Affinity's core strength is its relationship graph—the way it maps who knows whom, how deep those connections run, and the interaction history that proves it. When you export from Affinity to Salesforce, Pipedrive, or any other CRM, that graph collapses. You get contacts and companies. You get notes. You lose the network scoring, the relationship depth, and the ability to see at a glance which account has the warmest introduction path. This is not a bug. It's an architectural boundary, and understanding what survives and what doesn't will save you 60 days of chaos in your next migration. What actually leaves Affinity in the export Affinity's standard CSV export includes: Contact records: Name, email, phone, title, location. These port cleanly to any CRM. Company records: Organization name, domain, industry, employee count, funding stage (if Affinity's data enrichment captured it). These also move without loss. Notes and attachments: The long-form text and any files linked to a contact or deal. These survive intact, though formatting sometimes breaks depending on your target CRM's text editor. Deal records: Deal name, amount, close date, stage, owner. The basic deal structure transfers, though custom fields require manual mapping. Custom fields: Any additional fields you created in Affinity (e.g., 'Decision Timeline', 'Budget Approval Status') will export as columns, but you must manually recreate the field structure in your destination CRM. Activity history: Call logs, email sends, meeting notes tied to contacts. Most CRMs ingest these, but the format depends on your export method and the receiving platform's parser. On paper, this looks complete. In practice, you're left with a flat, decontextualized dataset. What dies in the export—and why you can't rebuild it from the export alone Affinity's relationship graph is algorithmic. It calculates: Relationship depth: A score that reflects shared connections, interaction frequency, and the age and strength of the relationship. This is computed in real time and never stored as a field. It does not export. Connection path: The specific chain of introductions between two people (Person A knows Person B, who knows Person C). Affinity renders this visually in the graph; it's not a exportable attribute. Salesforce and Pipedrive have no native equivalent. Interaction density: How many emails, calls, and meetings happened between two contacts in the last 90 days, and the trend (rising or falling engagement). This is surfaced in Affinity's UI but doesn't export as a discrete field. Entity resolution: Affinity's backend automatically deduplicates contacts and companies across your entire workspace based on shared identifiers and fuzzy matching. When you export, you get the consolidated view, but you lose the metadata about which records Affinity merged and why. Your new CRM won't know. Relationship type: Affinity lets you tag relationships (investor, advisor, customer, reference, etc.), but these tags are optional and often incomplete. If you relied on the graph's implicit relationship signals rather than explicit tags, that context is lost. Warm introduction opportunities: Affinity highlights mutual connections who could warm-introduce you. This is a derived insight , not a stored relationship. Once you're in Salesforce or Pipedrive, you have to manually map who could introduce whom—or accept that this advantage disappears. The hard truth: You cannot rebuild relationship depth from a contact list and activity log. Affinity's scoring is probabilistic, based on your entire historical dataset. Even if you log every subsequent interaction in your new CRM with perfect discipline, you'll never reconstruct the pre-export relationship map. You can build a forward-looking relationship layer, but the historical context is gone. Affinity data loss: the survival audit Before you export, run this audit to know exactly what you're losing: Count your relationships: In Affinity, go to each contact's profile and note the relationship strength score (usually 0–100). Export to a temp sheet. In your new CRM, you won't have this metric. This gap is your biggest risk—you'll lose the ability to rank warm paths by strength. Identify orphan relationships: Search for contacts whose only connection to your business is through another contact in Affinity. For example, 'John (CFO) knows Sarah (your champion)'. Export a report of all contacts, filtered by relationship type. Any relationship that lives only in Affinity's graph—not in explicit tags or notes—is at risk of being forgotten in your new system. Check tag coverage: Run a report of all contacts with the 'Relationship Type' field (or equivalent). If fewer than 60% of your warm contacts have an explicit tag, you're relying on Affinity's implicit relationship scoring. Document the gaps. Audit mutual connections: Pull a list of your top 20 accounts. For each, list all mutual connections between your team and the account's decision-makers. Affinit