As a someone returning to Power BI after taking time off to focus on school, I often asked myself whether I could still keep up with the fast pace of the data field. Everything moves so quickly, and coming back felt overwhelming at first. When I saw this challenge, I felt it could be the perfect opportunity to get my hands dusty again, rebuild confidence and learn something new. I interpreted the theme as simple, creative yet insightful and that influenced my thought-process.
My thought process was straightforward: If someone is exploring colleges, what do they care about?
Cost, career opportunities, student experience, satisfaction, programs and all the things that help people make informed decisions. That guided how I structured the report. First, an overview to gain insights about the colleges, then help users dive deeper. Lastly, offer a personalized recommendation by answering a few questions.
✨ Tools and Approach
I kept my process simple and focused by working entirely inside Power BI Desktop.
All data transformation, modeling, visual building, and calculation work was done directly in Power BI, using a combination of:
- Power Query for lightweight transformations
- DAX for calculated measures, normalization, scoring, and ranking logic
- Custom visuals where necessary, especially when the default visuals couldn’t communicate my ideas clearly
- Disconnected tables to power my recommendation quiz without affecting the main data model
- Buttons and navigation features to guide the user smoothly through the report
- Tooltips to add context and enrich the storytelling without cluttering the main pages
✨ Report Walkthrough
- Overview - Affordability & Career Outcomes
This report starts with a simple overview of the Fabric Land data; 10 colleges across 7 kingdoms, each with different strengths. One of the first patterns I noticed was how admission difficulty naturally grouped the colleges: premium schools with high tuition and high difficulty, medium-tier schools in the middle, and easier, lower-cost schools. I also noticed same for the placement rate.
But when I compared tuition to average graduate salary, I realized high tuition maynot always translate to high ROI (starting salary). A premium school like Silverleaf doesn’t necessarily provide better salary outcomes than some medium-tier schools like Goldspire or Ironforge. This already shows users that “expensive” isn’t always “better,” and that the best choice depends on what they care about.
- Overview - Student Experience, Satisfaction & Magic Score
The next page focuses on student life. I made use of two composite measures:
Magic Score: composite score showing ranking + placement + happiness
Student Experience Score: created from housing quality, diversity index, facility rating, support programs, number of clubs and campus events. These metrics were on different scales and were normalized to make fair comparisons.
This section helped reveal that even though premium schools dominate the Magic score, some medium-tier schools actually offer a better student experience. It shows how different metrics can tell different parts of the story.
- Deep Dive (Graduation Rates, Programs, Career Paths)
This page dives deeper into program-level insights such as graduation rates, program difficulty, and program offerings. Premium schools tend to have higher graduation rates and tougher programs, which aligns with their positioning.
I also added an image visual to display the campus type of each college i.e forest, coastal, mountain, or urban. Even though the dataset is fictional, I wanted users to see the environment each college sits in, because that can influence a person’s decision too.
- Recommendation Quiz
The final page is a simple recommendation engine. I didn’t want to overwhelm users with too many options, so the quiz is based directly on the four pillars of my report: Cost, Career outcomes, Student experience and Satisfaction. Users rate each on a 1–5 scale. Their inputs are normalized and fed into a weighted DAX scoring system that returns the Best Match, Next Best Match and the Least Match. From there, users can explore the suggested college on the Explore page.
✨ Challenges and Lessons Learned
- For the image visual, I had initially tried using a custom visual and even manipulated a card visual, but neither could give me the full experience I was looking for. Then, Power BI released the new Image visual right in the middle of this challenge, and it was perfect.
- Another challenge I faced was creating the career-path visual using Deneb. I didn’t want a table or bar chart, and building those small rounded rectangles in Deneb took trial and error, especially making them consistent and not random. But it pushed me to learn more about how flexible Deneb really is.
Overall, I learned a lot from this challenge and I am happy I got to do it.
✨ Connect with Me here: Linkedin
I’d appreciate feedback and would be excited to collaborate on future Power BI projects.
eyJrIjoiMzRlYzllZTQtN2YwMS00NWJhLWE2YWYtNDRkZGY1NTFjZGU4IiwidCI6ImU2MDMxZDZhLWRiOWMtNDRiMy1iYjA5LTRkYmM0NWJiOTAzYyJ9&embedImagePlaceholder=true&pageName=92cd11376e0dd0176656
