lesson
30 min

Build a Simple Report

Count and summarize records with Python, then design a dashboard people will actually use.

What you will learn

By the end of this lesson, you can count and summarize records with Python and sketch a dashboard built for one role.

Records in, answers out

A report takes many records and turns them into a few answers. At its core, that is counting, adding, and averaging. Spreadsheets and BI tools do it with clicks. Code does it with a few lines, which helps when you repeat the same report every week.

Try it

The records below are made up. Each one is a visit to a tutoring program. The code counts visits per night and finds the average.

Run it. Which night is busier? Then add two more Tuesday visits to the list and run it again. Watch the numbers change.
Python 3
Ctrl+Enter to run | Tab to indent | Python runs in your browser via Pyodide

Notice there are no names in these records. A report about patterns rarely needs to know who anyone is.

Dashboards people use

A dashboard is a page of key numbers and charts that updates as new data comes in. Many dashboards get built and then ignored. The ones people use share a few traits:

  • Built for one role. A program lead needs this week's attendance. A board member needs the trend across the year.
  • Few numbers. Three to five key measures, not thirty.
  • Plain labels. "Students this week," not "ACTV_STU_7D."
  • Shows change. Compared with last week or last year, so people can tell good from bad.
  • Words, not only color. Say "down 10 percent," do not rely on red alone. Some people cannot tell red from green.

A national example

Dashboard check

0 of 6 done

Go deeper: Business Intelligence Tools: Guide to Seeing What's Really Happening in Your Business

Ask. Answer. Build together.