Analytics and Dashboards on Databricks
Turning tables into answers people act on — the modelling underneath, the query, the chart, and the honest caveat.
Analytics fails more often from the wrong question than from the wrong chart. This course keeps returning to the question — what someone will do differently once they see the answer.
Everything is taught on Databricks, which is the platform this lab builds on and partners with.
Every lesson is free and needs no sign-up.
Syllabus
Starting from the question
Turning "how are we doing" into something a query can answer, before opening a tool.
Modelling for analysis
Facts, dimensions, grain, and why the wrong grain makes every number quietly wrong.
Querying with SQL warehouses
Running analytical SQL on Databricks, and what a warehouse does differently from a cluster.
Metrics that survive contact
Defining a metric once, and the arguments that follow when two teams define it twice.
Dashboards
Building one people read, and deleting the charts nobody does.
Being honest about the number
Confidence intervals, small samples, seasonality, and saying "we cannot tell yet".