TextQL

TextQL is an AI-powered data analytics and automation platform designed to give organizations a natural-language interface for working with complex business data. Its main product, Ana, functions as an AI data analyst that can connect to data warehouses, databases, business-intelligence platforms, semantic layers, catalogs, and SaaS applications, allowing users to ask questions in plain English and receive answers, charts, reports, and analytical explanations.

Unlike a simple text-to-SQL chatbot, TextQL is designed to operate across an organization’s existing data stack. Ana can inspect an organization’s data sources, understand business definitions and relationships through TextQL’s ontology, select the appropriate sources, perform analysis, and return a traceable result. This allows business users to ask analytical questions without needing to know the underlying database structure or SQL dialect.

A central component of TextQL is its Ontology. TextQL describes the ontology as a version-controlled representation of an organization’s business knowledge, including metrics, definitions, relationships, permissions, methodologies, and supporting documentation. The ontology is designed to provide context to AI agents so they understand what terms such as revenue, active customer, pipeline, or retention mean within a specific organization.

The ontology can be built from existing systems rather than requiring an organization to move all of its data into a new platform. TextQL says Ana can read schemas, relationships, queries, underlying data, business logic, dashboards, and documentation and reconcile this information into a shared model. Changes to the ontology can be reviewed, approved, versioned, and synchronized with Git.

Ana uses this ontology before answering questions. Instead of starting from scratch every time a user asks something, it searches the organization’s stored business context and retrieves the relevant definitions and sources. This can help reduce unnecessary model context and improve consistency between AI-generated analysis and the organization’s established business definitions.

TextQL’s AI analyst can query multiple sources in the same workflow. For example, a question can combine historical information from Snowflake with current information from Salesforce, allowing users to analyze data that would normally require manually combining different systems. TextQL supports both SQL-based connectors and API-based connectors, with API sources accessed through a secure Python sandbox.

The platform supports complex analytical tasks that require more than a single database query. Ana can break larger questions into multiple subtasks, use parallel subagents to process large datasets or information from several sources, and then combine the results into a single response. This agentic architecture is intended to replicate more of the investigative process used by human data analysts.

TextQL also uses Python for analytical work. After retrieving data, Ana can perform statistical analysis, transformations, calculations, and other analytical operations before producing the final result. This makes the platform capable of going beyond straightforward SQL generation and into deeper data investigation.

Visualization is another part of the product. Ana can generate charts and other analytical outputs based on the question being asked. The platform can also produce reports, alerts, and data applications rather than limiting the output to a text response. TextQL has expanded this approach with Data Apps, which allow users to turn governed data and ontology definitions into interactive applications.

TextQL also supports embedded analytics. Organizations can embed Ana inside their own applications so that customers or employees can ask natural-language questions about their data without leaving the organization’s product. This allows companies to provide AI-powered analytics as part of their own software experiences.

The platform is designed with enterprise security and governance in mind. TextQL states that credentials are stored in its secure vault and injected at the network layer rather than being exposed directly to the AI agent. Administrators can whitelist domains and control which external systems the agent is allowed to access. SQL connections inherit permissions from the organization’s existing data warehouse environment.

TextQL also provides auditability. Ana’s queries and runs can be logged, and the platform is designed so organizations can trace analytical activity. Enterprise deployments add capabilities such as SAML/OIDC single sign-on, SCIM provisioning, dedicated infrastructure, VPC and on-premises deployments, private links, HIPAA support, SOX support, custom model fine-tuning, and priority support.

The platform supports both traditional analytics systems and modern AI workflows. TextQL can connect to semantic layers such as dbt Semantic Layer and LookML, business-intelligence systems such as Tableau, Power BI, and Looker, and data workspaces such as Hex and Mode. This means organizations can use existing analytical definitions and validated dashboards as part of the context available to Ana.

TextQL provides a broad connector ecosystem. Its current connector directory lists 63 available data sources, including 18 SQL connectors and 45 API connectors. SQL connectors include Snowflake, Databricks, BigQuery, Redshift, PostgreSQL, MySQL, SQL Server, Oracle, Athena, SAP HANA, and Trino. API connectors include Salesforce, HubSpot, Slack, GitHub, Jira, Notion, Airtable, Linear, Zendesk, OpenAI, Anthropic, Stripe, Datadog, Sentry, PagerDuty, and many others.

