Actian AI Analyst
Also known as: Wobby, wobby.ai
Wobby, now sold as Actian AI Analyst, answers plain language questions over a company's data warehouse through a governed semantic layer, with a Steward agent that maintains it, scheduled insights and answers in Slack and Microsoft Teams.
Wobby, now sold as Actian AI Analyst, is an AI data analyst for teams whose data sits in a warehouse. Actian, the data and AI division of HCLSoftware, took over the product, and wobby.ai now redirects to Actian's AI Analyst pages. Business users ask questions in plain language in a web app or in Slack and Microsoft Teams, and the analyst queries Snowflake, BigQuery, Databricks, Redshift, Microsoft Fabric, SQL Server, PostgreSQL, ClickHouse and other sources through a governed semantic layer.
The semantic layer holds models, dimensions, measures, metrics, relationships, filters and a glossary, and a Steward agent helps build it. Steward also periodically reviews where the analysts struggled and what users asked, and proposes improvement plans that run only after an admin approves them. In Plan Mode the analyst sets out the steps it intends and waits for approval before running them, and Scheduled Insights rerun a prompt on a cron schedule and raise an inbox item or an email when a finding is worth surfacing.
Traces show each conversation with its reasoning steps, and audit logs record which user or agent took each action, with field level changes. Admins and users have separate roles, access to each analyst can be limited to named users, and a Public API with bearer keys is available on the Scale and Enterprise plans. Data is stored on Google Cloud in the European Union, and Actian says it uses models from Microsoft Azure, Anthropic and Google Cloud depending on the task. Plans start at $499 a month, with a 14 day free trial.
Vendor details
Canonical URL
https://www.actian.com/ai-analyst/
Category
Data analyst agent
Company status
first party product
Use cases & customers
In practice
A data team uses the Steward Agent to define metrics and relationships once, then lets business users ask questions in Slack that always resolve against the same governed semantic layer.
An analyst asks a deep question about why revenue fell in a region, and the agent reasons across many tables, then returns a report showing every join and calculation behind the answer.
A finance team sets Actian AI Analyst to rerun its key metric questions on a schedule, so an unexpected anomaly arrives as an inbox item and an email before anyone thinks to run the query.
Agentic Index coverage score
9.5 / 14 capabilities · 68%
| Integrations & Tool Calling | Partial |
|---|---|
|
Actian AI Analyst, formerly Wobby, connects to Snowflake, BigQuery, Databricks, Redshift, Azure Synapse, Microsoft Fabric, SQL Server, PostgreSQL, MySQL, MariaDB, ClickHouse, MotherDuck, Actian's own databases and CSV files, runs the queries it writes, and posts answers and alerts to Slack and Microsoft Teams. It takes no action in the connected systems and has no custom tools. Sourcedocs.actian.com/ai-analyst/settings/integrations.htmlread 2026-09-28 |
|
| Workflow Orchestration | Partial |
|
The analyst runs multi step analyses and in Plan Mode lays out the steps it intends before running them, and reports and scheduled insights rerun a prompt. It has no branching, no conditions, no multiple cooperating agents and no flow the customer composes. The sequence is the product's own. Sourcedocs.actian.com/ai-analyst/agent/working-with-agents/plan-mode.htmlread 2026-09-28 |
|
| Knowledge Grounding & RAG | Full |
|
Answers are grounded in a governed semantic layer of models, dimensions, measures, metrics, relationships, pre defined filters and a glossary, compiled to SQL through Semantic SQL, which the Steward agent helps build and admins edit in Studio, and which can be exported and imported in the Open Semantic Interchange format. Sourcedocs.actian.com/ai-analyst/semantic-layer/overview.htmlread 2026-09-28 |
|
| Human Oversight & Guardrails | Full |
|
Steward's proposed plans run only after an admin approves them, in Plan Mode the analyst waits for the user's approval of its plan before running the analysis, and admins restrict which models, dimensions and measures each analyst may use. Sourcedocs.actian.com/ai-analyst/steward-ai-agent/steward-inbox.htmlread 2026-09-28 |
|
| Security, Identity & Governance | Partial |
|
There are two roles (Admin with Studio access, User with Explorer only), per analyst restriction to named users, two factor authentication, session management and audit logs, and no SSO. Actian's trust center lists SOC 2 Type II for the Actian Data Platform and Data Intelligence Platform and ISO/IEC 27001:2022 for the Data Intelligence Platform and Actian Corporation without naming AI Analyst in scope, so neither certification is confirmed to cover this product. Sourceactian.com/company/trust-centerread 2026-09-28 |
|
| Observability & Auditability | Full |
|
Traces record each conversation with every response and its reasoning steps, charts and tables, audit logs record which user or agent initiated and executed each create, update, delete, execute, download and view with field level diffs, both export as CSV, and every answer shows its joins, filters and metric calculations. Sourcedocs.actian.com/ai-analyst/ai-analysts/monitoring/traces.htmlread 2026-09-28 |
|
| Memory & State Persistence | Partial |
|
Conversation history is kept and follow up questions keep their context within a thread. The standing instructions and saved prompts admins write for each analyst are instructions, not memory, and the agent keeps no memory store it writes and reads across sessions. Sourcedocs.actian.com/ai-analyst/governance/how-ai-analyst-handles-your-data.htmlread 2026-09-28 |
