Customization Guide — On-Call & Incident Tracker

Adapt this incident management system to your team's size, tooling, and processes.

Team Size Adaptations

Small Team (2-4 engineers)

• Simplify to Primary On-Call only (remove Secondary column)
• Use weekly rotations — each person is on-call every 2-4 weeks
• Collapse Tier 2 and Tier 3 escalation into one level
• Everyone handles all severity levels

Medium Team (5-15 engineers)

• Keep Primary + Secondary on-call
• Use biweekly rotations to reduce fatigue
• Add specialist on-call for specific domains (database, security)
• Split Tier 1 (first responder) and Tier 2 (specialist) clearly

Large Team (15+ engineers)

• Add follow-the-sun rotations across time zones
• Create per-service on-call rotations (platform, infrastructure, product)
• Add Tier 3 management escalation for SEV1
• Consider separate "Incident Commander" and "Technical Lead" roles

Adding Follow-the-Sun Coverage

Add a "Region" select to the On-Call Schedule:

• AMER (Americas): coverage 09:00-21:00 UTC-5
• EMEA (Europe/Middle East): coverage 08:00-20:00 UTC+1
• APAC (Asia-Pacific): coverage 09:00-21:00 UTC+9

Create a Timeline view grouped by Region to visualize 24-hour coverage.

Severity Level Customization

Adding SEV0 (Existential)

For companies where outages have extreme business impact:

SEV0 should page everyone in Tier 1 + Tier 2 simultaneously and auto-notify Tier 3 management.

Simplifying to 3 Levels

If SEV3 and SEV4 feel redundant:

Update the Severity select options and adjust all formula references.

Integrating with Alerting Tools

This Notion system complements (doesn't replace) alerting tools. Here's how they fit together:

Notion API Automation Ideas

1. Auto-create incident — Webhook from PagerDuty creates a new Incidents database entry

2. SLA monthly update — Cron job pulls uptime data from monitoring and updates SLA Targets

3. On-call reminder — Send Slack message on rotation day linking to handoff notes

4. Postmortem deadline — Auto-create postmortem entry 24h after incident resolution

Postmortem Process Customization

Adding a Review Meeting Workflow

Add these properties to the Postmortems database:

• Review Date (Date) — when the postmortem review meeting is scheduled
• Attendees (Relation → Escalation Contacts) — who should attend
• Review Status (Select) — Draft → Ready for Review → Reviewed → Actions Complete

Custom Contributing Factor Categories

Replace the free-text Contributing Factors with a Multi-select:

• Configuration Error
• Code Bug
• Capacity Planning
• Monitoring Gap
• Process Failure
• External Dependency
• Communication Breakdown
• Documentation Gap

This enables filtering postmortems by common failure patterns over time.

Action Item Tracking Enhancement

Create a separate Action Items database instead of tracking counts:

• Action Item (Title)
• Postmortem (Relation → Postmortems)
• Owner (Relation → Escalation Contacts)
• Priority (Select: P0, P1, P2)
• Due Date (Date)
• Status (Select: Not Started, In Progress, Done)
• Verification (Text — how to verify the action is effective)

This gives you a dedicated board view of all outstanding action items across incidents.

SLA Tracking Enhancements

Adding Historical SLA Data

Create a separate SLA Monthly History database:

• Service (Relation → SLA Targets)
• Month (Date)
• Uptime (Number, percent)
• MTTR (Number, minutes)
• Incident Count (Number)
• Status (Formula — calculate if SLA was met)

This enables trend analysis over time without overwriting current-month data.

Error Budget Tracking

Add these properties to SLA Targets:

• Monthly Error Budget (minutes) — calculated from uptime target
• Budget Consumed (minutes) — sum of downtime this month
• Budget Remaining (minutes) — formula: budget - consumed
• Budget Status — formula: green if >50% remaining, yellow if 20-50%, red if <20%

View Customization Ideas

War Room View

Create a filtered Board on Incidents:

• Filter: Status is "Active" or "Investigating"
• Group by: Severity
• Properties: Title, Affected Services, Incident Commander, time since detection
• This is your "active fires" view during incident response

Weekly Report View

Create a Table on Incidents:

• Filter: Detected At is within the last 7 days
• Group by: Severity
• Calculate: Count of incidents per severity
• Include: Duration (minutes) with average and total

On-Call Fatigue Dashboard

Track on-call health with a table showing:

• Each team member's total on-call weeks this quarter
• Number of pages received per rotation
• Average sleep interruptions (add as a field in On-Call Schedule)

Notification Preferences

Add a Notification Rules page documenting when to use which channel: