Affinity's relationship graph is beautiful until you export it. Interactions disappear into a flat timeline. Organization hierarchies collapse. People records lose their deal context. What arrives in your new platform is a skeleton of what you had—contact names, dates, maybe a few notes, but the relational intelligence that made Affinity useful is gone. For wealth advisors, relationship managers, and deal-heavy sales teams, that loss is catastrophic. You're migrating platforms, not rebuilding from scratch. But the export process doesn't preserve the connective tissue. This playbook maps exactly which nine fields break, how they flatten during export, and a field-by-field recovery sequence that restores relationship depth in 60 days without manual re-entry. The nine fields that disappear on Affinity export Affinity's power lies in relational structure: people link to organizations, interactions attach to deals and relationships, and historical context stays connected. On export, that structure collapses into CSV rows. Here's what breaks: Relationship type (investor, board member, advisor, LP, etc.) – Often exported as text in a single field, losing hierarchy and role context. Interaction-to-deal mapping – Calls, emails, meetings attached to specific opportunities become floating entries with no deal link. Organization hierarchy – Parent company, subsidiary, and relationship ownership flatten into a single "org name" field. Historical interaction sequence – Affinity's timeline order is lost; timestamps may survive, but the narrative arc doesn't. Person-to-organization multiple links – A person tied to three companies becomes three separate rows or loses two relationships entirely. Custom relationship fields – Affinity's extensible relationship metadata (e.g., "first intro date," "warm intro source," "deal involvement") often has no destination field in the new platform. Deal stage at interaction time – You lose the context of what stage a deal was in when each interaction occurred. Relationship strength scoring – Affinity's engagement metrics and relationship scoring vanish. Shared team ownership – Multiple team members tied to a single relationship get flattened into a single "owner" field. Why exports flatten: the structural mismatch Affinity stores relationships as a graph: vertices (people, orgs, deals) and edges (who knows whom, who works where, what happened when). CSV and JSON exports are row-based. When you export, every interaction needs a row, every person-org link needs a row, and context gets pushed into text fields or dropped entirely. That's not a bug in Affinity's export; it's a fundamental format limitation. Most CRM import engines expect flat data: one row per contact, one row per deal, one row per activity. Restoring relationship depth requires you to manually reverse that flattening—to rebuild the edges from the rows. The export preserves the data points. It doesn't preserve the relationships between them. You have the vertices; you need to rebuild the graph. 60-day recovery sequence: the field-by-field map Days 1–7: Audit and reconcile the export Before you import anywhere, validate what you actually have: Count unique people, organizations, and deals. Does the export count match your Affinity record count? If not, some relationships were dropped during export. Check interaction volume. Affinity should export every call, email, and meeting. If the row count drops significantly, flag which interaction types were excluded. Identify multi-link records. Search your export for people with multiple organizations or deals. How many rows do they occupy? How many are missing? Map custom fields. List every custom field in Affinity. Does it have a corresponding column in the export? If not, it's gone. Check relationship type coding. If Affinity exported "investor, board member" as comma-separated text in a single cell, note it. You'll need to split and re-map these. Document everything in a spreadsheet: column name, data type, what it represents, what's missing, how to validate it. This becomes your import checklist. Days 8–21: Clean and enrich the export Raw exports are messy. Fix them before import: Deduplicate people. Affinity may have exported the same person twice under slightly different names. Cross-reference email addresses and phone numbers. Merge duplicates. Standardize organization names. "Acme Corp," "Acme Corporation," and "ACME" need to be one record. Use fuzzy matching or manual review for high-value accounts. Split multi-value relationship fields. If Affinity exported "investor, board member" in one cell, split into separate rows or create multiple relationship entries in your new platform. Reformat dates. Ensure all timestamps are in ISO 8601 format (YYYY-MM-DD HH:MM:SS). Many import tools choke on ambiguous date formats. Add missing context in notes. For interactions that lost their deal mapping, add a note like "[DEAL: Acme Series B - Original stage: Negotiation]". This becomes sea