Abacum vs Pigment (2026)

Data comes into Abacum through more than 700 native integrations across ERP, HRIS, CRM, billing and data warehouse systems, plus SFTP and an Excel connector, and the AI works on the model those feeds keep current. Teams that keep hybrid models in spreadsheets stay in step through a two way Google Sheets integration or Excel uploads.

The MCP connection runs the other way, carrying Abacum's governed answers out to Claude, Gemini and ChatGPT. None of the agents creates, updates or sends anything in the source systems themselves, so their actions stay inside Abacum's own models, spaces and mappings.

Sourceabacum.ai/product/dataread 2026-10-13

Data reaches the model through scheduled imports from ERP, CRM, HR, data warehouse, BI, spreadsheet and file storage sources, and the agents work on that model instead of inside the source systems. Agent connectors reach outward to Notion and Slack. An admin switches a connector on, each member signs in with their own account, and each read, write or delete tool is set to always allow, needs approval or blocked. Google Drive is still to come, and a team cannot add an MCP server of its own, so agent action outside Pigment stops at those two products.

Sourcekb.pigment.com/docs/set-up-agent-connectorsread 2026-10-13

AI Modeler works from a plan. Asked for a model, it scans the connected datasets, picks a structure from its blueprints and proposes every variable and formula, and once approved it builds the model with its dimensions and totals in one pass, the same way it adds a variable and a new breakdown to a model that already exists. Workflow automation carries planning across people. Department spaces are duplicated by dimension, open and lock by status, and route budget and headcount requests to finance for a decision.

Sourceabacum.ai/blog/ai-modelerread 2026-10-13

In Plan & Build mode the Modeler Agent gathers context, writes a structured action plan and then carries it out across Dimensions, Lists, Metrics, formulas and data imports, asking before it switches to another mode. The Analyst Agent takes a Mission, a task written in plain language, and works through its subtasks.

It sources the data, drills into the drivers, runs its own code against the model's metrics and assembles a report of text and charts. A user can leave an agent working and come back to an unread badge on the chats that have answered. Custom Agents draw on the same analysis and documentation skills, though they cannot schedule Missions.

Sourcekb.pigment.com/docs/configure-and-run-the-modeler-agentread 2026-10-13

Incoming data starts work without anyone asking. As actuals land from connected systems, the AI classifies new accounts, maps and cleans the records and flags anomalies as they appear, and it pulls in business signals such as deal sentiment and product usage. Modeler and the chat agent start from a person's request inside Abacum, by typing or by voice, and the MCP connection lets the same question start in Claude, Gemini or ChatGPT.

Sourceabacum.ai/newsroom/abacum-rethinks-ai-for-finance-with-the-launch-of-abacum-intelligence-a-platform-wide-intelligence-layerread 2026-10-13

Saving an Analyst report as a Mission makes it repeatable, and switching on Scheduling sets a start time and a cadence for later runs, with each run's output kept in the report or exported to a text Metric. The Missions tab shows every saved Mission with its run frequency. People also start the agents from the AI sidebar on any Board, from the AI Analysis button beside a Metric, Table or numeric Widget, and from a prompt typed at the top of the Workspace homepage.

Sourcekb.pigment.com/docs/configure-and-run-the-conversational-analyst-agentread 2026-10-13

Answers come from the governed model Abacum keeps current from connected systems, so a question about pipeline conversion or spend by vendor rests on the same actuals, plans and dimensions finance reports from. AI Modeler also draws on a curated library of model blueprints that Abacum keeps extending. Board packs, contracts, benchmarks or meeting notes attached to a chat add outside context to that conversation alone.

Sourceabacum.ai/watch-and-learn/how-can-i-add-context-to-abacum-intelligence-agent-conversationsread 2026-10-13

Answers come from the customer's live planning model, its Applications, Blocks, Metrics and Boards, and an admin can switch AI data access off on any Block that should stay out of reach. The Analyst prepares itself once per Application on the first question, a setup that takes longer the more Blocks there are.

Custom Agents carry standing knowledge a team supplies, such as process documents, glossaries, FAQs, guidelines and chosen Boards and Metrics, and users can attach spreadsheets, text and images to a chat. When a security admin allows it, the agents also search the web, optionally only on approved domains, and look inside the model first when the answer may be there.

Sourcekb.pigment.com/docs/create-custom-agentsread 2026-10-13

Within a conversation the agent keeps what it has been given, so a question asked again after a board pack is attached comes back with the pack's drivers folded into the answer. Approved models, published spaces and budget changes persist as the planning data itself. There is no sign of a memory that follows a user from one chat to the next or holds a team's conventions for later questions.

