
Aleph is an AI-native FP&A data layer for startup and scale-up finance teams: it syncs 150+ business systems into the Excel and Google Sheets models the team already runs, and adds an agent and a Claude connector that can read and change the workspace.
Independently scored, one dimension at a time. Every dimension is sourced, and we say whether a claim is published, reported, or estimated.
Written by the Audit Friendly research team. No vendor edits, no sponsored placement.
Aleph sells the plumbing under a spreadsheet-run finance team. It pulls actuals from 150+ systems into a cloud data layer, lets finance map and reshape them with calculated tables, and pushes live numbers into existing Excel and Google Sheets models through two-way add-ins. Dashboards, Scan (AI variance analysis), AI mappings and Aleph Agent in Slack, Teams and Claude sit on top of the same governed tables.
The catch is that Aleph does not give you a planning engine. Driver logic, three-statement links and scenario math stay in your workbooks, so a tangled model stays tangled. Prices are not published, consolidation means data consolidation with FX rates rather than a statutory close with eliminations, and the Claude connector exposes 155 tools that include delete and permission changes.
Before signing, get in writing: the billing unit and whether an ERP such as NetSuite or Sage Intacct forces the higher plan, the renewal cap, which of your connectors Aleph builds itself versus runs through Fivetran and how often each refreshes, the data deletion window after exit, and how an admin limits what Claude and other agents can write.
Jobs, mapped to how finance teams actually buy.
Forget the 40-row feature grid. Pick a finance workflow and see how this tool and its closest peers actually do it, with our verdict on who wins for whom.
Here's what it really costs: published rates where a vendor lists them, aggregated quote data where they don't, plus a path to your own number.
Scored on what an outside agent can actually reach: the documented API, how it authenticates, and whether the vendor ships an MCP server.
Exists: Aleph runs a first-party remote MCP server at mcp.getaleph.com/mcp, documented on its MCP configuration page with setup steps for Claude, Claude Code, ChatGPT, Cursor and VS Code. Claude's connector directory lists it as Anthropic verified and added in August 2026 (checked 1 November 2026); Aleph announced the MCP on 6 May 2026 and the directory approval on 6 October 2026.
Scope: Read and write. The directory lists 155 tools, from list_tables, run_sql_query and pull_data to push_table, manage_mapping, manage_exchange_rate, manage_permissions and delete tools for tables, dashboards, users and automations. The vendor's own page documents a smaller core set over tables, blocks, dashboards and integrations. Aleph says consequential changes are staged for preview and confirmation and versioned with the initiating user.
Turnkey: Mostly. A Claude user adds the connector from the directory or as a custom connector and signs in with OAuth; there are no API keys to manage. A workspace admin must first enable AI features in Settings, which also routes data to the AI subprocessors Aleph lists (OpenAI, Anthropic, Microsoft Azure, Vercel and Langfuse). No public REST API is documented, so the MCP is the external surface.
Inside the product, Scan runs flux and variance analysis against a materiality threshold the user sets and drafts commentary with drill-down links, AI mappings classify new source records, and Aleph Agent answers questions in the web app, Slack and Teams. Aleph says its agent gives the same answer every time and that weekly MCP tool calls reached 35,998 in mid-September 2026; both are vendor claims.
AI / agent readiness
Aleph runs a hosted first-party MCP server at mcp.getaleph.com/mcp, documented on its MCP configuration page and listed as Anthropic verified in Claude's connector directory with 155 tools (added August 2026, checked 1 November 2026). Tools read and write tables, dashboards, mappings, exchange rates and permissions under OAuth. An admin must enable AI features first, and no public REST API is documented.
Implementation
Implementation specialists run onboarding, and the vendor promises value on day one, power users by week one and core workflows automated in month one (customer success page). G2 data cited on Aleph's comparison pages puts go-live at about 1 month, and a full-access free trial lets a team test on its own data. Because models stay in spreadsheets, little is rebuilt, but messy workbooks remain the customer's job.
ERP & data connectivity
The integrations page lists 150+ connectors, including NetSuite, QuickBooks, Xero, Sage Intacct, Rillet, Campfire, Salesforce, HubSpot, Rippling, ADP, Snowflake, BigQuery and Databricks, and says new ones can be built in 48 hours. The security FAQ says core connectors are Aleph-built API integrations and others run on Fivetran. Data flows into Aleph and out to spreadsheets; no ERP write-back is documented.
