Analytics Specialist Job Description Template

SQL for analytics · 28%
Most analytics work here depends on accurate, performant SQL to produce trustworthy datasets and metrics. Evaluated in Round 1.

• Weak: Writes SQL that produces incorrect results in edge cases or cannot explain query logic and assumptions.
• Meets the bar: Builds correct SQL queries for joins, aggregations, and window calculations; identifies and improves obvious performance issues.
• Strong: Designs robust SQL with clear grain, handles edge cases, validates outputs with checks, and optimizes queries for performance and maintainability.

BI dashboard building · 20%
Dashboards and reporting assets are the primary interface to insights for stakeholders. Evaluated in Round 1.

• Weak: Creates dashboards without consistent filters/definitions or relies on unclear metrics that cause confusion.
• Meets the bar: Builds usable dashboards with correct filters, clear metric labels, and an understandable layout aligned to stakeholder needs.
• Strong: Creates self-serve dashboards that reduce repeat questions, includes appropriate drill-downs, and ensures performance and consistency across views.

Metric definition consistency · 16%
Consistent KPI definitions prevent reporting disputes and enable reliable trend tracking. Evaluated in Round 2.

• Weak: Defines metrics in an ad hoc way that differs between reports or lacks documented assumptions.
• Meets the bar: Creates and documents consistent metric definitions and applies them consistently across dashboards and analyses.
• Strong: Leads alignment on metric definitions across teams, anticipates ambiguity (grain, inclusion/exclusion rules), and prevents future drift with documentation and validation.

Analytics data modeling · 14%
Correct grain and data modeling ensure metrics are computed reliably across multiple use cases. Evaluated in Round 2.

• Weak: Builds datasets without clear grain and inclusion rules, causing metrics to double-count or misalign across reports.
• Meets the bar: Creates fact/dimension-aligned models with an explicit grain, correct joins, and stable keys.
• Strong: Designs analytics models that support multiple KPIs cleanly, minimizes duplication, and enables easy extension while preserving correctness.

Data pipeline debugging · 12%
Analytics depend on scheduled data transformations; issues must be isolated and fixed quickly. Evaluated in Round 3.

• Weak: Cannot isolate pipeline failures or lacks a repeatable approach to debugging data transformations.
• Meets the bar: Diagnoses common pipeline issues using logs/outputs, fixes transformation problems, and ensures outputs match expectations.
• Strong: Proactively improves pipeline reliability with checks, clear error handling, and validation steps that prevent bad data from reaching reporting.

Data quality validation · 10%
Trustworthiness depends on systematic validation and strong root-cause investigation. Evaluated in Round 3.

• Weak: Detects issues late, relies on intuition, and provides no clear validation method or documented root cause.
• Meets the bar: Runs reconciliation and validation checks, documents findings, and applies fixes with verification steps.
• Strong: Implements scalable data quality checks, uses trend/threshold-based thinking for early detection, and drives preventive improvements.