Sourceabacum.ai/watch-and-learn/how-can-i-add-context-to-abacum-intelligence-agent-conversationsread 2026-10-13

Each member keeps a private chat history with the agents, where chats can be pinned, filtered by Application and reopened to carry on from where they stopped. The Modeler remembers which mode was last used in each chat, and a saved Mission keeps an analysis so it can run again. A team that wants an agent to hold on to its conventions writes them into a Custom Agent's instructions, since chats do not build up a memory of their own.

Sourcekb.pigment.com/docs/pigment-agentsread 2026-10-13

AI Modeler shows its whole proposal, variable by variable with the logic behind each, and saves nothing until a user approves it, after refining it in plain language if needed, or rejects it. A delete needs explicit confirmation, with its downstream impact shown first. The agent can read or change only what the user could already read or change by hand, and budget requests from department heads wait as pending until finance approves or rejects them.

Sourceabacum.ai/blog/ai-modelerread 2026-10-13

In Plan & Build mode the Modeler Agent puts its action plan up for validation and waits for a go ahead at each waypoint as it works. Auto accept lets it continue without stopping, except for deletions, which a person confirms every time.

Its changes to the model are permanent and attributed in Application History to the member who ran it, and Ask mode answers without touching anything.

Connector tools set to needs approval make the agent propose a create, edit or delete in Notion or Slack and wait for the member to confirm it, and a web search that might reveal sensitive data asks for approval first.

Sourcekb.pigment.com/docs/configure-and-run-the-modeler-agentread 2026-10-13

Single sign on works with any identity provider that uses SAML or OpenID Connect, and roles are assigned by group from one screen, down to sensitive payroll and compensation data, with the AI working inside that same permission structure. Abacum holds a SOC 2 Type II report and an ISO/IEC 27001:2022 certificate issued by Armanino LLP and declares GDPR compliance. Data is encrypted at rest with AES 256 GCM and in transit over TLS 1.2 or later, and its trust center shares reports and policies on request.

Sourceabacum.ai/product/securityread 2026-10-13

Sign in runs through SAML 2.0 single sign on with SCIM provisioning for Okta, Microsoft Entra ID or any compliant identity provider, or through Google Login, with multifactor authentication and domain allowlisting on top.

Group based roles and fine grained data access rights decide who can see and change each piece of data, and the agents, Custom Agents and MCP sessions act only within the rights of the member using them, under separate permissions to use agents and to configure them. Pigment holds ISO 27001 certification and SOC 1 Type 2 and SOC 2 Type 2 reports, audited every year, offers a SOC 3 report, sits at CSA STAR level one and covers GDPR and CCPA.

An independent penetration test summary was refreshed in June 2026, and API keys created since May 2026 follow a structured format that secret scanning tools can catch.

Sourcepigment.com/securityread 2026-10-13

Any result the AI produces can be inspected from its cell, showing the data sources, formulas and assumptions behind it. Abacum keeps a history of every edit to models, spaces and dimension mappings, and asking the agent what changed in a forecast returns the changelog with previous and new values, the dimensions touched and the person behind each change, inside Abacum or through an assistant connected over MCP. Each department head can pull the same trail for their own changes.

Sourceabacum.ai/product/intelligenceread 2026-10-13

The Missions tab shows every Mission the agents have run, and each change the Modeler makes is recorded in Application History against the member who ran it, inside edit histories that go down to the block level. Agent outputs explain how they were reached, with the data and logic behind them, and the Modeler writes up each change as it goes. The Audit Trail API sends login, administration, security and application events to a SIEM, and admins follow AI credit use in near real time on the Plan & Usage and AI credits screens.

Sourcekb.pigment.com/docs/pigment-agentsread 2026-10-13

Every customer runs on Abacum's AWS environment in the European Economic Area, in Ireland and Germany, with no other region on offer. AI processing on AWS stays inside the EEA, Microsoft's AI services run in the EEA and Anthropic in the United States. Abacum's data processing terms place the primary processing in the European Union and allow transfers to the United States where the service needs them. A buyer that must keep data in its own country or cloud has no option to choose.

