Questions people actually ask

A visual BI workspace you drive by conversation. You bring a dataset, ask questions in plain language, and each answer lands on an infinite canvas as a chart, a metric, a table or a written insight. Follow-up questions build on what's already there, so an investigation stays in one place.

A dashboard answers the questions someone anticipated. DashboardGPT answers the question you have right now, then the one it prompts. There's no build step: the visualisation is chosen from the shape of the answer, and the canvas keeps the chain of reasoning rather than a fixed layout.

No. Questions resolve to real queries against your dataset, and the result is rendered as an interactive visualisation you can modify, rearrange and follow up on. The conversation is the interface; the canvas is the artefact.

No. The playground runs on a curated demo dataset with no sign-up. You can ask questions, create charts, change how they're visualised and rearrange the canvas. An account is only needed to analyse your own data.

Not anonymously. Uploads, your own AI provider key, saved workspaces, export and sharing all live in the authenticated workspace. Choosing "Try your own data" takes you to sign-up with that intent carried across.

CSV and Excel uploads on every plan. Warehouse connections are coming soon on the Team plan; analysis will run through a Superset-powered query layer, so the data stays in the system of record.

Uploads are sized per plan; once warehouse connections arrive, warehouse-backed analysis will be bounded by your warehouse, not by us. Questions are answered with aggregated queries, so the volume that matters is the result set rather than the table.

DashboardGPT is provider-agnostic. Every plan includes managed AI, which is rate-limited and slower at busy times. You can also bring your own key for OpenAI, Anthropic, Azure OpenAI, Google or any OpenAI-compatible endpoint: there are no rate limits, responses are faster, and model usage is billed to your account under your own terms. On-premise installs can also point at a local model served through an OpenAI-compatible API, such as Ollama or vLLM, so no prompt leaves your network.

No. Your data is used to answer your questions. When you bring your own provider key, requests go to that provider under the agreement you hold with them.

Every workspace is a separate tenant. Datasets, canvases, conversations and credentials are scoped to the workspace that created them, and query execution happens inside that boundary.

We don't hold SOC 2 yet. We complete your security questionnaire, and the on-premise edition keeps all data inside your network.

Yes. The on-premise edition installs with Docker Compose on a single host or with a Helm chart on Kubernetes, and an offline bundle covers air-gapped environments. You bring your own LLM key or run a local model, and your data, queries and prompts stay inside your network. See the on-premise pages for the architecture, system requirements and self-hosted plans.

Every new account starts on a free 15-day trial of the Team plan, with no card required. The trial covers up to 3 people, 5 uploaded files, 2 canvases and 500 AI runs a month. When it ends, choose Explore or Team and carry on where you left off.

Yes. Every plan includes managed AI, with rate limits and slower responses at busy times. If you need more speed or volume, add your own AI provider key: there are no rate limits on your key, and the provider bills you directly with no markup from us.

Plans differ in how many people can join, how many queries you can run, how many charts you can create, how many files you can upload and how much data you can process. Explore allows up to 2 users and Team up to 50; Enterprise has no user cap. Usage limits are per workspace and reset every month. The compare plans table above lists every limit.

Yes. Annual billing on the Team plan is 20% off: $32 per user per month, billed once a year, instead of $40 billed monthly.

The Team plan is paid by card. Annual contracts over $5,000, Enterprise plans and self-hosted licenses can be paid by invoice, and we accept purchase orders.

Yes. Many teams start on the cloud while their security review runs. When you move to a self-hosted license, we help you move your workspaces, data connections and canvases across.

Your workspace stays intact and drops to the Explore plan. Nothing is deleted; the features above that plan are paused until you upgrade.