All lessons

Module 9: Capstone and interviews · Lesson 9.3 · about 12 min

Writing it up for your portfolio

Turn the capstone into a portfolio page and an interview story. What to include, a template, screenshots, and how to talk about it in two minutes.

Module 9 · about 12 minutes · read first: 9.2. You’ll have a portfolio write-up and a two-minute interview story for the dashboard.

1. Why this matters

A dashboard on your laptop isn’t visible to employers. A clear write-up with screenshots, on your portfolio site and linked from LinkedIn and your CV, is.

2. The page template

  1. Title and one line: “Store performance dashboard for a 5-store retailer: weekly view of sales against last year and target.”
  2. The problem: managers answered “how are we doing?” with spreadsheets; numbers disagreed.
  3. The users and questions: the table from Lesson 9.1, shortened.
  4. The model: a screenshot of Model view; one sentence on the star schema and the date table; how targets at month grain were brought in (TREATAS).
  5. Key measures: 3 or 4, with the DAX and one line each on why (Pace %, vs Target %, Margin %, Product Rank).
  6. The pages: one screenshot each, with the question it answers.
  7. Security and performance: RLS roles; Performance Analyzer result.
  8. What I’d do next: like-for-like, stock data, a calculation group.
  9. Honest note: “Data is a realistic synthetic dataset (Northbay Goods, fibub.com); the modelling and DAX are my own.”

Keep it to one screen of text plus images. Put the .pbix and screenshots in a GitHub repository with a short README (the TMDL/.pbip format shows your model as readable code).

3. The two-minute story (STAR)

  • Situation: a retailer whose store managers had no reliable weekly view.
  • Task: one report for store, regional and head-office users, against last year and target.
  • Action: modelled a star schema with a marked date table; wrote time intelligence for pace and growth; brought monthly targets to daily data with TREATAS; drill through from region to store; RLS by store; checked every total against a test query.
  • Result: one Monday report for three audiences; numbers that reconcile; under one second per visual.

Practise it aloud until it takes two minutes.

Principle

Principle: show decisions, not just results. “I chose single-direction relationships and counted products in the fact table instead of making the relationship bidirectional, because…” tells an interviewer more than any screenshot.

Try it: write down three decisions you made in the build and the reason for each.

4. Practice

  1. Write your project’s one-line description.
  2. Pick the measure you’re proudest of and explain it in two sentences without saying “CALCULATE”.
  3. What would you answer to “what would you improve?”
Answers
  1. Example: “A weekly store performance report showing each manager their sales against last year and target, built in Power BI on a star schema.”
  2. Example: “Pace compares this year so far with the same days last year, so managers see whether the year is on track. It stays steady even when one week is unusual.”
  3. Example: “Add stock data to explain lost sales, like-for-like growth, and automated tests of key totals after each refresh.”

Check yourself

1What should the first line of a portfolio project say?
2Which is the better way to describe a measure in an interview?
3The data is made up. Should you say so?

Your ticks stay in this browser. Save them as a file on the course home.

Updated 2026-11-22. Northbay Goods is a made-up shop and its numbers are generated. Power BI changes every month: menu names can move, so check Microsoft Learn if a button is not where the lesson says. The practice editor runs in your browser; nothing you type is sent anywhere.