Make the question visible.
A useful dashboard starts with a conversation: what decision needs to be made, which measures describe it, and what comparison puts it in context? That shapes the report before we choose a chart.
Our work includes organization-wide revenue and profit reporting in Amazon QuickSight. Clear labels, meaningful comparisons, and thoughtful visual design help colleagues move from the broader picture into the detail relevant to their work.
In practice
From question to maintained report.
Define the measure
Agree on what a number means, the period it covers, and the question it should help answer.
Design for the reader
Choose views that make trends and comparisons understandable, with useful filters and clear context.
Review and publish
Check the report and its definitions, then make a reviewed version available to colleagues.
Keep the context
Maintain documentation and change history so the next person can understand how the report works.
BI Ops: the work behind the chart.
Our BI Ops tooling records report definitions, generates documentation, and tracks dependencies and version history. That makes it easier to find where a measure is defined, understand a change, and maintain a report over time.
We’re also making that context useful to AI assistants and coding agents. Searchable documentation gives them a starting point in the reporting that already exists, helping the team investigate questions and prepare changes for review. People remain responsible for what is published.