Count pricing guide details
Pricing Overview
Count’s pricing page lists three plans: Free, Pro, and Enterprise. Free is listed at $0 and does not require a card. It includes one project, three canvases, up to five users, daily AI-credit limits for limited models, one database or MCP connection, web search, file uploads, two automations, and community support. The plan also includes a 500MB DuckDB compute-layer capacity and a 500MB Python VM size.
Pro is listed at $49 per editor per month. Count describes Pro as including unlimited AI credits on a fair-use basis, unlimited canvases, projects, reports, and automations, three database or MCP connections, shared agent skills and context, Slack integration, an MCP client, and email, Slack, or Teams support. The detailed plan table lists up to 75 workspace users on Pro. Owners, Admins, and Editors are identified as billable seats, while Viewers and Report Viewers are free; the Pro role listing names Owners, Editors, Viewers, and Report Viewers.
The Pro overview states 4GB compute capacity for the Python VM and dataset extract size. Elsewhere in the same pricing table, Pro is listed with 2GB for both DuckDB compute-layer capacity and Python VM size. Buyers planning around compute limits should confirm the applicable Pro capacity with Count before purchase.
Enterprise is presented as “Let’s talk,” with a “Get in touch” path rather than a public price. It adds unlimited editors and database connections, customer model selection or API keys, a governed semantic layer with admin change approvals, up to 32GB compute capacity, REST API access, workspace telemetry, user groups, row-level security, embedded analytics, SSO, SCIM, HIPAA, and EU/US data residency. Buyers should confirm the Enterprise price, billing terms, included implementation services, and applicable licensing details directly with Count.
Plan Comparison
Count’s current pricing page organizes features across Free, Pro, and Enterprise. The published plan limits distinguish workspace access, AI credits, connections, governance, and compute capacity.
Scroll sideways to see all columns. With a keyboard, focus this table and use the arrow keys.
| Plan | Published price and term | Workspace and analysis limits |
|---|---|---|
| Free | $0 | Up to 5 users, 1 project, 3 canvases, daily AI-credit limits, 1 database/MCP connection, 2 agent automations, 500MB DuckDB capacity, and a 500MB Python VM |
| Pro | $49 per editor per month | Up to 75 workspace users, unlimited canvases, projects, reports, and automations, fair-use AI credits, 3 database/MCP connections, shared agent skills and context, and 14-day version history |
| Enterprise | No public price listed; “Let’s talk” | Unlimited users, editors, projects, canvases, reports, automations, and database connections, with Enterprise governance and security capabilities |
Free includes web search, file uploads, Google Sheets import, public sharing that can be disabled, and community support. The pricing page describes file uploads as supporting CSV, PDF, Excel, Word, and more. Free users receive automatic model selection with limited models. The page also lists SOC 2 and GDPR on every plan, no model training on customer data on every plan, and Google and Microsoft login.
Pro adds scheduled reports, Slack integration, Count’s MCP client, custom templates, internal embedding, and 14-day version history. Pro lists fair-use AI credits and says credit top-ups are optional with spend controls. Buyers should confirm the included allowance and top-up rates for their expected AI workload. Pro also lists three database or MCP connections, including connections to data sources such as BigQuery, Snowflake, and Postgres, plus dbt integration.
The Pro overview lists 4GB compute capacity, while the detailed analytics-engine table lists 2GB for both Pro DuckDB compute-layer capacity and Pro Python VM size. The pricing page defines the compute-layer measure as the largest table Count’s DuckDB compute layer can hold. Confirm the applicable limit before relying on either figure.
Enterprise includes custom model selection and admin controls, support for Anthropic, OpenAI, Google, or custom API endpoints, unlimited database connections subject to admin approval, and up to 32GB for both the DuckDB compute layer and Python VM. It also lists user groups, row-level permissions, white-label embedded analytics, a REST API, workspace telemetry, SSO, SCIM provisioning, session-duration controls, export controls, HIPAA compliance, and EU/US data residency.
Hidden Costs and Considerations
Count’s current pricing page makes several cost boundaries explicit. Pro is priced per editor per month, not as a single flat workspace fee. The page identifies Owners, Admins, and Editors as billable seats, while Viewers and Report Viewers are free. However, its Pro role listing names Owners, Editors, Viewers, and Report Viewers, while Admins appear in the Enterprise role listing. Teams should confirm which roles are billable in their selected plan and how many editor seats are required.
Pro lists fair-use AI credits, while Free has daily AI-credit limits for limited models. Pro credit top-ups are optional and have spend controls; Enterprise top-ups are optional and have spend controls and volume discounts. Buyers should confirm the included allowance, top-up rates, and expected AI usage before budgeting. Enterprise customers can choose their own models or bring API keys, so buyers should confirm the commercial treatment of any selected model or external API use.
The pricing page lists three database or MCP connections on Pro and unlimited database connections on Enterprise, with Enterprise connections subject to admin approval. It lists one database or MCP connection on Free. Count names BigQuery, Snowflake, and Postgres among database and MCP connection examples, but the page does not publish third-party warehouse, model-provider, or integration charges. Buyers should establish those separate costs with the applicable provider.
Compute capacity needs confirmation. Pro’s plan overview says 4GB compute capacity for the Python VM and dataset extract size, whereas the detailed feature table says 2GB for Pro DuckDB compute capacity and 2GB for the Pro Python VM. Enterprise is listed with up to 32GB for both measurements. Count’s page does not state which Pro figure governs, so capacity-sensitive teams should obtain clarification before committing.
