Fast graph + offline cache.
IndexedDB remains the local performance and cache layer rather than being removed in favor of cloud-only behavior.
Product
Do not start with a contact list.Start with what you are trying to accomplish.
RelationshipsIQ turns a professional-history import into a persistent relationship graph, then reads that graph through an objective. The result is not one universal relationship score. It is a set of explainable views for relevance, access, network capacity, relationship health, structure and next action.
01
Objective-first workspaces
What are you trying to accomplish?
The current product begins with the outcome. Thirty core objective templates accelerate setup across consulting, sales, career, founders, fractional work and network building, while the broader goal system supports fourteen canonical goals. Templates order the starting questions; they do not create a separate scoring engine or lock you into a persona.
14 canonical goals30 core templatesCustom objectiveExplicit targetsLegacy compatible
02
Opportunity Radar
Who deserves attention now?
Radar places objective-relevant relationships into Act Now, Develop, Maintain or Monitor. Strategic relevance and access remain distinct, and the objective-aware reading explains what the portfolio contains without rewriting the underlying relationship facts or changing the canonical Radar math.
Act NowDevelopMaintainMonitorObjective reading
03
People · Companies & Paths
Where is access real—and where is it only assumed?
Inspect people and company clusters, see direct relationships and bridge routes, and understand whether a named target has real relationship coverage. Objective context can make a target prominent in the presentation, but it does not inflate relationship scores or manufacture a company path that does not exist.
PeopleCompany clustersDirect accessBridge routesTarget coverage
04
Network Runway
Can this network actually carry the objective?
Runway separates relationships usable now from relationships that need warming and requirements that still have to be built. The question changes with the objective—pipeline, career, capital, partnerships, board work or another goal—while the underlying capacity engine stays the same.
Usable nowNeeds warmingMust be builtConcentration riskObjective framing
05
Network asset intelligence
What is the network made of—and where is it fragile?
Relationship Weather, Relationship Capital, Reciprocity, Pull, Advocacy, Reachability, Organizational Coverage, Dispersion, Authority Alignment, predictive intelligence and Opportunity Rooms each answer a different structural or relational question. None is treated as a substitute for all the others.
WeatherCapitalReciprocityPullAdvocacyCoverageRooms
06
Evidence + correction
What does the system know, infer, and still need you to verify?
Observed facts, inferred conclusions and user-verified relationship knowledge remain separate. Unknowns stay visible. Your corrections, suppressions and explicit context outrank inference rather than becoming cosmetic notes on top of a fixed model.
ObservedInferredUser verifiedUncertaintyProvenance
07
Network Agent
Ask the network directly.
The Agent uses the existing relationship graph and intelligence tools to plan an answer, surface the relevant people or companies, separate evidence from interpretation, and recommend an appropriate next move. It is not an excuse to turn unsupported guesses into confident prose.
EvidenceReadingRecommendationUnknownsAction plan
Persistent product state
Import once. Reopen the product without rebuilding the network.
RelationshipsIQ preserves the local performance layer in IndexedDB while securely persisting the factual graph to the cloud. A fresh device can restore the facts and recompute the existing intelligence without requiring the original LinkedIn ZIP or storing opaque precomputed scores as truth.
Identity, tenancy and factual persistence.
Authentication and cloud persistence follow tenant isolation and row-level security contracts.
Facts first, intelligence recomputed.
Restored facts feed the same intelligence producers so the app does not rely on stale saved scores.
What the product deliberately does not do
Relationship intelligence is useful because it keeps boundaries.
The system does not infer willingness to buy, hire, invest, refer or sponsor from a title alone. It does not turn endorsements into advocacy, inactivity into deterioration, CRM presence into reachability, or a named target into artificially higher relationship scores.
01
No universal value scoreA relationship can matter for one objective and not another. Strategic interpretation belongs to the workspace.
02
No hidden target inflationNaming a company changes what you inspect, not the factual strength of the relationships underneath it.
03
No certainty where the data is thinInsufficient evidence stays visible so the next move can include verification instead of pretending the answer is known.
Experience the workflow
Choose an objective and watch the same network reorganize around it.
The website demo uses synthetic data but mirrors the objective-first product experience.