next sprint planning)

Output Format

Generate retrospective documents in markdown using the appropriate format above. Include all sections even if some have placeholder text, so nothing is missed. When processing raw notes from a user, organize them into the structured format and suggest action items based on the themes identified.