Core concepts
Data source
A data source is a connection to your data. Basedash queries databases and warehouses such as Postgres, Snowflake, and BigQuery directly, and syncs SaaS tools such as Salesforce and Stripe into Basedash Warehouse through connectors. Admins add data sources and control who can query them. For example, you might connect your production Postgres replica and sync HubSpot. Read more in Data sources and Data source access control.Basedash Warehouse
Basedash Warehouse is a managed DuckDB warehouse that Basedash runs for you. SaaS connectors sync into it, so you can query data from your CRM, billing, and support tools without running your own warehouse. Read more in Basedash Warehouse.Chat
Chat is Basedash’s AI data analyst. You ask a question in plain English, and the AI writes SQL, runs it against your data, and answers with a chart, a table, or a short write-up. Each answer shows the sources and context the AI used, so you can check how it got there. You can also chat with Basedash from Slack and from MCP clients through the MCP server. For example: “What was our net revenue retention by quarter last year?” Read more in Chat.Chart
A chart is a saved SQL query and a visualization of its result. You create charts by asking chat, by describing one on a dashboard, or by writing SQL. Every chart has a chart type, such as a line, bar, number, or table, and you can edit its query and settings at any time. Read more in Charts.Dashboard
A dashboard is a shared page of charts, organized into tabs and folders. Filters and variables let viewers change the date range or segment without editing charts, and dashboards refresh automatically. You share a dashboard with everyone, with groups, or with specific members, and you can publish a view-only public link. Read more in Dashboards and Filters and variables.Model
A model is a reusable SQL dataset with business meaning attached: column descriptions, measures, segments, synonyms, and relationships. Models are Basedash’s semantic layer. When a model defines a metric like monthly recurring revenue, the AI uses it instead of writing its own version, and you can query it in SQL asmodels.<reference_name>. Admins can mark models as verified.
Read more in Models.
AI context
AI context is written guidance that helps the AI understand your organization. It works at several levels: global context for everyone, group context for a team, personal context for one member, and descriptions on data sources, tables, and columns. For example: “A customer is churned when their subscription has been canceled for 30 days.” Read more in AI context.Skill
A skill is a named set of instructions for a specific task, such as how to calculate activation rate or how to triage support tickets. Unlike global context, which the AI always reads, the AI loads a skill only when it’s relevant to the request. Read more in Skills.Insight
Insights are AI-written posts about what changed in your data and why it matters. Each post has a short explanation, supporting numbers, and a chart you can explore. Insights run on a schedule and can be delivered to Slack and email. Read more in Insights.Automation
An automation is a task the AI runs for you and delivers to Slack or email. It can run on a schedule, when new rows match a SQL query, or manually. Use automations for recurring reports, alerts, and workflows that use tools from MCP connectors. For example: “Every Monday, summarize last week’s signups and post to #growth.” Read more in Automations.Subscription
A subscription delivers a dashboard or chart to email or Slack on a recurring schedule. Queries run when the subscription is sent, so recipients always get current numbers. Read more in Subscriptions.Members and roles
Everyone in your organization is a member with one of three roles. Admins manage data sources, members, and settings. Members create and edit charts, dashboards, and automations. View-only members can view what’s shared with them. Read more in Members and roles.Group
A group is a set of members, such as Sales or Finance. You share dashboards, chats, and data sources with groups, and groups can have their own AI context and row-level security rules. Groups can be synced from your identity provider with SCIM. Read more in Groups.How it fits together
- Connect data sources, or sync SaaS tools into Basedash Warehouse.
- Teach the AI with models, AI context, and skills, so answers use your definitions.
- Ask questions in chat, in Slack, or from an MCP client.
- Build charts from the answers and arrange them on dashboards.
- Monitor with insights, automations, and subscriptions that come to you.
- Control access with roles, groups, and data source permissions.