Zoho Analytics adds agentic AI layer for business data

Zoho has turned Zoho Analytics into an agentic analytics platform.

The update adds a business intelligence layer designed to give AI agents access to company data and business context.

It expands Zoho Analytics beyond dashboards and reporting into a system intended to connect data analysis with agent-led tasks inside business software. The platform now combines data, semantic definitions, ontology models and workflow tools in a single product.

A central part of the update is the addition of native semantic and ontology layers. These are designed to help AI systems interpret business terms, metrics and relationships within an organisation, rather than rely solely on raw datasets.

According to Zoho, the semantic layer acts as a business glossary for AI, while the ontology layer maps business entities and the links between them. This is intended to help agents generate responses and recommendations more closely tied to a company's operations.

Tony Thomas, Chief Executive Officer of Zoho USA, said the update is intended to address a gap in how AI agents use business information.

"We are at the beginning of the next leap forward in data analytics, where a strong data foundation is the starting point for all agentic activity in business operations," said Tony Thomas, Chief Executive Officer of Zoho USA.

"Agents, however, cannot function only with data, but require a robust intelligence layer. Today's updates to Zoho Analytics provide that and more: one open and extensible trusted intelligence layer that gives every AI tool in your business the context, metrics, and governance to deliver answers users can rely on," Thomas said.

Agent tools

The company's Ask Zia assistant has also been reworked into what Zoho describes as a conversational super-agent for business intelligence. It now uses a multi-agent structure with six pre-built sub-agents covering data analysis, visualisation, data preparation, modelling, coding and SQL generation.

A central agent coordinates those functions through a single interface while retaining conversation context. Users can also build custom agents that use the same organisational data, semantic models and knowledge base as Ask Zia.

Support for the Model Context Protocol allows those agents to connect to external systems and tools, as well as functions inside Zoho Analytics. Agents can also trigger actions and workflows directly through a built-in workflow builder.

Governance focus

Zoho has put oversight and control at the centre of the update, reflecting broader business concerns about how AI tools access information and make decisions. The platform includes a unified data catalogue, lineage and impact analysis, access policy controls, and administrative tools to track agent usage and cost.

Administrators can review which agents ran, who used them, what data they accessed and how they answered queries. Sandbox and deployment management tools are also included so teams can test analytics assets and agents before moving them into production.

Zoho is also broadening the system's data reach. More than 30 new connectors have been added to a library of more than 500, including links to SAP, PayPal, Adobe Commerce, Twilio, Notion, Intercom, Supabase, Amazon DocumentDB and AlloyDB, alongside several Zoho products.

The semantic and ontology layers are built directly into Zoho Analytics rather than requiring a separate data platform or cloud data warehouse. Users can also create different semantic models for separate departments, such as finance and operations, and build or edit those models without code.

Customer view

Zoho said the platform is used by 100,000 businesses worldwide each day across cloud, on-premises, public cloud and hybrid deployments. It added that pricing remains unchanged, with Zoho LLM credits now included in each plan.

One customer said the changes fit its internal plans for AI and decision-making.

"We've had tremendous success with Zoho Analytics, using data to improve visibility, drive accountability, and support better business decisions across our organization. What excites us about this next evolution is the ability to combine data, business context, and AI into a single intelligence layer. These new capabilities align closely with our AI strategy, which is focused on turning information into action, helping our people make better decisions faster, improve productivity, and spend more time creating value for our customers," said David Fauser, Vice President of Sales, Marketing & Strategy at CIMCO Refrigeration.

Industry analyst Megha Kumar also pointed to the combination of context, governance and deployment options.

"With this release, Zoho is bringing together several elements needed for scalable, governed agentic analytics, including native semantic and ontology layers, governance and observability capabilities, and the option to deploy these agentic capabilities either embedded in applications or headless via MCP. Its stated commitment to openness and interoperability should help organisations leverage context-grounded insights," said Megha Kumar, Research Vice President of Analytics and AI at IDC.