Classic Models: Sales Order Lines
Every order line from a model-car wholesaler: customer, country, product line, quantity and price. Build a PivotTable and answer four questions.
Start here
Lessons and reference
Look things up
Start here
SQL
Power Query
Data modeling
Power BI and DAX
Finance playbooks
Run queries
Build and check
Practice files built from the live LearnSQL databases: Excel workbooks with instructions, questions and a Check Your Work answer sheet, and complete Power BI projects that teach DAX, relationships, time intelligence and security on real data.
Open in Excel (any recent version). Each has the data, instructions, questions and a Check Your Work sheet.
Every order line from a model-car wholesaler: customer, country, product line, quantity and price. Build a PivotTable and answer four questions.
Order lines with quantities, prices and discounts. Practice calculated columns, SUMIFS and lookups.
Three years of retail sales summarized by month, channel and category. A realistic starting point for PivotTables, slicers and margin analysis.
Population, area, GNP and life expectancy for every country. Good for filters, sorting and dealing with blanks.
Thousands of rental payments with customer, date and film category. Practice grouping dates and finding top customers.
The Northwind orders data with realistic problems planted in it: blanks, stray spaces, numbers stored as text and duplicate rows. Find and fix them.
Complete, working Power BI Desktop projects on the Contoso database: a data model, DAX measures, and report pages. Extract the whole zip, open the .pbip file in Power BI Desktop, click Refresh, and sign in with the SQL Server login that comes with your CPE Today course. Each project also contains a TMDL script that rebuilds the model if it will not open.
Calculation groups, field parameters, running totals, a user-defined function and quick measures.
CALCULATE with Boolean filters, REMOVEFILTERS, ALLSELECTED, KEEPFILTERS and FILTER.
SUM, COUNTROWS, DIVIDE, calculated columns, variables and the first measures.
SUMX, AVERAGEX, RANKX, TOPN and row context versus filter context.
Relationships, RELATED, RELATEDTABLE, inactive relationships and USERELATIONSHIP.
Drillthrough and tooltip pages, navigation buttons, conditional formatting, dynamic titles, mobile layout.
Row-level security roles (static and USERPRINCIPALNAME) plus an object-level security script.
Date table, YTD, prior year, year-over-year, rolling 12 months and fiscal periods.
Every core visual type once, on the Contoso star schema.
Excel: read the Instructions sheet first, do the work on Data and Output, then compare with Check Your Work. Power BI: extract the zip, open the .pbip, refresh, then open each page and the measures in Model view to see how the DAX works. The data is real rows from the sample databases, so the same questions can also be answered with SQL in the Playground, and you can pull the live data into Excel or Power BI yourself (how). Stuck on DAX? Ask an AI assistant connected through the MCP server: it can look up DAX patterns and explain them.
These databases contain made-up data for learning. The Contoso files use a 10% sample of the full sales table so they stay small and quick to open. The messy-data file has problems planted on purpose; the hidden About sheet lists them if you get stuck.