Cube vs Datarails (2026)

Cube connects to hundreds of source systems, with guides for NetSuite, Salesforce, Workday, Sage Intacct, Snowflake, Databricks, ADP Workforce Now, QuickBooks, HubSpot, Rillet, Dynamics 365 Business Central and Campfire, and syncs them on a schedule or on demand. Agents can start a fresh import from an existing connection before they answer, and they put their work into Cube scenarios, Boards, Tables and decks. None of them books an entry, updates a deal or changes an employee record back in the ERP, CRM or HR system.

Sourcecubesoftware.com/integrationsread 2026-10-13

FinanceOS connects to more than 600 systems and data sources, from ERPs, CRMs and HR platforms to banks, billing systems and spreadsheets, and keeps them synchronized and consolidated. The AI works on what arrives. Spend codes, matches and reconciles invoices and card transactions against ERP entries, Desk pulls invoice, payment, vendor and customer records into each request, and the assistant drafts a journal entry in the conversation for a person to check and post. Corrections and payments stay inside Datarails, and no agent posts a change back into the ERP or CRM by itself.

Sourcedatarails.com/platform/financeosread 2026-10-13

Charlie, the super agent, breaks a question into tasks, routes each piece to one of four agent teams (Data Managers, Analysts, Planners and Business Partners), lets them pull, reconcile, model and draft, then checks the combined answer against the source data before returning it. One request can run from a data refresh through variance analysis to a drafted board deck. Cube Workflows add task lists with owners, assignees, statuses and due dates for processes such as the month end close.

Sourcecubesoftware.com/ai-agentsread 2026-10-13

The strategy, planning and reporting agents have merged into one general assistant, which asks clarifying questions, offers analysis options, confirms the key decisions and then builds the output, so one request can run from a data pull to a finished Excel model and deck.

Task agents sit inside the screens where the work happens, such as flux analysis and journal entries in the close. Workflows route budgets, forecasts and reports through the right reviewers, Month-End Close assigns tasks with deadlines across the close, and Desk routes each finance request by team, type and priority with SLA timers.

Sourcesupport.datarails.com/hc/en-us/articles/25257415989916-introduction-to-datarails-ai-agentsread 2026-10-13

People reach Charlie in the Cube Workspace, in the Excel and Google Sheets add ins, in Slack and Teams, and from any MCP assistant such as Claude or ChatGPT. Anomaly Detection watches the numbers and flags lines that look off before anyone asks. Source systems sync on a schedule, so fresh actuals land in the model and the plans update around them without someone starting the work.

Sourcecubesoftware.com/ai-agents/analystsread 2026-10-13

Insights sends AI written summaries on the schedule a team sets, weekly, monthly or quarterly, to the people it chooses, and automated agents rerun recurring analyses without being asked. Desk takes finance requests from email, Slack, Teams and other channels into one queue, where the AI gathers the records and drafts an answer. People also reach the agents and the assistant in the web app, and outside assistants reach the data through the AI Connector.

Sourcedatarails.com/platform/airead 2026-10-13

Charlie answers from the customer's own Cube. It decides which slices of actuals, budgets, forecasts and scenarios it needs and fetches them through the governed layer, with the company's formulas, account and entity hierarchies and currency conversions already applied, so revenue means what the company means by revenue. It can drill from a summary figure to transaction lines in the connected source systems, and every number opens back to the records behind it.

Sourcehelp.cubesoftware.com/hc/en-us/articles/36376094163092-about-ai-at-cuberead 2026-10-13

Agents work on FinanceOS, a governed data model that consolidates, maps and harmonizes data across entities, currencies and systems, with mappings, allocations and version control kept by the finance team. A Semantic Layer defines business metrics such as revenue, gross margin, headcount, ARR and DSO once for use across Datarails, and the assistant says whether a number came from a governed metric or an aggregation of the tables. Every answer can drill from a summary to departments, accounts and transactions.

Sourcedatarails.com/platform/financeosread 2026-10-13

Each conversation with Charlie and each Planner session is saved for the user who started it, and a History view reopens any of them to carry on where it left off. Context stays inside one conversation, so starting a new one for an unrelated question keeps the earlier context from carrying over. Admins can also write standing instructions for Charlie, such as always showing budget against actuals or how the fiscal calendar runs, which apply in every conversation for every user. Nothing an agent learns in one conversation reaches the next.

Sourcehelp.cubesoftware.com/hc/en-us/articles/41283352702356-chat-with-charlie-cube-s-ai-analystread 2026-10-13

Every conversation with the assistant is saved automatically for the user who had it, across all areas of the product, and nobody else sees it. Each conversation runs as a session on Anthropic's infrastructure, which holds its history, files and outputs, so it falls outside Anthropic's zero data retention option. A team can also keep company profile documents, such as board presentations or financial models, that agents and outside assistants read for context, and the team reviews and updates them itself. An agent does not write down what it learned from a conversation for next time.

Sourcesupport.datarails.com/hc/en-us/articles/25257473410332-how-ai-agents-use-context-metrics-and-dataread 2026-10-13

Agent writes land only in scenarios opened for AI write back or created by the AI, so the live budget stays untouched unless someone opens it, and each write carries the user's name and an AI flag. Writing through MCP needs its own permission, which only Finance or Admin users can hold, new workflows and tasks arrive as drafts with no assignee, and a publish can be undone. Data Manager agents score each proposed mapping or bulk change by confidence, and the team approves it before anything lands.

Sourcehelp.cubesoftware.com/hc/en-us/articles/54111068392724-what-ai-tools-can-do-through-the-cube-mcp-serverread 2026-10-13

Before creating or changing anything, the assistant states exactly what will change and waits for the user's OK, so nothing is created or changed without approval. A journal entry arrives in the conversation as a draft for the user to check, and a person posts it, never the agent. In Spend, agents work under defined controls and approval requirements, so a person approves before any payment, and invoices with discrepancies wait for review. Desk drafts each reply for a team member to review and send.

Sourcesupport.datarails.com/hc/en-us/articles/30100850412444-using-the-datarails-ai-assistant-betaread 2026-10-13

Cube holds SOC 2 Type II and SOC 1 reports, the SOC 2 examined by Dansa D'Arata Soucia, and handles data to HIPAA requirements. Sign in runs through SAML or OIDC single sign on with Okta, Duo, Google or Microsoft Entra, which an admin can enforce, plus app or text based multifactor authentication. Role based access and each user's Data Access Scope reach down to the cell, the agents inherit the asker's permissions, and separate permissions govern Charlie, the Planner, MCP read, MCP write and AI administration, so AI can be switched on or off for each use through the same roles.

Sourcecubesoftware.com/securityread 2026-10-13

Datarails holds a SOC 2 Type II report and a SOC 1 report, both available on request from its trust center. Sign in supports SAML 2.0 single sign on with Okta, Microsoft Azure, Duo, Google Workspace and JumpCloud, with multifactor authentication. Access follows roles (Super Admin, Admin, Contributor and Viewer) and permission groups down to the entity, department or report, the AI tools read only what the user is already allowed to see, and authorization is checked again on every request.

Sourcetrust.datarails.comread 2026-10-13

Every agent query, answer and publish is recorded with the user, the permission scope applied and the sources used, and denied actions are recorded too. The audit trail shows each data change with who made it, the event type, the number of values changed and the before and after figures, and change history marks which changes came from AI. Because answers open to the transactions behind each figure, a reviewer can see exactly what an agent relied on.

Sourcecubesoftware.com/securityread 2026-10-13

FinanceOS keeps audit trails and full data lineage on the numbers, and the close records approvals and every change. Super Admins get an audit log of outside assistant sessions on the AI Connector, listing each session's member, AI platform, time, duration and status with a timeline of its events, while a durable record of each tool call is still being built. Conversations with the in app assistant stay visible only to the user who had them, so an admin has no record of an agent session to open later.

Sourcedatarails.com/platform/financeosread 2026-10-13

Cube runs as a hosted service on AWS, with all data and infrastructure in US based regions, and offers a data processing agreement for GDPR needs. A company that needs its financial data held in the EU or another country cannot get that, and there is no customer hosted edition.

Sourcesecurity.cubesoftware.com/faqread 2026-10-13

Datarails runs on Microsoft Azure in the United States, Canada and the UK, and each customer's environment sits in one of those regions, with the application database following the same Azure location. When a user works with the assistant, some data leaves Datarails for Anthropic's infrastructure, in a region left unnamed. The AI Connector is processed in the United States, with routing by region under evaluation and no delivery date. The platform is a hosted service with no customer hosted edition.

Sourcedatarails.com/sub-processorsread 2026-10-13

Four agent teams come with every plan. Data Managers handle mapping, bulk edits, connection setup and hierarchy upkeep, and Analysts bring slices, Smart Variance and Anomaly Detection. Planners carry the Planner Agent, Smart Forecaster and Formula Modeler, and Business Partners carry Visualization, Board Builder and Deck Builder. The Agent Exchange adds ready plays, such as a variance bridge to budget, a rolling forecast refresh, an investor update or a new entity mapping, which Charlie runs once the user fills in the bracketed inputs.

