Skip to content

Certification Roadmap - Data Analyst

A data analyst answers business questions with data: pull it with SQL, shape it, chart it, and explain what it means. This roadmap lists only the certifications that have a study guide in this repo. It is short on purpose. For an analyst, the portfolio matters more than the badge, and most of the work is SQL and judgment, which no exam fully tests.

If you have never written a query, start with Data from Scratch and the day-one pages, then come back.

The path

flowchart LR
    A[SQL and spreadsheets] --> B[DP-900 vocabulary]
    B --> C[PL-300 Power BI Data Analyst]
    C --> D{Next step}
    D --> E[DP-600 Fabric Analytics Engineer]
    D --> F[SnowPro Core]
    D --> G[Analytics engineer roadmap]

The diagram shows one main line (SQL, vocabulary, a BI tool exam) and three ways to go deeper.

Step 0 - Skills before any exam

Step 1 - Vocabulary: DP-900

Azure Data Fundamentals (DP-900) covers core data concepts: relational and non-relational data, analytics workloads. It is optional if you already work with data, and a gentle first exam if you do not.

Step 2 - The analyst exam: PL-300

Power BI Data Analyst (PL-300) is the one analyst-tier certification in this repo. It covers preparing data, modelling it, visualizing it and managing the result. Pair it with the BI tools comparison so you know where Power BI sits next to Looker, Tableau and the others.

Step 3 - Pick a direction

If you want Next Why
Deeper Microsoft-stack work DP-600 Fabric Analytics Engineer Semantic models, warehouses, lakehouses, SQL and DAX
A warehouse skill that travels across clouds SnowPro Core Snowflake is common in analytics jobs; see the warehouse comparison
To own the transformation layer Analytics engineer roadmap Moves you from reading tables to building them

What this roadmap does not cover

Some analyst-focused certifications from other vendors have no guide in this repo (Databricks Data Analyst is on the TODO list as a candidate). This page does not recommend a cert it cannot point to.

Build proof as you go

  • Rebuild a public dataset as a small star schema: Design a star schema.
  • Publish one dashboard with a clear question, a chart per answer and a note on data quality.
  • Write down every metric definition. See Semantic layer.