Entity model: investors (grid) vs people (contacts); fix double-count (0.1.0:48)

Per Grant's clarification of the real data model:
- Investor entities come from the fundraising grid, one per row, all labeled
  "investor" (drops the confusing lp/organization split). Grid is source of truth.
- People come ONLY from the contacts table. The grid's contacts (fundraising_
  contacts) are matched to a contact-person and recorded as member_of links to
  their investor, instead of creating duplicate person entities. This fixes the
  ~doubled people count (people now ≈ contacts, not contacts + grid contacts).
- System Status cards: Investors / People (resolved) / Contacts in CRM / Grid
  contacts, so resolved-vs-source is visible at a glance.

Verified on synthetic: people == contacts count (no double-count); multi-contact
investors preserved via member_of.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

backend/ingest/entity_resolution.py

@@ -143,11 +143,12 @@ def resolve_organizations(conn, merge_map=None):
 
     org_canon_by_orgid, org_canon_by_fundinv = {}, {}
     for key, g in groups.items():
-        # An org we are actively raising from (has a fundraising row) is an 'lp';
+        # Every firm group is one INVESTOR entity. The fundraising grid is the
-        # otherwise a plain 'organization'.
+        # source of truth for investor entities (each row = one investor, whether
-        kind = "lp" if g["investors"] else "organization"
+        # an institution/family-office or an individual); the organizations table
-        cid = _redirect(merge_map, _eid("lp" if kind == "lp" else "org", key))
+        # mirrors those names. So we no longer split into lp/organization.
-        _upsert_entity(conn, cid, kind, g["name"], g["email"])
+        cid = _redirect(merge_map, _eid("inv", key))
+        _upsert_entity(conn, cid, "investor", g["name"], g["email"])
         for oid in g["orgs"]:
             _link(conn, cid, "organizations", oid, key, "exact_name", 1.0)
             org_canon_by_orgid[oid] = cid
@@ -158,57 +159,77 @@ def resolve_organizations(conn, merge_map=None):
     return org_canon_by_orgid, org_canon_by_fundinv
 
