cd /news/artificial-intelligence/introducing-new-explores-and-merge-q… · home topics artificial-intelligence article
[ARTICLE · art-29755] src=cloud.google.com ↗ pub= topic=artificial-intelligence verified=true sentiment=↑ positive

Introducing new Explores and Merge Queries in Looker

Google announced new AI-powered features for Looker Explore, including natural language querying and automated query generation using Gemini models, alongside a redesigned Merge Query workflow. The updates aim to help enterprise users uncover data insights without technical expertise.

read3 min views1 publishedJun 16, 2026

A key goal for many enterprises in the AI era is to empower their employees to uncover actionable data insights on their own. To help, we are evolving Looker Explore with a streamlined interface and integrated AI, so every usey can confidently turn data into a clear path to action.

At the heart of the new Explore release is a suite of AI capabilities that guides users from their very first click with new insight and expression assistants.

We are virtually eliminating the cold start from an empty canvas. If the data modeler hasn't built predefined Quick Starts, Looker automatically generates a query for the user, tapping into Google’s latest Gemini models to generate ad hoc Quick Starts that can help users dive deep into the data, beyond visible fields, and surface potential questions the data can tackle.

Users can now prompt Looker Explores in natural language to modify data tables and visualizations. The Insight Assistant uses the Conversational Analytics API to identify relevant fields, apply filters, sort data, and construct the data table. We expect this feature to be a significant time-saver that can provide a rapid starting point for complex analysis.

Users can also use natural language to describe their custom calculation, and Looker will automatically fill in the appropriate syntax, without having to learn Looker Expression (Lexp) syntax. Users can also re-prompt the assistant to iterate on custom field expressions.

If a user-generated description does not exist for an Explore, Looker will provide an AI-generated summary, to help data analysts rapidly gain familiarity. In addition to these new assistants, we’ve updated the Looker user interface to be more modern and polished. There, you’ll find:

A customizable workspace: The new interface features a resizable field picker pane, with more easily readable long field names.

Data table contextual menus: Looker now offers powerful functionality right in the data table. Users can access quick menus on columns to switch data granularities, apply filters like 'IS NOT BLANK' or 'IS NOT NULL', and instantly add complex table calculations like '% of column' or 'running total'.

**Visual pivots: **Users will soon be able to drag and drop fields into a panel to pivot data into columns, rows, and aggregated values.

Connect data with redesigned merge queries

In addition, we redesigned Looker Explore’s Merge Query workflow with a unified, in-window architecture that includes:

A dynamic three-panel interface: The new design maintains context beautifully by displaying three simultaneous panels: a "configure joins" list on the left, a dynamic field picker in the middle, and your data preview/visualization on the right. You can edit a source query without losing the context of the overarching join configuration.

Smart join suggestions: The new panel automatically suggests optimal join fields, such as state and month, and shows the combined fields.

Instant query linking: If you have an existing query you want to use, you can paste a prebuilt query URL to start a join.

Expanded row limits: We've increased the default row limit for non-BigQuery sources to 50,000 rows.

By pairing conversational AI with a dramatically simplified user interface, Looker’s new Explore experience gives your business users the tools they need to investigate their data with confidence. Reach out to your Looker administrator today to enable this feature. For more information, click here for detailed documentation.

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @google 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/introducing-new-expl…] indexed:0 read:3min 2026-06-16 ·