From “What Should We Measure?” to Live Dashboards: What We Learned
We recently built a set of dashboards for MISBO. You can take a look at them here (you need to be logged in to view): https://www.misbo.com/misbo-dashboard-mockup. This was the great work of the Emerging Issues and Solutions committee of the MISBO board.
They turned out great. Clean, useful, and honestly kind of fun to look at (which is not something we usually say about operational data). But here’s the truth: building the dashboards wasn’t the hardest part. Figuring out what we actually wanted to see was.
Step One: The Hard Question
Before we wrote a single line of code or prompted a single AI tool, we had to answer a deceptively simple question: What matters most? For us, as an association, that led to five core areas: Membership, Finances, Purchasing consortium, Professional development and engagement, Retirement plan
Each of those turned into a dashboard with clear metrics. Things like member retention, revenue streams, consortium savings, and event participation. That worked well for MISBO because we are an association, but you need something school specific. Your dashboard should reflect your daily reality. Different mission, different pressures, different “what keeps you up at night.” Here’s a starting point we’d suggest:
1. Enrollment & Retention
- Current enrollment vs. capacity
- Applications, acceptances, yield
- Retention by grade
- Attrition trends
2. Financial Health
- Tuition revenue vs. budget
- Financial aid awarded
- Net tuition per student
- Expense categories vs. plan
3. Advancement & Fundraising
- Annual fund participation (by constituency)
- Total dollars raised
- Major gift pipeline
- Event ROI
4. Student Experience
- Attendance trends
- Student-teacher ratios
- Program participation (arts, athletics, clubs)
- Survey or satisfaction indicators
5. Staffing & Operations
- Faculty retention
- Open positions / time to hire
- Compensation benchmarks
- Facilities or maintenance metrics
That’s not a perfect list. It’s a starting point. Your version should reflect your strategy, your culture, and your board’s priorities.
Step Two: Prompting Matters More Than You Think
We built our dashboards using Claude. Not because it’s magic, but because it’s fast, flexible, and surprisingly good at visual thinking when you give it the right inputs. The key wasn’t just asking for a dashboard. It was giving context: Who we are (a 400-school association). What we care about (those five areas). How we wanted it to feel (clean, minimal, lots of white space). Where it would live (web + presentation).
Once we did that, the outputs got dramatically better. Same idea applies to schools. If you say “build me a dashboard,” you’ll get something generic. If you say “build me a dashboard for a college-prep school focused on enrollment stability and financial sustainability,” now you’re getting somewhere.
Step Three: Design Still Matters (A Lot)
We didn’t want something cluttered or overly “techy.” We leaned into: simple layouts, clear hierarchy, lots of white space, strong use of our brand. That aligns with our broader design principles, especially clarity and simplicity. Translation: if people can’t read it in five seconds, it’s not helping.
Step Four: From Static to Live
Right now, our dashboards are essentially a polished prototype. The visuals are there. The structure is there. The next step is connecting live data sources so everything updates automatically. For schools, that might mean pulling from:
- Your SIS
- Your financial system
- Your advancement platform
- Survey tools
This is where things move from “interesting” to “indispensable.”
Final Thought: Don’t Start with the Tool
It’s tempting to jump straight into AI, dashboards, or software. Don’t. Start with the question we wrestled with: What do we actually need to see to make better decisions? Answer that well, and the rest gets a lot easier. Skip that step, and you’ll end up with a beautiful dashboard that no one uses, and we’ve all got enough of those already.