Business Intelligence
Business Intelligence turns raw operational data into decisions a business owner can act on.
Outcome
This method produces a working dashboard and a short decision brief. The output is one metric per decision, wired to a live data source, so a stakeholder can act without waiting on an analyst.
Steps
- Name the decision the dashboard must serve, and the one metric that moves it.
- Trace the metric to its source data footprint and confirm the data is clean, fast, and open.
- Model the metric in a BI tool or a spreadsheet while the pattern is still early.
- Build the smallest view that answers the decision; resist adding charts nobody reads.
- Review with the decision owner and cut anything that does not change a choice.
Proof of Done
Done when a decision owner uses the dashboard to make a real call without asking for a custom pull, and the metric refreshes on its own.
Signals
- Check that each chart maps to a named decision, and drop orphan charts.
- Measure refresh latency and whether the numbers reconcile with the source of truth.
- Watch for dashboards that get built but never opened; that is the signal the decision was never real.
Context
Products
Spreadsheets
- Airtable
- Excel
- Google Sheets
Questions
Which engineering decision related to this topic has the highest switching cost once made — and how do you make it well with incomplete information?
- At what scale or complexity level does the right answer to this topic change significantly?
- How does the introduction of AI-native workflows change the conventional wisdom about this technology?
- Which anti-pattern in this area is most commonly introduced by developers who know enough to be dangerous but not enough to know what they don't know?
Changes my mind: if decision owners act faster from a raw AI query than from a maintained dashboard, the build-a-dashboard-first step here is wrong.
Next question: which single metric, if wrong, would cause the most costly business decision — and is it the one on the dashboard?