Support
The customer success page promises a dedicated Slack or Teams channel, CSM office hours and a team of former bankers, consultants and FP&A operators. G2's plan listing, as quoted by CFO Shortlist on 24 September 2026, shows ticket-based support on Essentials and a dedicated CSM with a Slack or Teams channel on Premium, so confirm which applies to the quoted plan. Aleph's comparison page cites G2 Fall 2026 data rating its quality of support 9.8 against 9.2 for Cube. No SLA, response-time commitment or 24/7 coverage is published.
Compliance depth
The security page (checked 1 November 2026) states SOC 1 Type II and SOC 2 Type II, annual third-party penetration tests, AES-256 at rest, SAML SSO, SCIM directory sync, data-level permissions by table or dimension, version control and audit logs. The subprocessor list hosts everything in the US, and AI subprocessors apply only when a customer opts in. No ISO 27001 certificate or budget approval workflow beyond agent change approvals is documented.
Scalability
Data is stored and calculated in Aleph's cloud layer, and its comparison page says unlimited entities are included. Customers named in the 17 September 2025 Series B post include Zapier, Turo, Harvey and Chess.com. No limits on rows, users or refresh speed are published, and model size still depends on the customer's workbooks.
Modeling depth
Aleph stores scenarios, forecast versions and calculated tables and feeds them into Excel or Sheets by cell, pivot or custom function (spreadsheet add-ins page). Driver logic, three-statement links and headcount math live in the customer's workbooks. Aleph markets headcount and budget workflows, and its MCP tools include exchange rates, but it documents no native driver engine or intercompany eliminations.
Pricing transparency
There is no pricing page. G2's listing, as quoted by CFO Shortlist on 24 September 2026, names Essentials and Premium plans with no prices, and CFO Shortlist's own $20k to $40k a year band discloses no sample. Aleph does publish that implementation, support, unlimited entities and AI usage are included.
Aleph's custom functions pull mapped actuals into existing report cells, and a refresh runs on a click or a schedule. Envoy's FP&A director says financial statements that took days now take minutes. The hours figure is the vendor's case-study headline.
Turo started with revenue modeling and expanded to actuals, forecasts and budgets while still running Adaptive for planning. Its CFO said on a podcast quoted by Aleph that drivers tracked grew from 10 to 15 to about 50.
Scan compares any two datasets, flags lines above a materiality threshold the user sets, drills by department, class or vendor and drafts the explanation with links to the rows. The timing comes from Aleph's own product demo; reviewers should still check each driver it names.
With the connector, Claude can list tables, run SQL, push a table, create a calculated table, edit mappings and build dashboards inside Aleph. A named head of FP&A quoted by Aleph says Claude built and deployed a calculated table to production from one conversation.
Aleph pitches fractional CFO firms on connecting each client's ERP, pushing a budget, automating budget versus actuals and shipping a client dashboard from a reused template. The 40-minute figure is from Aleph's own August 2026 webinar description.
How Chess.com unified their global finance and gained real-time visibility with Aleph
Chess.com's CFO explains replacing manual ERP report exports with Aleph across many currencies and product lines, and opening reporting to budget owners. Relevant for a global, multi-currency business, though it is a vendor-produced testimonial. 4 minutes 38 seconds, posted 4 January 2026, about 430 views. Transcript not read.
How Zapier accelerated finance clarity and decision-making with Aleph
Zapier's senior director of accounting describes using Aleph for forecasting, close reporting, budget versus actuals and daily account-level ARR tracking from the warehouse. A credible named reference for a multi-entity, warehouse-heavy company, but produced by the vendor. 3 minutes 31 seconds, posted 4 January 2026, about 55 views. Transcript not read.
Aleph - The Future of Finance
A two-minute official overview of the platform posted in January 2026. Useful as a quick orientation before a demo; the description gives no feature detail, and it predates the April 2026 agent release. 2 minutes, posted 4 January 2026, about 580 views, made by the vendor. Transcript not read.
Welcome to the future of AI-powered FP&A
A two-minute launch reel from Aleph's April 2025 AI announcement covering the web app, spreadsheet add-ins and AI layer. Good for a first look at the interface, thin on detail and older than the agent and MCP releases. 1 minute 58 seconds, posted 10 April 2025, about 1,900 views, made by the vendor. Transcript not read.
Aleph for Fractional CFOs
A recorded session showing how a fractional CFO firm connects a client ERP, pushes a budget, automates budget versus actuals and ships a client dashboard in Aleph. The most complete public walkthrough of the workflow, useful for advisory firms and lean in-house teams. 44 minutes, posted 27 August 2026, about 170 views, made by the vendor; the 40-minute close claim in its description is the vendor's. Transcript not read.
"Can Claude do that? Maybe. But who challenges Claude?"
A clip from an Aleph and Graphite Financial webinar on what AI can automate in finance and where human judgment still has to challenge the output. Relevant to a lead planning how far to trust an agent like Aleph's, though it is an opinion piece with no product shown. 1 minute 18 seconds, posted 23 August 2026, about 25 views, made by the vendor. Transcript not read.
Our AI spend tripled in 2 months.
A fireside clip with Aleph's head of finance on how finance should measure the return on fast-growing AI spend. It frames the problem Aleph's AI budgeting content targets but shows no product. 52 seconds, posted 18 May 2026, about 200 views, made by the vendor. Transcript not read.
Your AI budget shouldn't be one big line item
A webinar clip in which Aleph's head of solutions argues for splitting AI spend by department, owner and project in the FY27 budget. It is planning advice rather than a product demo, useful as context for how Aleph thinks about budget structure. 1 minute 9 seconds, posted 28 September 2026, about 20 views, made by the vendor. Transcript not read.
Postscript was live with Aleph in a day.
A short customer clip about Postscript going live on Aleph within a day without rebuilding workflows. It speaks to the implementation-speed question every FP&A lead asks, but it is one vendor-selected customer. 58 seconds, posted 13 May 2026, about 95 views, made by Aleph. Transcript not read.
The same answer 10 out of 10 times: AI variance analysis in Aleph
Aleph's head of CFO partnerships explains how its agent answers a marketing budget variance question and then shows the tables it pulled from. Useful for a finance lead weighing whether AI variance output is repeatable and checkable. 1 minute 37 seconds, posted 7 September 2026, about 40 views, made by the vendor to sell the feature; the repeatability claim is the vendor's. Transcript not read.
Compliance depth
SOC 1 Type II and SOC 2 Type II, SAML SSO, SCIM, data-level permissions, version control and audit logs; US hosting on Google Cloud, on-premise on request.
Scalability
No published caps; Aleph says unlimited entities are included and data is calculated in its cloud layer, but model size still rests on your workbooks.
AI / agent readiness
Anthropic-verified Claude connector, added August 2026, with 155 read and write tools over tables, dashboards, mappings and permissions; OAuth sign-in.
Support
Vendor says every customer gets a dedicated Slack or Teams channel and CSM office hours; G2's plan listing shows ticket support on Essentials. No SLA published.
Implementation
Vendor-led by implementation specialists; Aleph promises value on day one and core workflows automated in month one; G2 data it cites shows 1 month to go live.
Pricing transparency
Quote only with no pricing page; G2 lists Essentials and Premium without prices (per CFO Shortlist), which reports $20k to $40k a year for software.
Modeling depth
Model logic lives in Excel or Sheets; Aleph stores scenarios, forecast versions and calculated tables, pulls actuals by cell and runs AI variance scans.
ERP & data connectivity
150+ connectors incl. NetSuite, Sage Intacct, QuickBooks, Xero, Snowflake, BigQuery, Salesforce and Rippling; some run on Fivetran; no ERP write-back documented.
AF Score
Audit Friendly score for Aleph across eight dimensions, 5 November 2026.
Price signal
Typical annual software cost for mid-market teams reported by CFO Shortlist on 24 September 2026; Aleph itself is quote only (checked 1 November 2026).
Ramp FP&A adoption
Share of Ramp businesses buying FP&A software in the past 12 months that paid Aleph, ranked first in the category (Ramp Rate, updated 1 November 2026).
Essentials
QuickBooks Online and Xero, 2 standard integrations, daily refresh, add-ins, dashboards, guided onboarding and ticket-based support (G2 listing, reported).
Premium
All ERP and data warehouse integrations, unlimited standard integrations, real-time refresh, dedicated CSM and a Slack or Teams channel (G2 listing, reported).
Implementation, support and AI usage
Aleph's comparison page lists implementation, support, unlimited entities and AI usage as included.
Free trial
Full-access trial; the vendor says a team can cancel and remove its data at any time.
Extra services for complex integrations
CFO Shortlist reports this band for integration work beyond the included hours.
Connect source systems and existing models; first live report
Hands-on training with implementation specialists to power-user level
Automate routine reporting and analysis before the next close
Ongoing CSM office hours and a shared Slack or Teams channel
The model still lives in your spreadsheets
Aleph's spreadsheet add-ins page describes pulling and pushing whole sheets or selected cells, and its modeling page adds a scenario repository and version control, but driver logic, three-statement links and headcount math stay in Excel or Sheets. A fragile 40-tab model becomes a fragile 40-tab model with live data. Teams that want driver-based, multi-dimensional planning in one engine should compare Abacum; Cube adds an OLAP cube but also keeps model logic in spreadsheets. Checked 1 November 2026.
No prices, and the ERP sets the plan
Aleph has no pricing page. CFO Shortlist, quoting G2 on 24 September 2026, says Essentials covers QuickBooks Online and Xero with 2 standard integrations, while ERP and data warehouse connections need Premium. A NetSuite or Sage Intacct company should price Premium from the start and ask what an upgrade mid-term would cost.
The Claude connector can delete and change permissions
Claude's directory listing for Aleph (checked 1 November 2026) shows write and delete tools across tables, dashboards, users, mappings and automations. Aleph's 6 October 2026 post says consequential actions are previewed and confirmed, destructive ones need explicit approval, and every change is versioned with the initiating user. Pilot with a few users and set tool permissions in the AI client before rolling it out.
Data consolidation is not a statutory close
Aleph maps data from several accounting systems into one view, and its MCP tools manage exchange rates; Aleph's 6 October 2026 blog post describes a customer building a multi-currency functional P&L with spot and average rates through Claude. Its pages describe no intercompany elimination, minority interest or multi-GAAP reporting module. Groups with a real consolidation close still need their ERP or a consolidation tool.
Turning on AI sends data to model providers
Aleph's subprocessor list marks Vercel, OpenAI, Anthropic, Microsoft Azure and Langfuse as used only for customers that opt in to AI features, all in the US. The MCP connector and Aleph Agent require that opt-in. Run it past security review before the first Claude pilot.
Subscription terms are not public
Aleph's terms page governs use of the website and was last updated on 13 May 2021, so renewal, uplift and termination terms sit in the order form. The security FAQ says data is deleted on request or automatically some days after service ends. Get the deletion window, a renewal cap and an export format in writing; CFO Shortlist reports typical increases of 5 to 7 percent a year.
Claude directory approves Aleph's read and write connector
Aleph announced its connector was approved in Claude's directory and says it is also verified in ChatGPT, Cursor and Grok. It reports 35,998 customer MCP tool calls in the week of 14 September 2026 (vendor figure).
Aleph pitches agentic budgeting for the 2027 cycle
Aleph published a 2027 agentic budgeting guide and product page describing agents that build baselines, collect budget-owner inputs in Slack and log approved changes. Treat it as announced until the workflow appears in documentation.
Aleph MCP launches for Claude, ChatGPT and Cursor
A remote MCP server lets AI clients query governed Aleph data and act on it, including inside Claude for Excel. Setup uses OAuth and requires an admin to enable AI features.
Aleph Agent builds and edits dashboards from prompts
Aleph extended its agent into dashboards so users can build and refine charts and KPI blocks in plain language.
Aleph Agent arrives in the web app, Slack and Teams
The agent answers questions from Aleph's consolidated data with cited sources and respects existing permissions. Aleph called it the first in a series of agent releases.
Aleph raises a $29M Series B led by Khosla Ventures
Picus Capital, Bain Capital Ventures and Y Combinator joined, bringing total funding to $46M. Aleph says it grew 10X since its Series A.
Aleph does not publish prices; every deal is quoted (checked 1 November 2026). CFO Shortlist reported on 24 September 2026 a typical $20,000 to $40,000 a year in software for mid-market deployments, and $30,000 to $60,000 in year one including implementation, without disclosing a sample.
G2 lists two plans, Essentials and Premium, both priced on request, per CFO Shortlist. Aleph's comparison page says implementation, support, unlimited entities and AI usage are included, so the main levers are plan tier and integration count.
Subscription terms are not public. The website terms of service, last updated in May 2021, cover the website and related services but set out no subscription term, renewal or fee rules, so ask for the term length, renewal cap and termination rights in the order form.
Models stay in your own Excel and Google Sheets files, and the security FAQ says all data can be synced and saved into spreadsheets through the add-ins. Aleph deletes company data on request or automatically some days after service ends, so export before the contract lapses.
Aleph's implementation specialists run it. The vendor promises value on day one, power users within the first week and core workflows automated within the first month, and G2 data cited on its comparison pages puts go-live at about 1 month.
Finance owns the workbooks, and system admins authorize the ERP, HRIS and CRM connections. CFO Shortlist reports 40 to 80 implementation hours included in typical contracts.
Yes. Aleph's connector is listed as Anthropic verified in Claude's directory (added August 2026) and also works in ChatGPT, Cursor and other MCP clients through mcp.getaleph.com/mcp with OAuth sign-in.
The listing shows 155 tools that read and write tables, dashboards, mappings, exchange rates, automations and permissions. A workspace admin must enable AI features first, and Aleph says consequential changes are previewed, confirmed and versioned.
Answered only from our own published research on this tool, never from general internet noise. If we cannot answer it well, our research agents will dig in and publish a sourced answer.