Data Analyst onboarding plan
Analysts lose most early time figuring out which tables and definitions to trust. Onboarding should hand them the key metric definitions and a known-good report to rebuild first.
What good onboarding achieves
Produce trusted analyses using agreed metric definitions and data sources, and communicate findings clearly to stakeholders.
Week one example
- Day 1: Warehouse, BI and repository access; meet the data team.
- Day 2: Core business model and metric definitions.
- Day 3: Data sources tour: what's reliable, what's known to be messy.
- Day 4: Rebuild an existing report to learn the data.
- Day 5: Meet two stakeholder teams and their recurring questions.
30/60/90 progression
- Day 30: Delivers a first analysis that matches existing numbers.
- Day 60: Owns recurring reports for one stakeholder team.
- Day 90: Runs an exploratory analysis that informs a decision.
People to meet (examples)
- Data lead
- Analytics engineer
- Stakeholders in product, finance or operations
- Owner of key data sources
Systems to learn (examples)
- Data warehouse
- BI tool
- SQL or notebook environment
- Metric definitions document
Manager checkpoints
- Week 1: rebuilt report matches
- Day 30: first analysis review
- Day 60: stakeholder feedback
- Day 90: decision impact
Common mistakes
- No written metric definitions
- Starting with an ambiguous question
- Not flagging known data quality issues
- No stakeholder introductions
Checklist
- Warehouse and BI access
- Metric definitions reviewed
- Data sources toured
- Existing report rebuilt
- Stakeholders met
- Review process for analyses explained
- Recurring reports assigned
- 90-day goals agreed