|
| Deployment & Data Residency | Partial |
|
Data is stored in PostgreSQL on Google Cloud Platform in the European Union with no region choice, and the Enterprise plan mentions custom deployment requirements without naming a region, customer cloud or on premises option. Zero copy means queries run in place, which is not a deployment mode. Sourceactian.com/ai-analyst/pricingread 2026-09-28 |
|
| Prebuilt Agents, Templates & Packs | Partial |
|
Admins configure each analyst with instructions, saved prompts and pre configured suggested questions, and users build and share reports, but no packaged analysts, templates or catalog are offered. Sourcedocs.actian.com/ai-analyst/ai-analysts/creating-an-agent/suggestions.htmlread 2026-09-28 |
|
| Triggers & Channel Coverage | Full |
|
Scheduled Insights run a prompt automatically on a cron schedule in a chosen timezone, without anyone opening the app, and raise an inbox item and an email when the analyst judges a finding worth surfacing. Analysts also answer in Slack and Microsoft Teams channels and the web app. Sourcedocs.actian.com/ai-analyst/agent/working-with-agents/scheduled-insights.htmlread 2026-09-28 |
|
| Model Flexibility & Routing | Partial |
|
Actian says AI Analyst uses models from Microsoft Azure, Anthropic and Google Cloud, with the provider depending on the task. The routing is Actian's, and customers and admins have no choice of model. Sourcedocs.actian.com/ai-analyst/governance/how-ai-analyst-handles-your-data.htmlread 2026-09-28 |
|
| APIs, SDKs & MCP Extensibility | Full |
|
A Public API authenticated with admin created bearer keys (sk_ secrets with optional expiry) has endpoints under app.wobby.ai/api/public/v1 such as the semantic layer, and the Scale and Enterprise plans include the Actian AI Analyst API. The full endpoint list is on the API reference page. Sourcedocs.actian.com/ai-analyst/public-api/api-keys.htmlread 2026-09-28 |
|
| Testing, Debugging & Optimization | Full |
|
The Steward agent periodically reviews where the analysts struggled, what users asked and how well the semantic layer served those questions, and proposes improvement plans. That is an optimization loop on the deployed analysts' behavior after deployment. There are no test sets or scored cases, and the Analytics page counts usage, not quality. Sourcedocs.actian.com/ai-analyst/steward-ai-agent/steward-inbox.htmlread 2026-09-28 |
|
| Browser & Computer Use | Not documented |
|
AI Analyst works through warehouse connections, the semantic layer and messaging apps, with no browser, desktop or computer control. Sourcedocs.actian.com/ai-analyst/index.htmlread 2026-09-28 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Product updates
Self hosted deployment is now available for Actian AI Analyst, formerly Wobby, in a private cloud or on premises, where until now it ran only as managed SaaS. A bring your own LLM option connects the analyst to a customer's own OpenAI or Anthropic account, or to models reached through AWS Bedrock, Google Vertex AI, Azure AI or an enterprise AI gateway. The self hosted version keeps the same governed semantic layer and visible answer logic as the cloud one.
Bears on: Deployment / data residency
View sourceActian expanded AI Analyst (formerly Wobby) with a Data Steward Agent that synchronizes business definitions and enforces data policies. The update also introduces Data Observability Agents to monitor pipeline health and a Model Context Protocol (MCP) server, allowing third-party AI agents to query governed business logic via open standards.
Bears on: Security / enterprise
View sourcePricing
Starter from $499 per month, or $5,950 billed annually. A 14 day free trial is offered.
Per plan per month, sized by message allowance, users, analysts and tables.
What is public
Starter, Growth and Scale prices are published with their messages, users, analysts and tables. Enterprise is custom.
Overage / add-ons
No automatic overage charges. When the monthly message allowance runs out, the account is prompted to upgrade. A report create or update uses 10 messages.
Sales call required
Mixed (some tiers require a call)
Free / trial
14 day free trial, no credit card
Lowest paid plan
Starter at $499 per month ($5,950 billed annually), with 200 messages a month
Cancellation / refund
Cancel at any time. The plan stays active until the end of the current billing period.
Related vendors
- AnswerRocket — Enterprise AI agent platform for analytics built around Max, which…
- AskEarth — Geospatial intelligence agent (GIA) that answers plain language…
- AskYourDatabase — AskYourDatabase is a conversational data analyst that turns plain…
- Athenic AI — Athenic AI is a data analyst agent that connects databases,…
- BlazeSQL — BlazeSQL is an AI data analyst that turns plain English into SQL…
- Brewit — Brewit is a conversational data analyst that writes SQL and…
Alternatives to Actian AI Analyst
The closest documented capability profiles to Actian AI Analyst among data analyst agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- BlazeSQL9.0 / 14Fuller documented coverage on Integrations & Tool Calling and Deployment & Data Residency
- Opensee10.0 / 14Fuller documented coverage on Integrations & Tool Calling and Deployment & Data Residency
- Querio9.0 / 14Fuller documented coverage on Security, Identity & Governance and Deployment & Data Residency
- Brewit6.5 / 14A lighter documented profile than Actian AI Analyst
- Datafold11.5 / 14Fuller documented coverage on Integrations & Tool Calling and Security, Identity & Governance
- Hex12.5 / 14Fuller documented coverage on Integrations & Tool Calling and Workflow Orchestration
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded