Looker put the model in git. Kater puts everything in git, in SQL.
LookML gave data teams one governed model, and that idea was right. But LookML is a language of its own, and the dashboards people use live in Looker’s database. In Kater, the model is SQL, and every query, dashboard and app is a file your agent can change.
You like Looker’s governed model, but not LookML or its price.
You want dashboards and apps in git, not only the model.
Your team builds with Claude Code, Codex or Cursor.
Your warehouse is Snowflake or DuckDB.
Stay with Looker if
You run on BigQuery or another warehouse that Kater does not connect to yet.
You embed Looker in your own product.
You are committed to Google Cloud, BigQuery and Gemini.
Your LookML project is large, stable and well staffed.
Why teams leave Looker
01
LookML is a second language.
Every new field or join waits for a LookML developer. Business users find Looker hard to use, so they queue behind the few people who know LookML.
“Someone with knowledge of Looker can understand it, but a person without any technical background finds it somewhat advanced or difficult to comprehend.”
In Kater
The model is Kater SQL: your warehouse’s own SQL with a few additions. Your agent already writes it, and anyone who reads SQL can review it.
02
Only the model is in git.
LookML lives in git. The dashboards and Looks that people build in the browser live in Looker’s database. An agent cannot search them, and nobody can review a change to them as a diff.
“UDD not part of git tracking is a big inconvenience for multi-instance deployments.”
In Kater
Kater ships every week. The changelog lists each release.
Kater vs Looker at a glance
Kater and Looker, feature by feature
Feature
Kater
Looker
Building with agents
Source of truth
Yes.Files in git: models, queries, charts, dashboards and apps. Every save is a commit.
In part.LookML is in git. Dashboards and Looks built in the browser live in Looker’s database.
Coding agents
Yes.Claude Code, Codex and Cursor edit the files through the Kater MCP, or in a local copy.
In part.The VS Code extension and Looker Skills support coding agents for LookML. Query and dashboard skills are planned.
Checks on changes
Yes.A write through the MCP compiles first and is refused on error. Every change has a diff, History and Undo.
Yes.The LookML validator, and CI on pull requests.
Answers in your own AI
Yes.People ask in Slack, Claude or ChatGPT. The agent sees only what that person can see.
Yes.Conversational Analytics in Looker and Gemini Enterprise. The managed MCP server is in preview.
Metrics and trust
Semantic layer
Yes.Kater SQL: your warehouse’s SQL with DIMENSION, MEASURE and EVAL. Defined once, read everywhere.
Yes.LookML, a modeling language of its own, with SQL inside.
Tests on numbers
Yes.Tests check a query’s rows, or compare a measure with SQL you trust. They run after each sync.
Yes.LookML data tests, run in the IDE or in CI.
Dashboards and apps
Build without code
Yes.Select fields, a chart and a place on the dashboard, all in the browser.
Yes.Business users build queries and dashboards in Explores.
Custom apps
Yes.HTML apps, reports and decks on live queries, built by your agent. Export to PDF, PowerPoint, Word or Excel.
In part.Extensions and embedded apps, built by developers.
Paginated reports
Yes.Report templates for PDF, PowerPoint and Word, on live queries.
Yes.Paginated reports are generally available.
Platform
Warehouses
In part.Snowflake and DuckDB today.
Yes.BigQuery, Snowflake, Redshift, Databricks, Postgres and many more.
Embedding
No.No embedding (for now). It is in development.
Yes.A mature embedding SDK.
Row-level access
Yes.Row-level security, access filters and grants, with access by folder and file.
Yes.Access filters and user attributes.
Where Looker is ahead
Looker does some things better today. Here is each one, and why teams still
select Kater.
Where Looker is aheadWhy teams still select Kater
More warehouses
Looker connects to BigQuery, Redshift, Databricks, Postgres and dozens more. Kater connects to Snowflake and DuckDB today.
Why teams still select Kater
Kater goes deep on each warehouse it supports. Queries run live, and the compiler checks each SQL dialect before a query runs. Have a warehouse we do not support? Request it.
Looker comes with Google’s scale, a long track record and a large partner network. A vendor that size also sets one roadmap for thousands of customers.
Why teams still select Kater
Kater is ready for enterprise: SOC 2 Type II, row-level security, access filters, grants and SSO. And you work with the people who build it. You decide how you pay, and you help decide what agentic BI becomes.
Looker sells a platform edition (Standard, Enterprise or Embed) plus Developer, Standard and Viewer users, on annual terms. Google lists no prices. The Standard edition lists at $60,000 a year on AWS Marketplace, before users. Conversational Analytics will be metered in data tokens after its promotional period.
Kater
A 30-day free trial with every feature, then a subscription quoted for your team. Kater does not charge per viewer or per dashboard. Kater runs no chat of its own, so answers run in the agent your team already uses.
Moving from Looker
LookML and Kater SQL describe the same things, so most of a Looker project maps one to one. Your agent reads your LookML files and writes the Kater SQL. You review each diff, and tests compare the numbers.
What each Looker object becomes in Kater
In Looker
In Kater
Connection
Connection, on Snowflake or DuckDB
View, with dimensions and measures
Kater dataset, with DIMENSION and MEASURE in SQL
Derived table
Custom SQL dataset
Explore and its joins
Semantic model, with RELATIONSHIP joins
Percent of total, or a measure in another context
A measure with EVAL and CLEAR, REMOVE or SET
Look or user-defined dashboard
Query and chart files, and a dashboard.yaml
Data test
Test, with expected rows or a same_as check
Access filter and user attribute
Access filter and grant, in Kater’s permissions
Ask your agent to migrate one dashboard. Kater rebuilds it as a private draft on your warehouse, and you compare the numbers before you publish.
Yes. Kater gives you a governed semantic layer, dashboards, apps and AI answers on one model. The model is written in SQL, and every layer is a file in git.
Can our agent convert LookML to Kater SQL?
Yes. Give your agent the LookML files. It writes Kater datasets and semantic models, and Kater compiles each file as it saves it. A same_as test then compares each measure with SQL you trust.
Does Kater work with BigQuery?
Not today. Kater connects to Snowflake and DuckDB. If you run on BigQuery, tell us.
Do we lose Looker’s governance?
No. The data team owns the shared layer. A field enters it only through a promotion that the data team approves, and every change has a diff and an undo.
Can we keep Looker and add Kater?
Yes, for as long as you want. Both read the same warehouse. Keep Looker for its dashboards, and add Kater to give your team’s agent one definition for each metric, your context and your access rules. When you want, rebuild one dashboard in Kater, compare the numbers with Looker, then move the next one.