Sourcecubesoftware.com/ai-agentsread 2026-10-13

Agents sit ready in the screens where finance work happens, each switched on for an account separately: flux analysis and journal entry drafting in the close, a month end close assistant, cash transaction matching and an intercompany reconciliation assistant. Spend adds agents for invoice extraction and coding, transaction classification, matching and exception checks, and Desk adds an assistant that drafts answers to finance requests. Teams working in Claude get a Datarails plugin with ready made finance skills that produce Excel workbooks and PowerPoint decks.

Sourcesupport.datarails.com/hc/en-us/articles/30100850412444-using-the-datarails-ai-assistant-betaread 2026-10-13

Charlie and the agent teams run on language models from providers available through Amazon Bedrock, chosen by Cube, with Anthropic named among Cube's subprocessors since March 2026, and Cube trains no models of its own for these features. AI settings hold a company's standing instructions and nothing else, so a company cannot pick the model behind its agents. Outside assistants such as Claude or ChatGPT can work on Cube data through the MCP server, but those are the customer's own tools reading Cube.

Sourcesecurity.cubesoftware.com/faqread 2026-10-13

The assistant and the agents built into Datarails screens run on Anthropic's Claude models through Anthropic's Claude Managed Agents platform, and other AI features use OpenAI models, directly and through Azure OpenAI. A customer cannot switch its agents to another provider or model. Teams that prefer Claude or ChatGPT can also query Datarails data through the AI Connector, but that is their own assistant reading Datarails.

Sourcesupport.datarails.com/hc/en-us/articles/30100850412444-using-the-datarails-ai-assistant-betaread 2026-10-13

The Cube MCP Server, generally available on every plan with usage limits, signs users in with OAuth and offers a full tool set, from get_financial_slice_of_data and get_drilldown_data to publish_data, duplicate_scenario, create_dashboard, create_planning_table, create_workflow and initiate_import, so a team can point any agent framework at the governed model and write plans, Boards, Tables, scenarios and workflows back under the user's permissions. The Cube REST API has a public OpenAPI reference covering agent chat and Planner sessions, AI settings and credits, canvases, workflow runs and connection syncs, authenticates through OAuth with a PKCE flow, and is sold as an add on.

Sourcehelp.cubesoftware.com/hc/en-us/articles/54111068392724-what-ai-tools-can-do-through-the-cube-mcp-serverread 2026-10-13

The FinanceOS AI Connector is a hosted MCP server with OAuth 2.0 and PKCE sign in per user and a public tool list for discovering tables, reading metrics, running filtered queries and aggregations. Every connector tool is read only for now, so an outside agent can analyze Datarails data but not change it. Any Datarails customer can also pull its tables over an OData feed at no extra cost, and the Data Gateway Service API accepts CSV and Excel uploads into Fileboxes from other systems.

Sourcesupport.datarails.com/hc/en-us/articles/25849710214556-datarails-financeos-mcp-server-technical-documentationread 2026-10-13

Before Charlie returns an answer it checks each figure against the source data, and every query, answer and AI write sits in the log for review afterward. The Planner, still in open beta, asks clarifying questions before it builds a scenario. Cube tested its own accuracy on its own books and does not warrant AI output, reminding users on every AI surface to check results. A finance team cannot run an agent against a set of known questions to see how it does before giving it to the wider team.

Sourcecubesoftware.com/ai-agentsread 2026-10-13

Custom AI workflows built under the AI Transformation package are tested against real data by a Datarails forward deployed engineer before they go live, which is a service, not something a team runs itself. Datarails' AI terms leave it to the customer to review outputs for accuracy before relying on them, and within the product a finance team cannot score an agent's answers or try it on known questions before rolling it out.

Sourcedatarails.com/datarails-ai-transformation-servicesread 2026-10-13

The agents work through Cube's governed data layer, its source connections and the spreadsheet, Slack, Teams and MCP surfaces. None of them opens another application and clicks through it for a user.

Sourcecubesoftware.com/ai-agentsread 2026-10-13

Agents work on FinanceOS data and produce Excel files, decks, PDFs and widgets inside Datarails. Data from other systems arrives through connectors and uploads, and when Claude works in Excel the Datarails add in carries out refreshes and drill downs rather than an agent reading the screen. None of Datarails' agents operates another application's screen for a user.

Sourcedatarails.com/platform/airead 2026-10-13