Why this matters
The CRM already knows the shape of the firm
Affinity's structures match a venture firm exactly: companies, persons, opportunities, lists, list entries, fields, and saved views. A company's profile carries the stable facts, its domain and location. Its row on the portfolio list carries the firm-specific truth: ownership partner, current status, valuation, reserve strategy, reporting cadence. That row is the portfolio, as the firm actually sees it.
The honest qualifier from the research: Affinity can only provide what a firm actually captures there. Structured pipeline fields, authored notes, and Notetaker transcripts can be rich. Interaction records may contain only metadata, and email bodies are not a confirmed surface. Affinity is therefore likely to complement, not replace, PostMoney's direct company-update ingestion.
What if
Eight things a connected fund could do
Each scenario traces to an idea in the research. The chips follow the research's own ranking.
Your portfolio roster just appears
A fund designates an Affinity saved view such as "Active Portfolio Companies" or "Portfolio Review Export" as the PostMoney source. PostMoney mirrors its membership, names, domains, and the firm's own columns: owner, status, valuation, reserve strategy, reporting cadence.
No duplicate portfolio setup. Ownership, status, and internal classifications stay aligned. The Affinity user stays in control of the companies, filters, and fields.
"What changed since the last review?"
Field-value history and webhook notifications could power a concise digest: stage changes, status changes, ownership changes, changed financing expectations, new next steps, modified portfolio classifications, and companies added to or removed from the selected list. Including who changed the status or owner, and what the field contained before.
The Monday question answered before anyone opens the CRM. The review starts from what moved, not from a blank page.
Investor notes beside founder updates
A partner opens Acme and sees recent CRM notes on the company timeline alongside company-submitted updates, each attributed to its Affinity author and timestamp. A note attached directly to the company is kept distinct from one inherited through a related person or meeting.
New risks, asks, hires, and financing discussions that never became a formal update stop being invisible. The qualitative record joins the quantitative one.
Portfolio calls become evidence
Where the firm uses Affinity AI Notetaker, transcript fragments, with speaker and timestamps, could flow in as transcript evidence: revenue and growth commentary, runway and financing plans, hiring changes, sales-pipeline developments, product launches, risks and blockers, investor asks, commitments and follow-ups.
The research calls this potentially the richest Affinity source, because an ordinary portfolio call contains updates that never become structured fields. Only where the firm uses Notetaker, and only what the API user is permitted to see.
Relationship health as a layer
Relationship timestamps and interaction scores could add an engagement layer: companies not contacted recently, founders with no upcoming meeting, companies with no clear internal owner, relationships concentrated in a single departing or inactive team member, communication that has become unusually quiet or loud.
Prioritize attention across the portfolio. The research's rule: these are behavioral signals, engagement context, not proof that a company is healthy or in trouble.
The review packet writes itself
One packet per company combining the current Affinity profile, its portfolio-list fields, recent field changes, recent notes and replies, recent and upcoming meetings, available Notetaker transcripts, and relationship recency and strength, next to PostMoney's existing updates, metrics, reports, and tasks.
The research calls this the clearest way Affinity adds context without becoming the authoritative system for every PostMoney domain object.
Two systems, one set of discrepancies
PostMoney could use Affinity as a comparison source and flag mismatches in company identity or domain data, financing dates and amounts, employee counts, portfolio status, and internal owner assignments, versus facts extracted from company-submitted updates.
Cross-source data quality without pretending both systems mean the same thing. Surfaced for review, never auto-resolved, because the two records may carry different timestamps, scopes, or meanings.
Pipeline snapshots for follow-ons
PostMoney could ingest a point-in-time representation of pipeline or follow-on state: stage, amount, owner, expected timing, and the firm's custom fields for reserves, follow-on classification, and financing workflows that overlap with company review.
The follow-on conversation and the company review finally share a record. Read-only: the research does not recommend or scope two-way sync.
Where PostMoney is different
Same meeting, two different records
Affinity receives a meeting as relationship history. PostMoney reads the same meeting as part of the company's operating and financial record, next to reporting periods, tracked metrics, and extracted datapoints with provenance.
Company-submitted updates, investor-authored Affinity observations, AI Notetaker summaries, meeting transcript evidence, and structured CRM fields stay separately provenanced. Their authors, timestamps, associations, and source systems carry different trust, and they are never collapsed into one undifferentiated truth.
Highest value first
Where the research says to start
The research's recommended starting point is a designated Affinity portfolio list or saved view, enriched in this order:
- A designated portfolio list or saved view as the source of membership and structured CRM state.
- Recent notes and replies, attributed to author and timestamp.
- Recent field changes, from field-value history and webhook notifications.
- Recent and upcoming meetings, as metadata for review preparation.
- Available AI Notetaker transcripts, where the firm records them.
That would give PostMoney useful company and relationship context while preserving company-submitted updates and PostMoney's extracted metrics as separately provenanced sources. No fundamental platform limitation found in the research makes this impractical; the main determinants of value are the quality of each customer's Affinity usage, the permitted data scope, and which sources PostMoney treats as authoritative versus contextual.
Under the hood
Platform facts, as the research found them
Research current as of September 1, 2026. Full requirements and open questions are gathered on Foundations.
Surfaces
- REST API V2 at
https://api.affinity.co/v2; V1 still required for some capabilities. - V1 webhooks:
field_value.updated,note.created,list_entry.created,list_entry.deleted,organization.mergedand others. - Data Share via Snowflake secure views or Databricks Delta Sharing, Advanced and Enterprise only, refreshing every two hours.
- First-party MCP server on Scale, Advanced, and Enterprise.
- Bearer API keys act on behalf of their owning user; IP allowlists supported.
Limits
- No API allowance on Essentials.
- 100,000 calls per month on Scale and Advanced; unlimited on Enterprise. V1 and V2 share the allowance.
- 900 calls per user per minute, plus an account-level concurrency limit. Over-limit requests return HTTP 429.
- At most three webhook subscriptions per instance.
- Failed deliveries retry with exponential backoff for up to ten hours, then the subscription is disabled.
Caveats
- API results are scoped to the key owner's visibility, so a firm-wide sync can be silently incomplete.
- No webhook events for meetings, emails, calls, chat messages, transcripts, or enrichment-field changes. No generic
list_entry.updated. - Email bodies are not a confirmed surface; the email endpoint returns metadata and subjects.
- Data Share bypasses Affinity ACLs: warehouse-grade export, not a bigger API token.
- Webhook signature verification is undocumented; confirm with Affinity.
- PitchBook, Crunchbase, and some Dealroom fields are not exportable. Beta endpoints may change.
Full requirements and open questions on the Foundations page.