Live now
User-provided professional-history import.
LinkedIn complete export
Connections
Messages metadata
Positions
Skills + endorsements
Recommendations
Posts / topics
Jobs + companies
Data + integration architecture
Start with the data path that works now.Add connected execution without changing the relationship truth.
The current RelationshipsIQ product is already portable across devices without turning raw professional history into a permanent cloud archive. A user-provided LinkedIn export becomes a derived relationship graph; that factual graph can be restored securely and fed back through the same intelligence engines.
Relationship intelligence
Persistent state
Local performance + secure cloud continuity.
IndexedDBFast local graph and cache layer remains intact.
Supabase AuthIdentity and secure session foundation.
Tenant persistenceFactual network state is protected by row-level security.
Fresh-device restoreFacts restore and existing intelligence recomputes.
Current privacy boundary
The raw ZIP is not the ongoing application database.
The cloud portability work is intentionally fact-centric. RelationshipsIQ persists the relationship facts required to rebuild the intelligence without uploading the original LinkedIn archive as the permanent source of truth.
01
Message bodies are not retainedCommunication metadata can support relationship analysis without preserving private message content.
02
Facts round-tripPeople, positions, recommendations, endorsements, messages metadata, jobs, companies and authority topics can restore from cloud persistence.
03
Scores recomputeThe cloud stores factual inputs; the existing intelligence engines recompute the current interpretation rather than trusting stale saved scores.
Connected execution architecture
RelationshipsIQ should decide. External systems should execute.
The integration architecture has a deliberate boundary: RelationshipsIQ owns relationship evidence, objective-aware interpretation and recommendation. CRM and workflow systems own operational execution after the user approves the move.
RelationshipsIQ
Who matters? Why now? What does the evidence support?
Radar · Runway · Companies & Paths · Weather · Capital · Rooms · Reciprocity · predictive intelligence · Agent.
01 · Analyze
Relationship graph and objective-aware intelligence.
RIQ02 · Approve
The user decides whether the recommended action is appropriate.
YOU03 · Execute
CRM / workflow provider carries the approved operational step.
EXT04 · Learn
Normalized outcome evidence can return without directly mutating scores.
FACTGoHighLevel
The primary execution architecture is being built around GoHighLevel.
The target design is multi-tenant, server-side for secrets, idempotent for writes and human-approved for relationship actions. The website does not claim the closed-loop connector is generally available before that architecture is certified.
Provider boundary
No UI-direct provider calls. Integrations sit behind a provider architecture so relationship intelligence remains independent of one CRM.
Identity resolution
Confirmed, probable, ambiguous, unmatched and conflict states are kept distinct; name-only matching does not silently merge people.
Human approval
Act Now is not permission to auto-send. Recommendations, drafts, approval and execution remain separate stages.
Idempotent writes
External actions must avoid duplicate blind retries and classify failures explicitly.
Outcome normalization
External events become evidence facts; they do not directly rewrite Pull, Advocacy, Capital or other intelligence.
Tenant isolation
Disconnecting an integration must not destroy the RelationshipsIQ graph or cross tenant boundaries.
What we are not claiming
No fictional connector grid.
Google Contacts, Outlook, HubSpot, Salesforce, Monday.com and other systems are reasonable future provider targets, but they are not represented here as generally available live connectors until the product actually supports and certifies them.
Need another source?
Bring the workflow problem, not a logo checklist.
We can evaluate whether a data source materially improves relationship evidence and how it should enter the provider boundary without weakening identity or provenance.