 
-def resolve_people(conn, org_canon_by_orgid, org_canon_by_fundinv, merge_map=None):
+def _member_of(conn, person_id, investor_id):
-    """Merge contacts + fundraising_contacts by exact email, else exact name within
+    """Record that a person (contact) belongs to an investor entity."""
-    the same canonical org. Returns contact_id -> person canonical id (for lp_profiles)."""
+    if not investor_id or person_id == investor_id:
-    merge_map = merge_map or {}
+        return
-    # gather (model, source_id, full_name, email, org_canon)
-    people = []
-    for r in conn.execute("SELECT id, first_name, last_name, email, organization_id FROM contacts"):
-        full = f"{r['first_name'] or ''} {r['last_name'] or ''}".strip()
-        people.append(("contacts", r["id"], full, norm_email(r["email"]),
-                       org_canon_by_orgid.get(r["organization_id"])))
-    for r in conn.execute("SELECT id, full_name, email, investor_id FROM fundraising_contacts"):
-        people.append(("fundraising_contacts", r["id"], r["full_name"] or "", norm_email(r["email"]),
-                       org_canon_by_fundinv.get(r["investor_id"])))
-
-    contact_to_person = {}
-    person_meta = {}  # canonical_id -> {"org": org_canon, "last": last_norm, "name": display, "email": email}
-
-    for model, sid, full, email, org_canon in people:
-        name_norm = norm_text(full)
-        if email:
-            key = f"e|{email}"
-            match_kind, conf, match_value = "exact_email", 1.0, email
-        elif name_norm:
-            key = f"n|{name_norm}|{org_canon or ''}"
-            match_kind, conf, match_value = "name_org", 0.8, name_norm
-        else:
-            continue
-        cid = _redirect(merge_map, _eid("per", key))
-        display = full.strip() or email
-        _upsert_entity(conn, cid, "person", display, email)
-        _link(conn, cid, model, sid, match_value, match_kind, conf)
-        # Record that this contact (person) belongs to its investor/org entity, so
-        # one investor can own many contacts (e.g. a family office with several
-        # people) — and a 1-contact HNWI is just the N=1 case.
-        if org_canon and cid != org_canon:
     conn.execute("""
         INSERT INTO relationship_edges (id, src_id, dst_id, edge_type, source, strength, directed,
                                         first_seen_at, last_seen_at, created_at, updated_at)
         VALUES (?,?,?, 'member_of', 'entity_resolution', 1.0, 1, ?, ?, ?, ?)
         ON CONFLICT(src_id, dst_id, edge_type, source)
         DO UPDATE SET last_seen_at=excluded.last_seen_at, updated_at=excluded.updated_at
-            """, (str(uuid.uuid4()), cid, org_canon, _now(), _now(), _now(), _now()))
+    """, (str(uuid.uuid4()), person_id, investor_id, _now(), _now(), _now(), _now()))
-        if model == "contacts":
+
-            contact_to_person[sid] = cid
+
-        meta = person_meta.setdefault(cid, {"org": org_canon, "last": _split_name(full)[1],
+def resolve_people(conn, org_canon_by_orgid, org_canon_by_fundinv, merge_map=None):
-                                             "name": display, "email": email})
+    """People come from the CONTACTS table (one person per contact, where the
-        if org_canon and not meta["org"]:
+    emails/LinkedIn live). The fundraising grid's contacts are NOT a second set of
-            meta["org"] = org_canon
+    people — each is matched to a contact-person and recorded only as a member_of
+    edge to its investor entity (the grid's 'Contacts' column says who belongs to
+    which investor). This is what stops the double-count.
+    Returns contact_id -> person canonical id (for lp_profiles)."""
+    merge_map = merge_map or {}
+    contact_to_person = {}
+    person_meta = {}
+    by_email = {}            # norm_email -> person cid
+    by_name_inv = {}         # (name_norm, investor_canon) -> person cid
+
+    def _person(full, email, inv_canon, model, sid):
+        name_norm = norm_text(full)
+        if email:
+            key, mk, conf, mv = f"e|{email}", "exact_email", 1.0, email
+        elif name_norm:
+            key, mk, conf, mv = f"n|{name_norm}|{inv_canon or ''}", "name_org", 0.8, name_norm
+        else:
+            return None
+        cid = _redirect(merge_map, _eid("per", key))
+        _upsert_entity(conn, cid, "person", full.strip() or email, email)
+        _link(conn, cid, model, sid, mv, mk, conf)
+        if email:
+            by_email[email] = cid
+        if name_norm:
+            by_name_inv[(name_norm, inv_canon or "")] = cid
+        _member_of(conn, cid, inv_canon)
+        m = person_meta.setdefault(cid, {"org": inv_canon, "last": _split_name(full)[1],
+                                         "name": full.strip() or email, "email": email})
+        if inv_canon and not m["org"]:
+            m["org"] = inv_canon
+        return cid
+
+    # 1. People = the contacts table.
+    for r in conn.execute("SELECT id, first_name, last_name, email, organization_id FROM contacts WHERE deleted_at IS NULL"):
+        full = f"{r['first_name'] or ''} {r['last_name'] or ''}".strip()
+        cid = _person(full, norm_email(r["email"]), org_canon_by_orgid.get(r["organization_id"]), "contacts", r["id"])
+        if cid:
+            contact_to_person[r["id"]] = cid
+
+    # 2. Grid contacts are associations, not new people: match to a contact-person
+    #    (by email, else name within the same investor) and just add membership.
+    #    Only create a person when there is genuinely no matching contact.
+    for r in conn.execute("SELECT id, full_name, email, investor_id FROM fundraising_contacts"):
+        email = norm_email(r["email"])
+        name_norm = norm_text(r["full_name"] or "")
+        inv_canon = org_canon_by_fundinv.get(r["investor_id"])
+        cid = (by_email.get(email) if email else None) or by_name_inv.get((name_norm, inv_canon or ""))
+        if cid:
+            _link(conn, cid, "fundraising_contacts", r["id"], email or name_norm, "grid_assoc", 0.9)
+            _member_of(conn, cid, inv_canon)
+        else:
+            _person(r["full_name"] or "", email, inv_canon, "fundraising_contacts", r["id"])
 
     # lp_profiles -> the person entity of its contact
-    for r in conn.execute("SELECT id, contact_id FROM lp_profiles"):
+    for r in conn.execute("SELECT id, contact_id FROM lp_profiles WHERE deleted_at IS NULL"):
         cid = contact_to_person.get(r["contact_id"])
         if cid:
             _link(conn, cid, "lp_profiles", r["id"], r["contact_id"], "contact_fk", 1.0)
@@ -256,8 +277,7 @@ def run(db_path: str):
     live = "deleted_at IS NULL"
     counts = {
         "canonical_total": conn.execute(f"SELECT COUNT(*) FROM canonical_entities WHERE {live}").fetchone()[0],
-        "lp": conn.execute(f"SELECT COUNT(*) FROM canonical_entities WHERE entity_kind='lp' AND {live}").fetchone()[0],
+        "investor": conn.execute(f"SELECT COUNT(*) FROM canonical_entities WHERE entity_kind='investor' AND {live}").fetchone()[0],
-        "organization": conn.execute(f"SELECT COUNT(*) FROM canonical_entities WHERE entity_kind='organization' AND {live}").fetchone()[0],
         "person": conn.execute(f"SELECT COUNT(*) FROM canonical_entities WHERE entity_kind='person' AND {live}").fetchone()[0],
         "links": conn.execute("SELECT COUNT(*) FROM entity_links").fetchone()[0],
         "fuzzy_candidates": len(candidates),

