Looker visual update adds native charts and AI tools
Google Cloud says the November 2026 release brings KPI, Sankey, histogram and network charts, plus dashboard filters and an AI skill for custom visuals.

Key takeaways
- Looker's November 2026 update adds KPI, bullet, gauge, Sankey, histogram and network visuals.
- Dashboards receive updated styling and a preview filter control for individual tiles.
- A new looker-skills repository lets AI coding agents build custom charts against Looker's Visualization API.
- Existing Marketplace plugins will keep working, while related plugins are retired.
In this article
What changed

Google Cloud has announced a broad visual refresh for Looker, its enterprise business intelligence product. The update was described by Google Cloud Blog on 17 November 2026. It adds built-in chart types, dashboard controls and AI-assisted tools for custom visualisations. Google Cloud says the changes are meant to help business users interpret data more clearly. It says metrics will remain governed by Looker's semantic layer.
Looker is Google Cloud's business intelligence product for exploring and reporting on enterprise data. Google Cloud says the visual refresh follows earlier work on exploration. That work included drag-and-drop workspaces, simpler field pickers and conversational analytics. Those changes were intended to make it easier to start from a blank canvas. The November release focuses on how data is displayed and interpreted.
The update applies to dashboards, Explores and Looker's generative Data Apps. Google Cloud says it runs on Looker's governed semantic layer and uses context supplied by Knowledge Catalog.
New native charts
The release adds several built-in visual types. KPI visuals can show a core metric and its change from the previous period. They can also show a percentage trend and a directional status. Bullet charts and gauges can place actual results against targets, quotas or threshold bands in compact layouts. Google Cloud also adds compact number formatting, so large values can appear as $2.4M rather than $2,400,000.
Other charts are aimed at deeper analysis. Sankey diagrams can map multi-stage flows, including conversion paths, user drop-offs, budget allocations and material flows. Hierarchical tables let users drill into nested data trees, such as category rollups or departmental cost centres. Histograms can plot frequency distributions and reveal skew or outliers in numeric variables without table calculations. Network charts can show relationships between entities and help identify clusters or bottlenecks.
Dashboard design and filters
Google Cloud is also modernising the look of dashboards. The company says the update brings newer typography, refined spacing, cleaner table padding and upgraded palettes. These changes are meant to give dashboards a more polished appearance without requiring custom styling. A preview feature called filters on visualisations will allow viewers to change filters directly on individual tiles. The aim is to let people isolate specific cuts of data without refreshing the whole page.
AI-assisted custom charts
Looker is also introducing a custom visualisation skill for AI coding agents. The skill is in the open-source looker-skills repository on GitHub. Google Cloud says a user can describe a desired chart in plain language. An agent can then produce code for Looker's Visualization API. The agent can then refine the design through conversation and deploy the result.
The company says this can help with specialised visual needs. Examples include defect density maps for silicon wafers. They also include warehouse floor plans combined with shipping metrics and multi-step variance bridges for finance teams. It says team members do not need to know JavaScript. If preferred, the agent can use common JavaScript charting libraries or build the graphic directly in SVG.
Charts created this way still receive data from LookML models. Google Cloud says they keep Looker's usual drill menus, cross-filtering and tile controls. This means custom visuals can behave like built-in Looker content rather than isolated add-ons.
Plugins and Data Apps
Some of the new native charts began as plugins in the Looker Marketplace. Google Cloud says it will retire the related plugins because the features are now supported natively. Dashboards already using those Marketplace plugins will keep working. The company also says the same design approach and agent-led workflow supports Looker Data Apps, which is in preview.
Google Cloud says Data Apps can extend a single chart into a multi-step operational workflow. These apps can use the same updated type styles, colour palettes and governed LookML metrics. The company says this helps teams keep a consistent look and governed metrics. It applies whether they are adding a chart to a dashboard or building a standalone application. It says the new chart options and dashboard controls can be explored in a Looker environment now. Documentation is available for users who want more detail.
Sources
Spotted a mistake? See our corrections policy. Read how News by OnDevice works in our editorial policy.



