How Inferred Connections Work: Relationship Intelligence Without Data Upload

How do inferred connections work?

Inferred connections use public data — shared work history, co-investments, board memberships, public interactions, and other signals — to map who knows who without requiring any data upload from users. Cabal's AI analyzes these signals to identify connections, score relationship strength, and rank warm intro paths. Users get answers from their first query, in under 30 seconds, with zero setup.

Why Inferred Connections Change Everything About Onboarding

Traditional relationship intelligence tools require a painful onboarding process before users see any value. Connect your email. Import your contacts. Link your CRM. Wait for processing. Come back in two weeks.

Inferred connections flip this model entirely:

• Zero setup time: No data upload, no CRM integration, no onboarding project required to start
• Instant value: Users see warm intro paths and connection data from their very first query
• No dependency on user behavior: Value doesn't require the team to remember to log interactions or update a system
• Progressive enrichment: When users do connect LinkedIn, email, and calendar, the intelligence gets dramatically richer — but it's not a prerequisite

What Data Sources Power Inferred Connections?

Cabal analyzes multiple categories of public signals to infer who knows who and how strong the connection is:

• Shared employment history: People who worked at the same company during overlapping time periods likely know each other. The closer the overlap in tenure and seniority, the stronger the inferred connection.
• Co-investment relationships: VCs who co-invest in the same companies, LPs who back the same funds, and founders who share investors all have inferred connections through these financial relationships.
• Board and advisory roles: Shared board memberships, advisory positions, and governance roles create strong professional connections.
• Public interactions: Conference appearances, panel discussions, co-authored content, and mutual public endorsements signal professional relationships.
• Educational overlap: Shared alma maters, especially in the same graduating class or program, create lasting professional networks.

How Connection Strength Is Scored for Inferred Connections

Not all inferred connections are equal. Cabal scores each connection based on multiple factors to help users prioritize the warmest paths.

• Signal density: More overlapping signals (worked together AND co-invested AND attended the same events) indicate a stronger connection
• Recency: Recent overlaps are weighted more heavily than older ones
• Duration: A 3-year overlap at the same company scores higher than a 3-month overlap
• Proximity: Same team or department scores higher than same company but different offices
• Reciprocity: Bidirectional signals (both people reference the connection publicly) score higher than unidirectional ones

Inferred Connections vs. Self-Reported Connections

Both types of connections have value. The key difference is coverage and effort.

• Inferred connections: Available instantly, cover the full public network, require zero effort from users. Coverage is broad but depth depends on available public signals.
• Self-reported connections (LinkedIn, email, calendar): Richer data, more accurate strength scoring, includes private relationships not visible publicly. