From: Hiroshi Lund, operations manager
For: Monday's team meeting
"Can you get me the April numbers? I need how many orders we took, the revenue, the average order, our best product, and how many new customers signed up. Then tell me which category to feature in the May newsletter: the one that earned the most in April.
Leave cancelled orders out of everything. Returned orders still count; finance handles refunds separately."
Data: orders, order_items, products, customers. April means April 1 to 30, 2025.
Deliverable: one query per question. The last step is the recommendation.
A case study is a business question answered in steps. Each step is one query, checked by its result, and the last step is the answer the manager asked for.
Read the brief first: it fixes the rules that apply to every step, here the date range and leaving out cancelled orders. Keeping those rules the same across every query is most of the work in real reporting, because two numbers built on different rules can't be compared.
You'll use filtering, aggregates and joins from the beginner units. Name your result columns as each step asks, so the numbers are easy to read out in the meeting.