Zoho is launching a new version of its analytics platform that has capabilities designed to connect AI agents with the business logic they require to be accurate, an upgrade that brings it in line with other data and analytics platforms.
To provide AI tools with relevant information, Zoho's latest update includes a semantic layer where customers can define data to make it consistent and discoverable, an ontology layer to show how business assets relate to one another, and over 30 new and updated connectors to data sources such as SAP, PayPal, Amazon DocumentDB and Google AlloyDB.
In addition, Zoho Analytics now includes a multi-agent version of its conversational interface Ask Zia, , headless BI capabilities that enable users to build and embed AI and analytics tools in any application, a data catalog to collect and govern data assets, AI observability capabilities and a sandbox environment for agent development.
Ensuring that agents have proper context -- relevant data, metrics and business logic -- has been the dominant trend in data and analytics this year because, even as model memory and reasoning improves, agents that lack situational awareness will never make it beyond the pilot stage or fail in production.
"The addition is the business-context layer is an important one for Zoho users," Michael Ni, an analyst at Constellation Research, told TechTarget. "Zoho is bringing ontology and semantic modeling together with its analytical agents, so users can work from consistent definitions. ... That matters as analytics moves beyond dashboards to support AI-driven applications and automated workflows.
While the new features are valuable additions for Zoho customers, many of the capabilities included in the vendor's latest update are commonly found in the data and AI platforms provided by competing vendors from hyperscalers AWS and Microsoft through platform providers Databricks and Snowflake to specialists such as Domo and Tableau.
However, Zoho is distinguishing itself from some of its peers by integrating the features within one platform rather than adding them in separate environments.
"With this release, Zoho is catching up on some individual AI features, but for customers and prospects, that is not the whole competitive story," Ni said. "Analytics, data management, business context and automation are converging, and vendors are competing to deliver more of that full stack. Zoho's advantage is its integrated approach and its established business application ecosystem."
Context key for agents
Despite enterprises continuing to increase their investments in AI development, far more projects are still failing than delivering a valuable return on investment. Problems with data, including discovering and delivering relevant data to AI tools, are a primary reason the high failure rate persists.
In response, a litany of vendors, including the hyperscalers and platform vendors, have made connecting agents with context a focal point of their product development. Zoho's latest update marks the vendor's foray into the fray.
"Analytics is entering into a new agentic era," Clarence Rozario, Zoho's director of product management," said during a virtual press conference on Oct. 22. "But is the architecture going to be just plugging in agents to help develop insights? We feel that's not the right way."
Instead, he noted that Zoho sees AI-powered insight generation flowing from a strong data foundation, through business context and agents that understand what is required of them based on that context, to decisions and actions.
Each of the new capabilities addresses part of that flow, such as the data catalog and connectors to strengthen the foundation, the semantic and ontology layers to provide context, and Ask Zia and headless BI capabilities to enable insight generation.
"It's a significant [update], mostly because the business context comes in the box," he told TechTarget. "A lot of Zoho customers are smaller companies without a data warehouse or a data team. They can now give AI agents their business definitions and how their business fits together without needing either one."
Mike Leone
Analyst, Moor Strategy & Insights
Feeding AI tools with context has been a struggle for many organizations. Therefore, including tools that ease the burden of enabling connecting agents with relevant data makes Zoho's platform update important for users, according to Mike Leone, an analyst at Moor Strategy & Insights.
"It's a significant [update], mostly because the business context comes in the box," he told TechTarget. "A lot of Zoho customers are smaller companies without a data warehouse or a data team. They can now give AI agents their business definitions and how their business fits together without needing either one."
In addition, like Ni, Leone noted that integrating capabilities such as a semantic layer, ontology layer and data catalog in one unified platform helps Zoho stand apart from vendors providing similar features.
"Most of these pieces are showing up across the market, so on features alone, Zoho is keeping in line," he said. "What sets it apart is who it's for. Zoho puts the context layers, the agents and the governance in one product, so a midsize company gets this without first buying a separate data platform."
Zoto puts the pieces in place for agentic success
It can't yet be determined whether Zoho's new capabilities perform as intended in production environments and truly enable customers to connect agents with context. However, Zoho at least has put the right features in place to do so, according to Ni.
"Zoho has put the right building blocks together," he said, noting that business context grounds the agents while tools such as Model Context Protocol and command-line interfaces allow users to enable their agents to operate beyond the confines of Zoho's platform.
"The challenge comes when multiple systems and agents depend on the same context," Ni continued. "Zoho will need to manage changes, permissions and execution consistently across those systems and data boundaries."
Leone likewise noted that Zoho's analytics platform update appears appropriately built to enable the vendor's users to connect agents with the context they need to perform in production. However, he pointed out that what is not yet is clear is how much work has to go into building semantic and ontology layers before they can be used to inform agents.
"What Zoho still has to show is how much work it takes for a customer to model their business in the ontology layer," he said. "The agents' answers are only as good as that model."
Eric Avidon is a senior news writer for Informa TechTarget and a journalist with more than three decades of experience. He covers analytics and data management.
