Looker
by Google Cloud · Embedded BI
Governed semantic-layer BI and embedded dashboards over warehoused ERP data.
- Works with
- NetSuite, SAP, Oracle
- Deployment
- Cloud
- Company size
- Mid-market, Enterprise
- Pricing
- Quote-based
- Founded
- 2012
- Headquarters
- Santa Cruz, California, United States (now part of Google Cloud)
Overview
Looker is a business intelligence and embedded analytics platform from Google Cloud, distinguished by its LookML modeling language, which defines a centralized, governed semantic layer over a database or cloud data warehouse. Metrics and business logic are defined once in LookML and reused consistently across dashboards, explores, embedded experiences, and AI agents, providing a single source of truth.
Looker takes an in-database approach: rather than extracting and caching data, it generates SQL and queries the underlying warehouse (such as BigQuery, Snowflake, or Redshift) directly, which suits large-scale, freshly-updated data including warehoused ERP datasets. The platform is heavily oriented toward embedded analytics, offering iframe-based embedding and SDKs so developers can deliver governed analytics inside their applications.
Now part of Google Cloud, Looker integrates Gemini-powered Conversational Analytics for natural-language questions grounded in the LookML semantic model, plus Looker Studio for free-form visualization. Its open semantic layer is also positioned as a trusted grounding source for AI agents.
Screenshots & demo
Demo video from the vendor's YouTube channel.
Features & capabilities
Semantic Modeling (LookML)
Governed, reusable data definitions.
- LookML language to define dimensions, measures, and business logic
- Centralized metrics as a single source of truth
- Version control and Git integration for models
- Reusable explores for self-service querying
- Open semantic layer consumable by APIs and AI agents
In-Database Querying
Query the warehouse directly.
- SQL generation against the source warehouse (no extract required)
- Native dialects for BigQuery, Snowflake, Redshift, and 50+ databases
- Always-fresh data with in-database execution
- Aggregate awareness and caching policies for performance
- Persistent derived tables (PDTs)
Embedded Analytics
Embed governed analytics into apps.
- Private and public embedding via signed/SSO iframes
- Embed SDK and full REST API
- Row-level data controls for multi-tenant embedding
- Embedded Conversational Analytics
- Dedicated Embed edition with API call allowances
AI & Conversational Analytics
Gemini-powered natural language.
- Conversational Analytics grounded in the LookML semantic model
- Gemini-powered natural-language querying
- Agentic insights and automated decisions
- Low-code conversational embedding
Dashboards & Visualization
Content authoring and delivery.
- Looker dashboards and explores
- Looker Studio for free-form visualization
- Custom visualizations and the Marketplace
- Scheduled delivery and alerting
- Data actions and write-back integrations
Governance & Administration
Enterprise control.
- Centralized permissions, model access, and content folders
- Row- and column-level access controls
- Usage and performance monitoring
- Google Cloud IAM and SSO integration
Common use cases
- Defining governed, consistent metrics over warehoused ERP and finance data
- Embedding multi-tenant analytics into SaaS applications
- Self-service exploration with explores backed by a semantic model
- Natural-language analytics grounded in trusted definitions
- Always-fresh operational dashboards querying the warehouse directly
- Serving a trusted semantic layer to downstream AI agents and APIs
- Data-driven workflows with write-back and data actions
Strengths & considerations
Strengths
- LookML semantic layer enforces consistent, governed metric definitions
- In-database architecture queries the warehouse directly (no extracts)
- Strong embedded analytics and developer API/SDK story
- Native Google Cloud and BigQuery integration with Gemini AI
Considerations
- Requires a performant cloud data warehouse to query against
- LookML modeling is developer-oriented and adds an upfront learning/setup cost
- Total cost (licensing plus warehouse compute) can be high for mid-market deployments
ERP integrations
Pricing
Looker uses platform pricing (per instance, with editions such as Standard, Enterprise, and Embed) plus per-user pricing; quoted by sales. Warehouse compute (e.g., BigQuery) is billed separately. Conversational Analytics usage is metered in data tokens. Get an independent shortlist with pricing guidance below.
Technical & security
- Hosting
- SaaS on Google Cloud (multi-tenant); customer-hosted/Google-hosted instance options
- Data residency
- US, EU, UK, APAC
- Compliance
- SOC 2, SOC 3, ISO 27001, HIPAA, GDPR
- Mobile app
- Yes
- Languages
- English, and others
About the vendor
- Founded
- 2012
- Headquarters
- Santa Cruz, California, United States (now part of Google Cloud)
- Employees
- Part of Google Cloud / Alphabet
- Ownership
- Public (subsidiary of Alphabet/Google, NASDAQ: GOOGL)
Alternatives to Looker in Embedded BI
Looker — frequently asked questions
What is LookML?
LookML is Looker's modeling language for defining a centralized, governed semantic layer; metrics and logic are defined once and reused consistently across dashboards, explores, embeds, and AI.
Does Looker store its own copy of data?
Generally no. Looker generates SQL and queries the underlying data warehouse directly, so data stays fresh and lives in the warehouse, with caching and PDTs for performance.
How does Looker handle natural-language questions?
Looker's Conversational Analytics uses Gemini and is grounded in the LookML semantic model to return governed, consistent answers.
Evaluating Embedded BI?
Tell us your ERP and requirements and we'll send an independent shortlist — including Looker and the best-fit alternatives — with honest pros and cons.