These new features empower data teams to easily ask smart, context-aware questions in plain language, while developers can now effortlessly integrate secure analytics agents across all their applications using unified API endpoints.
<div class="media-with-label__label">
Credit: Khanchit Khirisutchalual </div>
</figure>
</div>
</div>
</div>
</div>
<div id="remove_no_follow">
<div class="grid grid--cols-10@md grid--cols-8@lg article-column">
<div class="col-12 col-10@md col-6@lg col-start-3@lg">
<div class="article-column__content">Google just made its BigQuery data warehouse even smarter! They’ve introduced ‘Conversational Analytics,’ which means that now, anyone – from data experts to regular business users – can simply *talk* to their data, asking questions in everyday language. This is a game-changer for quickly analyzing data, especially for AI projects.
This cool new ‘agent’ is currently in its preview phase. You can find it tucked away under the ‘Conversations’ tab within BigQuery’s new Agents Hub. Getting started is easy – just point it to your data tables! Experts are saying it takes BigQuery’s existing ability to turn text into SQL queries to a whole new level.
“BigQuery already offers features like data canvas to make query generation and visual exploration easier. What changes with the agent is not the ability to ask questions in simple language, but the ability to carry a contextual conversation with the data over multiple steps, which can be defined as conversational analytics,” said Abhisekh Satapathy, principal analyst at Avasant.
Satapathy further elaborated, “Imagine the new agent actually *remembers* your previous questions, including all the details like datasets, filters, timeframes, and assumptions. It uses this memory to answer your follow-up questions much more intelligently. This means you can gradually fine-tune your analysis without having to begin anew every single time.”
He also highlighted a major benefit: this new capability significantly reduces the burden on developers. They won’t have to spend endless hours building dashboards or setting up predefined rules for every single question a data analyst or business user might come up with.
“Rather than encoding every scenario upfront, teams can let the agent interpret user intent dynamically, while still enforcing access controls, metric definitions, and governance rules already defined in BigQuery,” he said.
Easily Create and Deploy Custom Agents Using API Endpoints
Beyond just the built-in agent, Google has packed the Agent Hub with even more power! You’ll find tools there to help you create, roll out, and manage your very own custom agents. These can work across all your different applications and daily tasks, all controlled through straightforward API endpoints.
These tools, according to Satapathy, address three practical enterprise needs: “It reduces duplication of analytics logic across tools, ensures consistent definitions and policies across all analytics users, and centralizes access control and auditing rather than implementing them separately in each application.”
This clever reduction in duplicate work also gives developers a huge break! Satapathy explained that they won’t need to constantly rebuild the logic for understanding user questions, connecting them to the right data, enforcing security, or even explaining the outcomes.
What’s more, you can even deploy these custom agents through Looker, which already has its own handy conversational analytics feature built right in.
Constantly Getting Better: Text-to-SQL Improvements
For quite some time now, Google has been steadily enhancing BigQuery with natural language and SQL features. Their goal? To make SQL querying much simpler for both developers and data analysts.
Just this month, they gave us a sneak peek at a ‘Comments to SQL’ feature. This genius tool lets developers and data analysts jot down instructions in plain English right within their SQL comments, and then BigQuery Studio, powered by Gemini, will magically turn them into working queries.
Rewinding to last November, Google introduced three fantastic new AI-powered SQL functions: AI.IF, AI.CLASSIFY, and AI.SCORE. These are designed to drastically simplify complex, large-scale analytics for businesses, especially when dealing with messy, unstructured data. And back in August, BigQuery’s data engineering and data science agents also received some helpful updates.
Of course, Google isn’t alone in this race. Competitors like Snowflake and Databricks are also busy developing their own natural language to SQL features for their respective platforms.
To give you an idea, Databricks already has ‘AI Functions’ that let you use generative AI or large language model (LLM) inference directly from your SQL or Python code. Snowflake, on the other hand, offers handy functions like AI_PARSE_DOCUMENT, AISQL, and Cortex, which help with things like understanding documents, smart searching, and AI-powered analytics. Even other data warehouses, like Oracle’s Autonomous Data Warehouse, are getting in on the action, supporting AI processes right alongside traditional SQL.
