playbooks

home / skills / gtmagents / gtm-agents / data-governance

data-governance skill

/plugins/marketing-automation/skills/data-governance

This skill helps you govern data contracts, consent policies, and monitoring for automation programs across systems.

npx playbooks add skill gtmagents/gtm-agents --skill data-governance

Review the files below or copy the command above to add this skill to your agents.

Files (1)
SKILL.md
1.2 KB
---
name: data-governance
description: Use when defining data contracts, consent policies, and monitoring for
  automation programs.
---

# Automation Data Governance Skill

## When to Use
- Launching new journeys requiring cross-system data sharing.
- Auditing consent/suppression logic across email/SMS/in-app.
- Investigating data quality incidents impacting automation.

## Framework
1. **Data Contracts** – document required fields/events, owners, freshness SLAs, fallback behavior.
2. **Consent & Compliance** – track opt-in types, regional consent, TTL, audit trails.
3. **Identity Resolution** – ensure consistent IDs across product, CRM, MAP, CDP.
4. **Monitoring** – dashboards/alerts for data latency, schema changes, null spikes.
5. **Change Management** – versioning, rollback, and communication paths.

## Templates
- Data requirements matrix (journey → fields/events → source → owner → SLA).
- Consent policy doc (channel, region, legal basis, suppression rules).
- Incident log + RCA template.

## Tips
- Set automated kill switches when critical fields are stale or missing.
- Collaborate with security/legal on retention + privacy impact assessments.
- Align governance cadences with quarterly automation retros.

---

Overview

This skill helps teams define and enforce data contracts, consent policies, and monitoring for automation programs. It codifies ownership, SLAs, and fallback behavior so automated journeys run reliably and compliantly. Use it to reduce data incidents, speed launches, and maintain auditability across systems.

How this skill works

The skill provides a practical framework: create data contracts that list required fields, owners, and freshness SLAs; capture consent and suppression rules by channel and region; and map identity resolution across systems. It also specifies monitoring signals and alerting behavior, plus change management practices like versioning and rollback. Templates accelerate adoption: requirement matrices, consent policy docs, and incident/RCA logs.

When to use it

  • Launching cross-system journeys that rely on shared data feeds
  • Auditing consent, suppression, or opt-out logic across email, SMS, and in-app channels
  • Investigating data quality incidents affecting automation performance
  • Defining ownership and SLAs before onboarding new event sources or integrations
  • Adding automated kill-switches or safety checks to critical flows

Best practices

  • Document required fields, owners, freshness SLAs, and fallback behavior in a data contract
  • Record opt-in types, regional legal basis, TTLs, and an auditable suppression trail
  • Map consistent identity resolution across product, CRM, MAP, and CDP
  • Instrument dashboards and alerts for latency, schema changes, and null/empty spikes
  • Use versioning, rollout plans, and clear rollback paths for schema or policy changes

Example use cases

  • Define a data requirements matrix for a new onboarding journey driven by product events
  • Audit and remediate missing consent flags that caused an SMS suppression failure
  • Create alerts that automatically disable campaigns if critical fields are stale
  • Run quarterly governance reviews aligning retention and privacy with legal
  • Produce incident logs and RCAs after a broken identity sync that affected revenue metrics

FAQ

How quickly should data freshness SLAs be defined?

Define SLAs during design: near-real-time for behavioral triggers, hourly or daily for batch syncs. Match SLA to business impact and include fallback behavior.

What triggers an automated kill switch?

Kill switches should trigger on pre-defined signals like missing required fields, identity resolution failures, or large spikes in nulls that could cause incorrect customer actions.