# Integration is not _the project_. The company brain reads from the systems you already run. Nothing is replaced, nothing is migrated, and no eighteen-month data programme stands between you and the first answer. It connects to your systems once. Every use case, and every AI tool your people use, then reads from the brain - under your rules, on your own servers or in your own cloud account. § 01 Start with the questions ## Start with the questions, not the systems Integration is scoped by what the brain needs to answer, not by the size of your estate. The question audit shows which questions matter first, and only the connectors those answers need get built. A brain that starts with cash does not wait for the HR system; one that starts with accounts does not wait for the payments stack. --- § 02 What it connects to ## The systems it connects to. The connectors The systems we expect to connect to most often, and how each one connects: **public API** \- a well-documented interface a connector can be built on in days; **partner or deeper work** \- a partner agreement or deeper integration, scoped per engagement. Connectors are built for your engagement, around the systems you actually run, and a system that isn't listed is one we build for, not a wall you hit. In our own brain, Google Drive is live today. ### Email, calendar & documents * Microsoft 365 · Outlook · Teams · SharePoint public API * Google Workspace · Gmail · Calendar · Drive public API * Dropbox public API * iManage partner or deeper work * NetDocuments partner or deeper work ### CRM & pipeline * Salesforce public API * HubSpot public API * Dynamics 365 public API * Pipedrive public API ### Finance & accounting * Xero public API * QuickBooks public API * Sage partner or deeper work * NetSuite partner or deeper work ### Billing & payments * Stripe public API * GoCardless public API * Open banking - account & transaction feeds partner or deeper work ### Service & support * Zendesk public API * Freshdesk public API * Jira public API * ServiceNow partner or deeper work ### Practice, case & policy systems * Clio public API * LEAP partner or deeper work * Intapp partner or deeper work * Intelliflo partner or deeper work ### Registries, screening & regulatory * Companies House public API * FCA Register public API * Sanctions lists · OFSI · UN · OFAC public API * Screening providers · adverse media & PEPs partner or deeper work * Credit bureaus partner or deeper work ### Product & analytics * Google Analytics public API * Segment public API ### People & HR * BambooHR public API * HiBob public API ### Monitoring & incidents * PagerDuty public API * Datadog public API ### Internal & enterprise * Internal databases & on-prem systems partner or deeper work * SAP partner or deeper work --- § 03 Connected once ## Connected once, read by everything Each change in your systems is picked up once, recorded in the brain against everything it touches, and shared by every use case that needs it. An ageing invoice is one record, read by the cash loop and the account loop alike. That is why the integration work shrinks as use cases accumulate, rather than growing with every tool: the second use case finds most of what it needs already in the brain, because the first one put it there. What happens to it there - linked to everything it touches, with a source and an owner - is [how the brain understands it](/knowledge-graph/). --- § 04 How your AI tools read it ## How your AI tools read it Your people ask the brain through the AI tools they already use. Claude, ChatGPT, Copilot and Gemini all connect through the open standard AI assistants use, under the same rules: your rules decide what each tool and agent can see and do, and every answer carries its source. Change assistants next year and nothing moves. The brain stays where it is, and the next tool connects the same way. --- § 05 Where it runs, and what's yours ## Where it runs, and what's yours The brain runs on your own servers or in your own cloud account, which we can set up for you. It is plain files and an index that rebuilds from them, with nothing locked into any vendor, us included. The connectors built for it are yours on [either route](/#two-ways): documented, owned by you, and yours to extend. If we run the brain for you, they come in-house with it whenever you choose. Integration is not a dependency on us; it is an asset you keep. --- § 06 What we don't claim ## What we don't claim No universal schema. No zero-effort integration. The claim is narrower, and true: the first signals flow in days, not months, because the brain starts with only what it needs. --- § 07 Where to start ## Where to start It starts with a discovery call, then the question audit, which shows what the brain needs to connect to first. **Start with a conversation.** [Book a discovery call](mailto:hello@somai.studio?subject=Book%20a%20discovery%20call), or try the [ten-minute diagnostic (opens in a new tab)](https://benchmark.somai.studio) yourself.