Another important capability is the ability to connect AI systems to TextQL. TextQL provides integrations with AI services such as OpenAI and Anthropic, allowing Ana to use external AI capabilities alongside organizational data. The company also emphasizes that credentials are protected through network-level injection and administrative controls.

TextQL is particularly focused on data-intensive industries. Its current solutions include finance and accounting, financial services, manufacturing, healthcare, and government. These applications include financial reporting, portfolio analysis, predictive maintenance, claims processing, eligibility verification, population health, revenue-cycle analysis, government program oversight, budget analysis, and operational analytics.

The platform is also designed for enterprise deployment models. Analyst and Team plans operate as multi-tenant SaaS, while Enterprise customers can use dedicated or single-tenant infrastructure, VPC deployments, on-premises environments, bring-your-own-cloud configurations, private links, and embedded or white-label solutions.

TextQL was founded in 2022 by Ethan Ding and Mark Hay, according to the company’s current Data Ecosystem Wiki. The company is headquartered in New York and has developed Ana as its central AI analyst product.

TextQL therefore sits between business intelligence, data analytics, semantic-layer technology, and AI agents. Its main distinction is that it attempts to make an AI analyst understand an organization’s entire data environment rather than simply generating SQL against one database. The combination of an organization-owned ontology, multi-source connectivity, agentic analysis, Python computation, natural-language interaction, data applications, and enterprise deployment makes TextQL particularly suited to organizations with complex data environments.

Pricing

TextQL currently offers three main plans: Analyst, Team, and Enterprise. The Analyst plan starts at $0 per month plus compute and includes $100 per month in free credits, three seats, shared compute, unlimited connectors, and the platform’s integrations. The Team plan costs $250 per month plus compute and includes $250 per month in credits, unlimited seats, team collaboration, and email support. Enterprise pricing is customized and adds dedicated infrastructure, advanced identity, compliance, deployment, and support capabilities.

TextQL uses ACU (Agent Compute Units) for compute-based usage. The Analyst plan includes 50,000 ACUs, while Team includes approximately 83,333 ACUs. Overage rates differ between plans, and Enterprise usage is customized.

Review

TextQL stands out because it approaches AI analytics as an AI analyst and data infrastructure problem, rather than simply a natural-language SQL generator. Ana can work across warehouses, BI platforms, semantic layers, APIs, documentation, and business applications.

The ontology is one of its strongest differentiators. By storing business definitions, permissions, methodologies, and context in a version-controlled model, TextQL attempts to make AI-generated analysis more consistent with the organization’s existing understanding of its data.

Its connector ecosystem is another major advantage. The ability to combine warehouse data with live CRM, support, productivity, development, and SaaS information can eliminate many manual data-gathering steps. The current platform supports 63 connectors, and all connectors are included across the pricing tiers.

The agentic architecture is also useful for complex analytical questions. Ana can break down large tasks, query multiple sources, perform Python-based analysis, and produce visualizations and reports. This gives it a broader analytical scope than tools focused primarily on generating SQL.

Security is another strength for enterprise users. Credentials are protected through network-level injection, existing warehouse permissions can be inherited, and enterprise customers can use dedicated, VPC, on-premises, or bring-your-own-cloud deployments.

The main limitation is that TextQL is primarily aimed at organizations with substantial data infrastructure. Smaller teams that only need basic dashboards or simple spreadsheet analysis may find the platform more sophisticated than necessary.

Pricing also requires attention because the monthly subscription price is only part of the cost. Compute is separately accounted for through ACUs, meaning organizations with intensive analytical workloads need to monitor usage.

Another consideration is implementation. Although users can ask questions in natural language, getting the best results from an enterprise AI analyst still depends on connecting relevant systems, establishing useful business context, and maintaining a trustworthy ontology.

TextQL is therefore best suited to data-intensive organizations, enterprise analytics teams, financial services companies, healthcare organizations, manufacturers, government agencies, and businesses that need to make information accessible across complex data environments.