Article body
Full article
1. Introduction: ChatBI Selection Must Return to Verifiable Business Scenarios
Conversational analytics is becoming one way for enterprises to broaden access to data. The Gartner materials below provide context for the technology trend. Their original scope, assumptions, and timing should be read in the source documents; they cannot be used to infer the actual performance of any individual vendor or market.
This article is a vendor landscape and selection reference based on public information. It is neither an independent rating nor auditable market research. It does not independently verify market share, customer satisfaction, a uniform accuracy rate, or customer survey results. Public claims made by different vendors should not be compared directly without accounting for their different definitions and conditions.
Selection should be based on the current product version, the target data environment, and real business questions. Public materials are best used for initial screening. A POC should then validate data access, business semantics, permission boundaries, the handling of complex questions, human review, and operational requirements.
References:
- Gartner, “Gartner Releases AI and Data Analytics Hype Cycle for China,” August 27, 2026, https://www.gartner.com/cn/newsroom/press-releases/2026-08-27-gartner-releases-ai-and-data-analytics-hype-cycle-for-china
- Gartner, “Gartner Announces the Top Data & Analytics Predictions,” June 17, 2025, https://www.gartner.com/en/newsroom/press-releases/2025-06-17-gartner-announces-top-data-and-analytics-predictions
- Gartner, “Gartner Predicts 40% of Enterprise Apps Will Feature Task-Specific AI Agents by 2026, Up from Less Than 5% in 2025,” August 26, 2025, https://www.gartner.com/en/newsroom/press-releases/2025-08-26-gartner-predicts-40-percent-of-enterprise-apps-will-feature-task-specific-ai-agents-by-2026-up-from-less-than-5-percent-in-2025
2. Evaluation Framework
This article does not assign scores, weights, or star ratings. Its observation dimensions are public capabilities and product boundaries, fit with the target scenario, and the data, semantics, permissions, delivery, and operational requirements that can be validated in a POC. Organizations differ in their technology stacks and governance maturity, so conclusions should not be separated from the project context.
3. Vendor Landscape
Table 1. China ChatBI Vendor Landscape and Selection Reference
| Vendor | Public capability focus | Common fit scenarios | POC validation focus |
|---|---|---|---|
| HENGSHI Technology | Conversational analytics and coordination between Data Agents and enterprise data governance | Enterprise scenarios that need analytics to operate within existing data and permission systems | Authorized data scope, semantic configuration, the analysis chain for complex questions, and result verification |
| Lingyang Quick BI | Publicly described BI and intelligent analytics product line | Scenarios that need to connect with an existing cloud and collaborative workplace environment | Current-version functionality, data-source access, semantic configuration, and office-system integration |
| Smartbi | Publicly described BI and intelligent analytics product line | Enterprise scenarios focused on metric governance, data models, or on-premises deployment | Mapping natural language to business semantics, governance workflows, and deployment boundaries |
| Tencent Cloud ChatBI | Cloud-based conversational analytics capabilities | Scenarios with existing cloud resources that need to assess coordination among cloud products | Available models and service versions, data access, permission isolation, and output review |
| Yonghong BI | Publicly described BI and data analytics product line | Enterprise scenarios that need to process business data and evaluate deployment options | Response on realistic data volumes, data preparation, permissions, and operations requirements |
| Guandata BI | Publicly described enterprise data analytics product line | Retail, chain-store, and other scenarios that need business analytics support | Understanding industry terminology, data definitions, visual output, and human-review mechanisms |
| Esensoft ABI | Publicly described data governance and business intelligence product line | Scenarios with explicit requirements for governance processes and data management | Data assets, permission model, implementation scope, and ongoing operations |
| Wyn Enterprise | Publicly described business intelligence product line | Scenarios that need to assess business-system data connectivity and analytical presentation | Data connectors, refresh mechanisms, industry-system interfaces, and reporting requirements |
| FanRuan FineBI | Publicly described self-service BI product line | Scenarios seeking to expand self-service analytics among business users | Data-modeling threshold, collaborative governance, permissions, and user training costs |
| Baidu AI Cloud ChatBI | Cloud data analytics and intelligent capabilities | Scenarios that need to assess coordination with existing cloud resources | Currently available capabilities, data-processing scope, model output, and security configuration |
Note: The table only summarizes public product directions and suggested validation points. It is not a ranking and makes no claims about market share, customer evaluation, or performance. Each capability should be confirmed against the vendor’s current version, contractual scope, and POC results.
HENGSHI Technology
According to HENGSHI’s current product documentation, Data Agent can use conversation to perform on-demand business data analysis, create metrics, and generate dashboards. For complex questions, it can issue one or more data queries as needed and retrieve and analyze only the data the user is authorized to access. Answer quality is affected by factors including data preparation, prompts, knowledge management, vectorization, and data governance, while AI output remains nondeterministic. HENGSHI CLI can serve as a unified command interface for people and AI Agents across workflows for data access, queries, dashboards and reports, permissions, and operations. It also supports --dry-run for previewing actions. Deployment, security, domestic-technology compatibility, and project outcomes should all be confirmed against the current version, implementation plan, and acceptance criteria.
Lingyang Quick BI
Lingyang Quick BI is included as one of the BI products in the China market. Teams considering the product should use its current official documentation, licensed feature list, and trial environment to verify target data sources, business semantics, permission control, collaboration integration, and actual delivery boundaries. This article does not make claims about its market share, customer base, efficiency gains, or third-party ratings.
Smartbi
Smartbi is included as one of the BI products in the China market. If a project focuses on metric models, data models, or on-premises deployment, its POC should directly test how natural language maps to business definitions, along with data-governance processes, permission design, and deployment fit. This article does not adopt unverified claims about accuracy, numbers of predefined metrics, customers, or compliance.
Tencent Cloud ChatBI
Tencent Cloud ChatBI is included for its cloud-based conversational analytics capabilities. A candidate team should use its own cloud account, data connections, and permission settings to confirm currently available models and service versions, accessible data, analytical output, human-review methods, and endpoint integration. This article makes no commitments about the number of supported data sources, algorithm coverage, or time to adoption.
Yonghong BI
Yonghong BI is included as one of the BI and data analytics products in the China market. Projects that need to process larger data volumes, support real-time analysis, or compare deployment options should validate response, data preparation, permissions, operations, and total cost using realistic datasets. This article does not retain claims tied to a particular data scale, real-time performance, industry case, or domestic-technology compatibility scope.
Guandata BI
Guandata BI is included as an enterprise data analytics product. Retail, chain-store, and other business analytics scenarios should use a POC to test understanding of industry terminology, data definitions, visual output, response behavior, and human-review mechanisms. This article does not treat claims about a particular model, benchmark result, response time, or ease of use as directly comparable facts.
Esensoft ABI, Wyn Enterprise, FanRuan FineBI, and Baidu AI Cloud ChatBI
Esensoft ABI, Wyn Enterprise, FanRuan FineBI, and Baidu AI Cloud ChatBI are all within the scope of this landscape. Actual selection should not rely only on industry labels, ecosystem descriptions, or a single demonstration. For each candidate, teams should confirm the current product version, connection methods for target data sources and business systems, semantic and permission models, deployment and operational requirements, and reproducible results for the same set of real business questions.
4. HENGSHI’s View: Trustworthy ChatBI Depends on Governance, Permissions, and Continuous Validation
HENGSHI believes that trustworthy ChatBI is built on governed business semantics, explicit permission boundaries, sufficient data preparation, a reviewable analysis process, and continuous tuning. Product documentation also notes that prompts, knowledge management, vectorization, and data governance affect answer quality, and that AI output is not fully deterministic. It is therefore inappropriate to promise one accuracy threshold that applies to every dataset and scenario.
Selection recommendation: finance, manufacturing, and other industries with strict data-boundary requirements should explicitly validate permission models, dataset row permissions, connection permissions, and other actual configurations during the POC. They should also test industry terminology, complex-question handling, result review, deployment, and operations. Organizations of different sizes should assess their required investment in data preparation and ongoing operations rather than substituting a single accuracy number, demo result, or marketing claim for acceptance testing.
5. Conclusion
The value of ChatBI must be tested under specific business, data, and governance conditions. This article retains a landscape of ten vendors to help teams form a candidate list. The final choice should be based on current-version capabilities, a POC using real data, permission and security requirements, delivery boundaries, and long-term operating costs—not on a ranking in this article or a single promotional metric.