Introduction
Traditional customer rankings tell us who generates the most revenue, but they do not explain what those customers buy. A customer may be highly valuable and still buy almost entirely from one product category – which makes the next conversation much clearer than a generic “upsell” message.
This week, build a Customer Cross-Sell Map in Power BI. Use a ternary chart to position every customer by the share of their revenue that comes from Furniture, Office Supplies and Technology. A customer near one vertex is a category specialist; a customer in the middle has a balanced basket; a customer near an edge has an obvious missing category.
Your job is to turn that position into an action. Show the highest-value customers, make the category mix readable, and use a standard Power BI matrix to reveal the exact revenue, profit and recommended next step behind every point.
I built the example with the free HTML Content custom visual: https://html-content.com/. It uses HTML and SVG only – no external JavaScript, images or remote assets are required. You may use any visual approach that fulfils the requirements.
Requirements
1. Import and model the Sales Orders, Customer, Product and Location data. Ensure each customer can be analysed across the three product categories.
2. Create customer-level measures for Revenue, Profit, Furniture revenue, Office Supplies revenue and Technology revenue. Calculate the three category shares so that they total 100% for each customer in the active filter context.
3. Create a ternary chart in which one dot equals one customer. Aggregate every order for that customer within the active Year, Country and Customer Segment filters. Do not show one point per order.
4. Label the three vertices Furniture 100%, Office Supplies 100% and Technology 100%. Add a grid, readable share labels and a short explanation of how to interpret the three axes.
5. Classify customer baskets using transparent rules:
– Specialist: one category is at least 60% of revenue.
– Balanced: all three categories are at least 20% of revenue.
– Hybrid: every other basket.
6. Use colour to identify the leading category. Furniture, Office Supplies and Technology must use visibly different colours. Override the category colour with red when aggregate customer profit is negative.
7. Size bubbles by customer revenue. Show only a selectable Top 25, 60, 100 or 150 customers, ranked again after filters. Display how much of selected revenue the visible cohort covers.
8. Add slicers for Year, Country, Customer Segment, Customers shown and Customer drilldown. The report should open on Year = 2021 and Country = United States.
9. Add a native Power BI matrix below the map. It must show Customer Name, archetype, Revenue, Profit, the three category shares and a rule-based recommended action. Selecting one row must result in exactly one point on the ternary chart.
10. Add a compact “How to read it” panel. Explain that points near a vertex are category specialists, points near an edge are two-category baskets, and the centre is a balanced basket – not a performance score.
Dataset
Find this week’s dataset in Excel at: Workspaces / Workout Wednesday / 2025 / 2025W21 – Sales
Share
After you finish your workout, share on Bluesky or LinkedIn using the hashtags #WOW2026 and #PowerBI, and tag @MMarie, @shan_gsd, @KerryKolosko.