Runbook vs Playbook: Understanding the Difference
“Should this be a runbook or a playbook?” It’s a question that comes up in every DevOps team. The terms are often used interchangeably, but they serve different purposes.
Understanding the difference between a runbook and a playbook helps you create better documentation for your team. For runbook basics, see how to write a runbook.
Runbook vs Playbook: The Core Difference
What Is a Runbook?
A runbook is a step-by-step procedure for completing a specific operational task. Runbooks are prescriptive—they tell you exactly what to do.
Runbook characteristics:
- Specific, detailed steps
- One procedure per runbook
- Executable commands
- Little decision-making required
- Goal: Complete a known task reliably
Example runbook: “Restart the API service”
# Restart API Service
## Steps
1. Check current pod status
```bash
kubectl get pods -n production -l app=api
```
2. Perform rolling restart
```bash
kubectl rollout restart deployment/api -n production
```
3. Verify restart complete
```bash
kubectl rollout status deployment/api -n production
```
What Is a Playbook?
A playbook is a collection of strategies and decision trees for handling a category of situations. Playbooks are adaptive—they guide you through decisions.
Playbook characteristics:
- Multiple scenarios covered
- Decision points and branching
- Higher-level guidance
- Judgment calls required
- Goal: Navigate complex or uncertain situations
Example playbook: “API Performance Degradation”
# API Performance Degradation Playbook
## Initial Assessment
What symptoms are you seeing?
- **High latency only** → Go to Section A
- **High error rate** → Go to Section B
- **Both latency and errors** → Go to Section C
- **Intermittent issues** → Go to Section D
## Section A: High Latency
### Check Database
Is database CPU > 80%?
- Yes → Scale database, then reassess
- No → Continue to next check
### Check External Dependencies
Are third-party API calls timing out?
- Yes → Enable circuit breaker, notify vendor
- No → Check application-level issues
Runbook vs Playbook: When to Use Each
Use a Runbook When:
- The procedure is well-defined
- Steps rarely change
- Any team member should be able to execute
- The goal is consistency and speed
- Examples: deployments, backups, certificate renewals
Use a Playbook When:
- The situation requires diagnosis
- Multiple root causes are possible
- Experience and judgment matter
- The goal is effective problem-solving
- Examples: incident response, performance issues, outages
How Runbooks and Playbooks Work Together
The best teams use both. Playbooks guide decision-making; runbooks execute the decisions.
## Incident Response Playbook
### Step 1: Identify Impact
Check which services are affected:
```bash
curl https://status.internal/api/health
```
### Step 2: Categorize Issue
Based on symptoms:
- **API returning 500s** → Execute [API Error Runbook]
- **Database connection failures** → Execute [Database Recovery Runbook]
- **High memory usage** → Execute [Memory Leak Runbook]
- **Unknown** → Continue investigation below
### Step 3: Deep Investigation
...
Runbook vs Playbook: A Practical Framework
| Aspect | Runbook | Playbook |
|---|---|---|
| Structure | Linear steps | Branching decisions |
| Scope | Single task | Category of situations |
| Flexibility | Low (by design) | High (by necessity) |
| Author | Any team member | Experienced engineers |
| Updates | When procedure changes | As new scenarios emerge |
| Execution | Mechanical | Requires judgment |
Building Your Documentation Library
Start with playbooks for major incident categories. As patterns emerge, extract runbooks for common procedures.
## Documentation Evolution
1. Incident happens
2. Create playbook entry for this scenario
3. Notice repeated procedure in playbook
4. Extract procedure as standalone runbook
5. Reference runbook from playbook
6. Repeat
Making Both Executable
Whether runbook or playbook, documentation that can’t be executed gets ignored. Learn about interactive documentation and see our runbook examples for practical templates.
Stew makes your runbooks executable—every command runs with a click. Your playbooks become interactive decision trees. The gap between documentation and action disappears.
Join the waitlist and build better operational documentation.