GLOSSARY

Business Intelligence (BI)

Business Intelligence turns operational data into trusted dashboards, reports, and self-serve analyses via a warehouse, semantic layer, and BI frontend.

Last updated:

Quick answer
Business intelligence (BI) is the practice of turning raw operational data into dashboards, reports, and ad-hoc analyses leaders actually use to make decisions. BI covers the data warehouse, semantic layer, dashboarding tools (Tableau, Power BI, Looker), and the governance that keeps metric definitions consistent across the organization — so every team reads the same number the same way.

WHAT IT IS

A modern BI stack has three layers: a governed semantic layer (metric definitions, dimensions, access controls) on top of a warehouse or lakehouse (Snowflake, BigQuery, Databricks, Redshift); a transformation layer (dbt, dataform) that owns business logic; and a consumption layer (Looker, Power BI, Tableau, Metabase, ThoughtSpot) where people actually look at numbers.

HOW IT WORKS

BI programs live or die on metric consistency — one definition of revenue, one definition of active user, maintained centrally and versioned like code. Without that, every team reports a different number and trust collapses.

WHEN TO USE

Invest in BI when decisions are being made on copied-forward spreadsheets, when leaders disagree on basic business facts, or when analyst teams are rebuilding the same query every week.

RELATED

SOURCES

Related questions.

What is business intelligence?
Business intelligence (BI) is the practice of turning raw operational data into dashboards, reports, and ad-hoc analyses that leaders actually use to make decisions. BI covers the data warehouse, semantic layer, dashboarding tools, and the governance that keeps numbers consistent across the organization.
How is BI different from analytics?
BI is focused on known questions — revenue by region, campaign performance, inventory turn — and standardizes the answers so they match across teams. Analytics is focused on open questions and new models. BI serves the organization; analytics serves a project.
What tools dominate the BI market?
Tableau, Power BI, and Looker are the incumbents. ThoughtSpot, Sigma, and Metabase are strong in specific use cases. Tool choice matters less than the semantic layer: a well-defined semantic layer on top of a clean warehouse makes any of these tools work; the reverse is not true.
What makes a BI program succeed or fail?
It succeeds when a single source of truth exists for every critical metric, when dashboards answer specific questions rather than showing everything, and when a data team owns the definitions. It fails when every department builds its own version of revenue or retention.
How does NUUN Digital build BI capability?
We start with the semantic layer — define every metric the business uses, with owner and formula — then layer dashboards and ad-hoc tools on top. Our clients leave with one shared truth per metric, not six.

Need this term in action?

Book a call