backend/server.py

@@ -3467,8 +3467,7 @@ class CRMHandler(BaseHTTPRequestHandler):
         try:
             live = "deleted_at IS NULL"
             out['canonical_entities'] = {
-                'lp': conn.execute(f"SELECT COUNT(*) FROM canonical_entities WHERE entity_kind='lp' AND {live}").fetchone()[0],
+                'investor': conn.execute(f"SELECT COUNT(*) FROM canonical_entities WHERE entity_kind='investor' AND {live}").fetchone()[0],
-                'organization': conn.execute(f"SELECT COUNT(*) FROM canonical_entities WHERE entity_kind='organization' AND {live}").fetchone()[0],
                 'person': conn.execute(f"SELECT COUNT(*) FROM canonical_entities WHERE entity_kind='person' AND {live}").fetchone()[0],
             }
             out['entity_links'] = conn.execute("SELECT COUNT(*) FROM entity_links").fetchone()[0]

frontend/index.html

@@ -9382,20 +9382,20 @@
 
                     <div className="kpi-grid">
                         <div className="kpi-card">
-                            <div className="kpi-label">LPs</div>
+                            <div className="kpi-label">Investors</div>
-                            <div className="kpi-value">{entities.lp ?? 0}</div>
+                            <div className="kpi-value">{entities.investor ?? 0}</div>
                         </div>
                         <div className="kpi-card">
-                            <div className="kpi-label">Organizations</div>
+                            <div className="kpi-label">People (resolved)</div>
-                            <div className="kpi-value">{entities.organization ?? 0}</div>
-                        </div>
-                        <div className="kpi-card">
-                            <div className="kpi-label">People</div>
                             <div className="kpi-value">{entities.person ?? 0}</div>
                         </div>
                         <div className="kpi-card">
-                            <div className="kpi-label">Entity Links</div>
+                            <div className="kpi-label">Contacts in CRM</div>
-                            <div className="kpi-value">{data.entity_links ?? 0}</div>
+                            <div className="kpi-value">{(data.source_counts || {}).contacts ?? '—'}</div>
+                        </div>
+                        <div className="kpi-card">
+                            <div className="kpi-label">Grid contacts</div>
+                            <div className="kpi-value">{(data.source_counts || {}).fundraising_contacts ?? '—'}</div>
                         </div>
                     </div>
 

start9/0.4/startos/utils.ts

@@ -12,8 +12,9 @@ export const PACKAGE_TITLE = 'Ten31 Database'
 //   * 0.1.0:44 (Phase-0 ingest + MCP server in image; build-index action)
 //   * 0.1.0:45 (Phase-1 thesis system; dual approval; merge review; in-app index)
 //   * 0.1.0:46 (packaging fix: ship full backend so migrations run + endpoints work)
-//   * Current:        0.1.0:47 (soft-delete instead of hard-delete; source-count diagnostics)
+//   * 0.1.0:47 (soft-delete instead of hard-delete; source-count diagnostics)
-export const PACKAGE_VERSION = '0.1.0:47'
+//   * Current:        0.1.0:48 (entity model: investors vs people; fixes double-count)
+export const PACKAGE_VERSION = '0.1.0:48'
 
 export const DATA_MOUNT_PATH = '/data'
 export const WEB_PORT = 8080

start9/0.4/startos/versions/index.ts

@@ -8,8 +8,9 @@ import { v_0_1_0_44 } from './v0.1.0.44'
 import { v_0_1_0_45 } from './v0.1.0.45'
 import { v_0_1_0_46 } from './v0.1.0.46'
 import { v_0_1_0_47 } from './v0.1.0.47'
+import { v_0_1_0_48 } from './v0.1.0.48'
 
 export const versionGraph = VersionGraph.of({
-  current: v_0_1_0_47,
+  current: v_0_1_0_48,
-  other: [v_0_1_0_39, v_0_1_0_40, v_0_1_0_41, v_0_1_0_42, v_0_1_0_43, v_0_1_0_44, v_0_1_0_45, v_0_1_0_46],
+  other: [v_0_1_0_39, v_0_1_0_40, v_0_1_0_41, v_0_1_0_42, v_0_1_0_43, v_0_1_0_44, v_0_1_0_45, v_0_1_0_46, v_0_1_0_47],
 })

start9/0.4/startos/versions/v0.1.0.48.ts

@@ -0,0 +1,19 @@
+import { VersionInfo } from '@start9labs/start-sdk'
+
+// Entity-model fix. Investor entities now come from the fundraising grid (one per
+// row, all labeled "investor"; the confusing organization/lp split is gone), and
+// PEOPLE come only from the contacts table. The grid's contacts are recorded as
+// membership links to their investor rather than as duplicate people, which fixes
+// the roughly-doubled people count. Rebuild the search index after upgrading.
+export const v_0_1_0_48 = VersionInfo.of({
+  version: '0.1.0:48',
+  releaseNotes: {
+    en_US: [
+      'Fixes the entity counts: investors now come from the fundraising grid and',
+      'people come from the contacts table, with the grid\'s contacts treated as',
+      'memberships instead of duplicate people. Run Rebuild search index after',
+      'upgrading to refresh the numbers.',
+    ].join(' '),
+  },
+  migrations: { up: async () => {}, down: async () => {} },
+})