Sales data analysis · 2025

A sales review the client still opens

ShippedIndependent engineer

Seven months of sales records raised the same questions at every review: what drove revenue, where the gaps were, and who returned. The answers lived in a spreadsheet only one person trusted.

The constraint

The deliverable had to survive the meeting. The client already worked in Excel and Power BI, so the result had to fit those tools rather than introduce a new platform.

Built with

  • Power BI
  • SQL
  • Excel

How it works

Records into a model. Excel for the working surface. Power BI for the view that is opened each quarter.

  1. Sales records
  2. Model
  3. Excel
  1. Power BI
  2. Quarterly review
Records into a model. Excel for the working surface. Power BI for the view that is opened each quarter.

The calls that shaped it

Each decision with the pressure that forced it and the price it keeps costing.

  1. Model first

    Drivers, regions, and returning-customer behaviour are defined in the model, then shown. Charts without that work are decoration.

  2. Workbook and Power BI

    Both, because the review still happens in those tools. I did not wrap the data in a web app they would stop opening.

The model is what survives the meeting. Whether the deliverable is a worksheet or a dashboard, the analysis is built first so the review still holds six months later, not a chart dropped on an unexplored sheet.

Where it stands

Shipped. A model the client still opens each quarter, built so the review holds up long after we stop working together.

What was handed over

  1. The workbook
  2. The Power BI view
  3. Which questions the model actually answers
Next project Technical SEO audit tool Technical checks after a template change