Excel and Power BI become genuinely useful when every row, calculation and filter has a clear meaning. Start with data, not chart colours.
1. Define one row
A row may represent an order, a sold item or a monthly total. Do not mix these grains in the same fact table: an order with several items could otherwise be counted multiple times.
2. Check types and values
A date stored as text or an amount with a misread separator can distort analysis. Check duplicates, missing values and units before aggregating. Keep a reference total to compare against the result.
3. Explain a metric
For every metric, write its business definition, period, exclusions and filters. Revenue may include or exclude taxes and returns: those choices must be explicit for two reports to remain comparable.
Date | Order | Amount (MAD)
2026-10-01 | A001 | 120
2026-10-01 | A002 | 80
2026-10-02 | A003 | 150
Control total: 350 MAD
One row = one order in this exampleYour turn to build
Describe a sales tracking table: what does one row represent, what is the amount’s unit and how will you validate the total? Then propose two metrics and their definitions.
Your discovery progress is saved on this device. You can clear it from My learning.