← Back to work
Automated BI Reporting Pipeline
dbtQuickSightData EngineeringAutomation

Automated BI Reporting Pipeline

2025Client project

A client reporting pipeline built over more than 1TB of source data and 7.2M reporting rows, using dbt, Amazon QuickSight, and scheduled daily refreshes.

The problem

The client produced quarterly and yearly reports manually in Excel. They described the process as taking days and requiring substantial effort.

How it works

  1. 01, Input
    Source data

    More than 1TB of client data, kept separate from reporting workloads.

  2. 02, Tool or code
    dbt reporting layer

    Tested, documented models producing 7.2M reporting rows.

  3. 03, Output
    QuickSight dashboards

    Client-facing dashboards in Amazon QuickSight.

  4. 04, Tool or code
    Daily refresh

    Scheduled refreshes replace manual Excel reporting.

↺ Every day: back to step 02
InputTool or codeOutput

What I did

  • Independently built a dbt reporting layer over more than 1TB of source data and 7.2M reporting rows.
  • Kept reporting workloads separate from the source data and added tests and documentation to the models.
  • Built the client-facing dashboards in Amazon QuickSight and scheduled daily refreshes.

Results

  • Replaced a reporting process the client described as taking days with dashboards refreshed each day.
  • Delivered a repeatable reporting workflow for the team and its clients.
← All projects