Zoho has launched a major architectural update for its Zoho Analytics platform, restructuring the software around four core pillars: trusted data and business context, agentic analytics and actions, intelligence everywhere, and governed AI. The central enhancement of this release is the introduction of native semantic and ontology layers built directly into the platform, positioning the software as a unified intelligence layer for enterprise users and artificial intelligence systems alike.
Understanding the Architectural Shift
The updated analytics platform addresses a critical challenge in modern enterprise data management by bridging the gap between raw data repositories and AI agents. According to company executives, the native semantic and ontology layers arrive fully integrated out of the box, removing the requirement for a separate standalone data platform or third-party cloud data warehouse. This design allows organizations to maintain their existing enterprise software stacks while adopting individual composable layers of the updated architecture independently.
The system relies on two distinct context layers positioned above raw data sources. The ontology layer maps canonical business entities and defines the structural relationships between them. Meanwhile, the semantic layer operates directly above the ontology to establish clear business definitions for entities and metrics, encompassing descriptions, synonyms, sample queries, metric calculations, and the corresponding SQL statements.
- Canonical business entity mapping through the native ontology layer
- Standardized metric definitions and business terms via the semantic layer
- Compatibility with the Open Semantic Interchange specification
Expanding Agentic Capabilities and Integration
Building upon its natural-language assistant originally introduced years ago, the updated platform transitions the assistant into a multi-agent orchestration system. The environment now features multiple pre-built sub-agents handling specific operational domains, including data visualization, data preparation, modeling, coding, data analysis, and action execution. Furthermore, a custom agent builder empowers enterprises to deploy bespoke agents grounded on specific semantic models or datasets, supported by knowledge-base files and external tools.
The intelligence everywhere initiative focuses on two distinct delivery models. First, embedded analytics allow independent software vendors to white-label the complete agentic feature set directly within their own applications. Second, headless business intelligence capabilities—facilitated through model context protocol servers, command-line interfaces, and dedicated plugins—enable external developers and third-party AI agents to query data, construct dashboards, and execute analytical actions from outside the platform environment entirely.
Governance, Pricing, and Infrastructure
To ensure oversight across complex deployments, the governed AI pillar provides a unified data catalog offering complete lineage and impact analysis spanning upstream data sources, processing pipelines, metrics, models, and active agents. This centralized view aims to maintain transparency as organizations scale their automated workflows.
Despite the significant functional additions, pricing remains unchanged, with bundled artificial intelligence credits included across plans starting at entry-level tiers. Organizations can utilize zero-copy connectors to pull metadata from external data warehouses while keeping raw data in place, highlighting Zoho’s ongoing emphasis on flexibility and composability within the enterprise software market.