Migrate from Jarvis CRM to Twenty CRM

Jarvis Contact records map to Twenty People. We extract via FileMaker export including contact name, email, phone, address, job title, and owner assignment. Custom contact fields (deployment-specific) map to Twenty custom fields that we create in Settings before import. The Jarvis contact-to-company relationship (Contact linked to Company via FileMaker relationship) resolves to the People.company field during Twenty import by matching Company foreign keys.

Jarvis Company records map to Twenty Companies. We export the full Company table including company name, domain, industry, address, phone, and any custom company fields. Company is the first object imported into Twenty because People, Opportunities, and Tasks reference it via lookup. We use the company name as the dedupe key during import to prevent duplicate Companies.

Jarvis Opportunity records map to Twenty Opportunities with stage, amount, expected close date, probability, and owner preserved. Pipeline stage names from Jarvis (which vary per deployment) map to Twenty Opportunity.stage values. We create the stage options in Twenty Settings before import. Opportunities link to People (the primary contact) and Companies via Twenty's Opportunity.company and Opportunity.person foreign keys.

Jarvis Projects map to a Twenty Custom Object named Projects. We extract project records including name, description, start date, end date, status, and budget. Task structures within projects export from FileMaker as a separate table; we flatten these into Twenty Tasks linked to the Project custom object via a lookup field. Gantt chart layout data exports as metadata but does not map to a native Twenty format; we store Gantt layout notes in a custom field for admin reference.

Jarvis time entries (billable and non-billable hours linked to projects, contacts, or vendors) map to a Twenty Custom Object named Time Entries. We extract date, duration, description, billable flag, and the linked project and contact references. Because Twenty does not have native time tracking, the custom object provides a data housing layer; billing workflow and time approval require manual rebuild in Twenty Settings or a separate time-tracking tool.

Jarvis vendor records (part of the ERP module) map to a Twenty Custom Object named Vendors. We extract vendor name, contact information, payment terms, and any custom fields. QuickBooks Online vendor data that lives in Jarvis through the native integration does not migrate through FileMaker export; the customer must re-export from QuickBooks Online directly if vendor history is needed in Twenty.

Jarvis product and service catalog records map to Twenty Products. We extract item names, descriptions, SKUs, unit prices, and any custom product fields. Product-to-opportunity associations from Jarvis map to the Opportunity.items relationship in Twenty. The Jarvis product catalog export uses FileMaker table exports; we align column headers to Twenty Product field names during the transform phase.

Jarvis task flows and activity records (follow-ups, assignments, meeting logs) map to Twenty Tasks. We extract activity type, title, description, due date, completion status, owner, and linked record references. Task types from Jarvis (call, email, meeting, note) map to Twenty Task.type values. Historical timestamps (created date, modified date) preserve as Task.createdAt and Task.updatedAt to maintain audit trails.

Jarvis user records and owner assignments export from the FileMaker ACL and record-level ownership fields. We map each Jarvis user to a corresponding Twenty Member by email address. Twenty Members must be invited and accept their invitations before we import any records with owner assignments; otherwise owner references cannot resolve. Users without a matching Twenty Member go to a reconciliation queue for the customer's admin to provision.

Every Jarvis deployment has custom fields on Contacts, Companies, Opportunities, and Projects that do not exist in a standard FileMaker schema. We identify all custom properties during the mandatory schema audit, then create matching custom fields in Twenty Settings under the relevant object before any data import. Custom fields must exist before CSV import because the CSV import creates records, not fields. We flag any custom properties that have no equivalent Twenty field type for admin decision during scoping.

File attachments stored within the FileMaker instance can be exported, but attachment storage format and location vary by deployment. We identify attachment storage paths during scoping and include them in the scope only if the FileMaker host grants access. Twenty does not support bulk file attachment import via CSV; we document the file locations and naming conventions so the customer's admin can re-upload them manually or via a separate file migration tool. Attachments linked to specific records via FileMaker relationships require manual re-attachment in Twenty after import.

Jarvis campaign and contact group records export as basic CRM data (campaign name, type, status, and group membership lists). We map these to a Twenty Custom Object named Campaigns with a many-to-many relationship to People for group membership. Because Jarvis does not have a native marketing automation engine, campaign data is typically basic (names and tags rather than engagement metrics); we preserve what exists and note that email campaign sending capability is not a Twenty feature in the free or standard tiers.