Sourcetrust.abacum.ai/subprocessorsread 2026-10-13

Customer data is stored in the region the customer picks, Frankfurt in Germany or Oregon in the United States, on Google Cloud across three independent zones, with backups in multiregional buckets inside the same geography. Buyers who need French or European sovereign hosting can run Pigment on S3NS PREMI3NS, a SecNumCloud 3.2 qualified cloud operated by a Thales subsidiary and shielded from jurisdiction outside Europe. Disaster recovery is rehearsed live once a year against a six hour recovery time and a 24 hour recovery point.

Sourcepigment.com/securityread 2026-10-13

AI Modeler can start from a curated library of model blueprints, so a requested P&L or headcount plan follows a proven structure from the first build. AI Spaces comes with a library of finance templates, ready prompts that build the reports finance runs each month on the customer's own model in a click. The AI Classifier, anomaly detection, AI Forecasting with backsolving and Modeler each do a separate job and work without the others.

Sourceabacum.ai/product/intelligenceread 2026-10-13

The Modeler, Analyst and Documentation Agents come ready to switch on for each Application, and each does its own job. The Modeler builds and maintains models, the Analyst analyzes and reports, and the Documentation Agent answers platform questions from Pigment's knowledge base. Custom Agents come with worked examples for user guidance and for FP&A, workforce and supply chain analysis, and the chat offers a library of suggested prompts. The Analyst comes with ready use cases by role, from monthly expense variance and EBITDA and cash checks to ARR cohorts, hiring gaps and supplier risk.

Sourcekb.pigment.com/docs/pigment-agentsread 2026-10-13

Abacum picks the models. Its AI features run through AWS inside the EEA, Microsoft's Azure AI services in the EEA and Anthropic in the United States, and no setting lets a customer choose or swap the model behind an agent. Calculations never go to a model, since Abacum's own engine performs every one. A team that prefers its own assistant can reach Abacum data from Claude, Gemini or ChatGPT, while the agents inside Abacum stay on Abacum's choice.

Sourcetrust.abacum.ai/subprocessorsread 2026-10-13

The agents run on pretrained models from AI vendors under enterprise agreements with zero data retention, and customer data is never used to train them. Which vendors and models sit behind the agents is not public, and an admin has no setting to pick or switch the model. Claude, ChatGPT or Mistral can reach Pigment through its MCP server, which puts a team's own assistant in front of Pigment data while Pigment's agents stay on Pigment's choice.

Sourcekb.pigment.com/docs/ai-faqread 2026-10-13

An MCP connection lets Claude, Gemini and ChatGPT query Abacum data under the same governance and pull the forecast change report. Its address, sign in method and tool list are not public, and no public API or developer reference for the platform is offered, so outside software works with Abacum through that connection, the Excel connector, SFTP and exports.

Sourceabacum.ai/product/intelligenceread 2026-10-13

The public REST API, with a Swagger reference, imports data, exports Metric, Table and List data, returns import configurations and starts data flows, under API keys that carry their owner's access rights, IP allow lists and set rate limits, beside an Audit Logs API.

The Pigment MCP server gives each Workspace its own endpoint and signs people in with their Pigment accounts. Its default tools read and analyze, and advanced mode adds tools that create Applications, build Dimensions, Metrics and Tables, write formulas, design Views and Boards and enter data. Pigment keeps Skills files and plugins for Claude Code and Cursor in a public repository.

Sourcekb.pigment.com/docs/supported-toolsread 2026-10-13

Nothing lets a customer test an agent before it reaches users, such as scored test cases or a staging step for AI changes. AI Modeler's Review mode reads a customer's model and suggests where its logic can be simplified, which checks the model, not the agent. Abacum runs its own observability and evaluation tooling for the language models behind its features, out of the customer's hands.

Sourceabacum.ai/blog/ai-modelerread 2026-10-13

A builder previews a Custom Agent and refines it before switching it on for members. Test & Deploy environments stage model changes, the Modeler's included, before they reach a production Application that can be locked, and in managed Applications the Modeler works in Ask mode only. Every answer can be traced to its data and logic. Nothing scores the agents' answers against a test set before rollout.

Sourcekb.pigment.com/docs/create-custom-agentsread 2026-10-13

Abacum's agents work through the platform's own model, its native integrations and its MCP connection, and none of them drives a browser or a desktop application. Data arrives through connectors and SFTP, and spreadsheets link through the Excel connector, with no screen automation involved.

Sourceabacum.ai/product/dataread 2026-10-13

The agents act inside Pigment's own model and reach Notion and Slack through connectors. Web search returns information to them, and none of them opens a browser, drives another application's screen or works a desktop session for a user.

Sourcekb.pigment.com/docs/pigment-agentsread 2026-10-13