Looker
Looker is a semantic layer first, BI tool second. You define metrics, dimensions, and joins in LookML (a YAML-like DSL), then business users explore data through a drag-and-drop interface. Every query hits your warehouse live; there are no extracts. Google acquired it in 2020 and has been weaving it into BigQuery and GCP. The trade-off: LookML has a learning curve, and you need someone technical to own the model.
Official site ↗Why it ranks here
- Governed metrics layer means every dashboard uses the same definition of ARR, pipeline, or churn. No metric drift.
- Live queries on the warehouse, so dashboards always reflect the latest data without extract refreshes.
- Version-controlled LookML models let you treat analytics like infrastructure: code review, CI/CD, rollback.
Where it sits
- Stronger than Tableau for metric governance; weaker for visual exploration by non-technical users.
- More opinionated than Power BI or Omni: you must invest in the semantic layer upfront.
- Better than Mode for end-user self-service; Mode is SQL-first and stays technical.
How GTM Operations teams use it
- RevOps defines LookML models for pipeline, bookings, and customer health; sales ops and marketing then build dashboards on top.
- Weekly pipeline reviews using live Looker dashboards that drill into Salesforce opportunity fields.
- Marketing attribution dashboards that join campaigns, leads, and closed revenue with consistent date logic.
- Executive scorecards with governed definitions of ARR, NRR, and logo count.
In-depth notes
- Pricing is per-viewer or per-developer. Expect thousands per month minimum; scales with user count and query volume.
- LookML is a DSL, not SQL. Budget time to learn it or hire someone who knows it. The payoff is metric consistency across the org.
- Looker does not store data; it generates SQL and hits your warehouse. Performance depends on your warehouse tuning (materialized views, clustering).
- Google is migrating Looker to Looker Studio for some use cases; watch the roadmap if you are betting on this long-term.
- Embedded analytics: Looker has strong APIs for embedding dashboards into apps. If you want white-label reporting, this is a good path.
Best for
Teams that want a single source of truth for GTM metrics, or orgs with technical RevOps who can own the LookML layer.
Avoid if
You need quick visual prototyping with no upfront modeling (use Tableau), or you are not on a modern cloud warehouse.
The rest of BI & analytics
Tableau Leader Visual analytics platform with drag-and-drop interface; Salesforce-owned market leader for interactive dashboards. #3 
Power BI Leader Microsoft's BI platform with deep Office and Teams integration; cost-effective for Microsoft shops with enterprise agreements. #4 
Omni Strong Modern warehouse-native BI with spreadsheet, SQL, and semantic model workflows; live queries with no extracts or cubes. #5 
Sigma Strong Warehouse-native BI with a spreadsheet interface; live queries on Snowflake, BigQuery, or Databricks with familiar Excel-like UX. #6 
Mode Strong SQL-first analytics platform with notebooks for analysis and reporting; popular with technical RevOps and data teams. #7 
ThoughtSpot Strong Search-driven analytics with natural language queries; AI-powered insights for business users who do not want to build dashboards. #8 
Hex Situational Notebooks for Python and SQL hybrid analysis; build interactive apps from notebooks with code and no-code cells. #9 
Domo Situational All-in-one platform with ETL, cloud data warehouse, and BI; higher total cost but unified vendor for orgs without a data stack. #10 
Qlik Situational Associative analytics engine for exploring complex data relationships; in-memory architecture with a steeper learning curve. #11 
Metabase Situational Open-source BI with simple setup and no-code query builder; good for small teams that need dashboards without enterprise complexity. #12 
Lightdash Niche dbt-native BI tool; metrics and dimensions defined in dbt YAML, dashboards built on transformed warehouse tables. #13 
Preset Niche Managed Apache Superset; open-source BI with commercial support, hosting, and enterprise features. #14 
Redash Niche Open-source SQL query and visualization tool; simple dashboards for technical teams with minimal setup. Learning guide
Setup
Connect Looker to your warehouse, define a LookML project in Git, and model your first view with dimensions and measures. Deploy the project and build an Explore for users.
First thing to build
Model opportunities as a view with fields like stage, close date, amount, and owner. Define a pipeline measure as sum of amount where stage is open. Users then drag fields to build dashboards.
What actually matters
- Version control LookML in Git with code review for metric changes
- Use derived tables or persistent derived tables for complex joins or aggregations
- Set access grants and model-level permissions for row-level security
- Cache expensive queries with datagroups to reduce warehouse costs
Watch out
LookML changes affect every dashboard using that model, so test in dev before deploying to production.