Stop 13 · BA Domain SpecializationsTopics · 3/7

Data / Reporting Business Analyst

In short

Every manager wants a dashboard. Few can say what the numbers should mean. A data BA defines the right metrics, finds the data and builds reports people actually make decisions with.

Focuses heavily on data analysis, reporting, and dashboards — leans on Phase 5's data skills more deeply than a generalist BA, often requiring stronger SQL and visualization skills.

The purple blob sits at a cricket boundary with a scorebook, drawing a bar chart while a batsman plays.

Think of it like this

The cricket analyst

The scorer records every ball. The team analyst turns that into 'their opener struggles against short balls on off stump'. The captain then sets the field. Raw data to insight to decision: that's the data BA's whole job.

Watch it happen

Breaking a vague goal into measurable metrics

Grow revenue

Customers

New signups

Repeat rate

Avg order value

Items per order

Avg item price

1/5The manager asks for 'a revenue dashboard'. Too vague to build yet.

The big ideas

What you need to know

Stronger SQL and BI

Expect to write SQL daily and build dashboards in tools like Power BI, Tableau or Looker Studio.

Define every metric

Write down exactly what 'active user' or 'revenue' means, including what's excluded. Two teams, two definitions, endless fights.

Data quality matters

Check for duplicates, gaps and odd spikes before you report. A wrong number destroys trust faster than no number.

Insight, not charts

Every report should answer a business question and suggest what to do next.

See it in code

Walk through the code

sql

1-- Metric: Repeat rate (Sept 2026)2-- Definition: share of customers with 2+ paid orders3SELECT4  ROUND(100.0 * COUNT(*) FILTER (WHERE orders >= 2)5        / COUNT(*), 1) AS repeat_rate_pct6FROM (7  SELECT customer_id, COUNT(*) AS orders8  FROM orders9  WHERE status = 'paid'10    AND order_date >= '2026-09-01'11    AND order_date <  '2026-10-01'12  GROUP BY customer_id13) AS per_customer;

1/3Name and definition first, so everyone counts the same thing.

A documented metric in PostgreSQL. The comment lines matter as much as the query: they say exactly what is being counted.

Save yourself the pain

Common mistakes

Oops

Building the dashboard before agreeing what each metric means.

Do this instead

Write and sign off metric definitions first. It saves weeks of 'your number is wrong' arguments.

Oops

Putting 25 charts on one dashboard.

Do this instead

Show the few numbers that answer the main question. Add drill-downs for the curious.

Your mission

Define and build one metric

~40 min

Quick check

Sales and Finance report different 'monthly revenue' numbers. What's the BA's first step?

Go deeper

1Read 2-3 Data/Reporting BA job descriptions and note the data-specific skill expectations.
2Revisit Phase 5 (Data Analysis) and consider deepening your SQL skills for this track.
3Consider this track if Phase 5 was your favorite part of the roadmap.