Enterprise has no public price on the pricing page. Its “Let’s talk” and “Get in touch” presentation does not disclose an amount, currency, billing period, minimum commitment, or implementation terms. Buyers should ask Count for those specific commercial details, along with scope for SSO, SCIM, HIPAA, data residency, embedded analytics, and custom training.
Cost Estimates by Team Size
Count publishes one paid numeric rate: Pro at $49 per editor per month. Count separately identifies Owners, Admins, and Editors as billable seats, with Viewers and Report Viewers free. Confirm how its published per-editor rate applies to the selected plan's billable roles before calculating a team total. Pro supports up to 75 workspace users and lists Owners, Editors, Viewers, and Report Viewers as its available roles. Viewers and Report Viewers are free, which means a workspace’s user count alone does not establish its Pro cost.
Scroll sideways to see all columns. With a keyboard, focus this table and use the arrow keys.
| Planning input | Published detail | What to confirm |
|---|---|---|
| Free workspace | $0; up to 5 users | Whether 1 project, 3 canvases, 1 database/MCP connection, and 2 automations fit the intended use |
| Pro editor pricing | $49 per editor per month | Billable Owner and Editor counts, any Admin roles, and how the published rate applies to each role |
| Pro workspace capacity | Up to 75 workspace users | Mix of Owners, Editors, Viewers, and Report Viewers |
| Pro AI usage | Fair-use credits; optional top-ups with spend controls | Expected workload and any potential top-up need |
| Enterprise | No public price listed | Price, currency, billing term, commitments, and implementation scope |
A Free workspace includes daily AI-credit limits, 500MB DuckDB compute-layer capacity, a 500MB Python VM, web search, file uploads, and community support. It is a published $0 option with no card required, but its limits should be evaluated against the intended project, canvas, connection, automation, and AI requirements.
For Pro planning, the public price is a per-editor monthly rate, not a team total. The plan includes unlimited canvases, projects, reports, and automations; three database or MCP connections; shared agent skills and context; Slack integration; an MCP client; and email, Slack, or Teams support. A team total requires confirmation of the billable role mix and applicable rate for each role. The page does not publish a total price for a five-, 15-, or 75-user deployment, and it does not state annual pricing.
Capacity planning warrants special attention because Count’s Pro overview states 4GB compute capacity for the Python VM and dataset extract size, while its detailed feature table lists 2GB for Pro DuckDB compute capacity and 2GB for the Python VM. Do not use either value as a pricing proxy; confirm the operative capacity limit with Count.
Enterprise adds unlimited editors, up to 32GB compute capacity, custom model or API-key options, REST API access, and governance capabilities such as SSO, SCIM, row-level security, and data residency. Its pricing page does not disclose an Enterprise amount, currency, term, minimum commitment, or license basis, so an Enterprise cost estimate requires those details from Count.
How Count Pricing Compares
Count’s pricing page provides a published $0 Free plan, a Pro plan at $49 per editor per month, and an Enterprise plan without a publicly listed price. The current page does not list a Scale plan. It also does not publish competitor pricing, so a like-for-like price comparison with other products cannot be made from Count’s pricing information alone.
Scroll sideways to see all columns. With a keyboard, focus this table and use the arrow keys.
| Count tier | Price disclosure | Selected published capabilities |
|---|---|---|
| Free | $0 | Up to 5 users, 1 project, 3 canvases, daily AI-credit limits, 1 database/MCP connection, 2 automations, web search, file uploads, and 500MB compute and Python VM limits |
| Pro | $49 per editor per month | Up to 75 workspace users, unlimited canvases, projects, reports, and automations, fair-use AI credits, 3 database/MCP connections, scheduled reports, and Slack integration |
| Enterprise | No public price listed | Unlimited editors and connections, up to 32GB compute capacity, custom models or API keys, REST API, SSO, SCIM, HIPAA, and EU/US data residency |
The relevant pricing distinction is between billable building roles and free viewing roles. Count identifies Owners, Admins, and Editors as billable seats, and identifies Viewers and Report Viewers as free. Pro specifically lists Owners, Editors, Viewers, and Report Viewers; Enterprise adds Admins. A buyer comparing Count with another tool would need the other product’s supported role definitions and billing basis before comparing total deployment cost.
AI allowances also vary by tier. Free has daily AI-credit limits with limited models. Pro includes fair-use AI credits and optional top-ups with spend controls. Enterprise includes fair-use AI credits, optional top-ups with spend controls and volume discounts, and the ability to choose models or bring API keys. Confirm the included AI allowance and top-up rates for the intended workload before comparing total cost.
The page has an unresolved Pro compute discrepancy: its overview cites 4GB compute capacity for the Python VM and dataset extract size, while the detailed table cites 2GB for both Pro DuckDB compute capacity and Python VM memory. Enterprise is listed at up to 32GB. Buyers comparing infrastructure limits should verify the current Pro entitlement directly with Count.
Enterprise is not comparable as a public starting-price offering because its page says “Let’s talk” and gives no amount, currency, billing period, or commitment. Confirm those commercial terms, along with requirements for SSO, SCIM, row-level permissions, export controls, and data residency, before evaluating an Enterprise deployment.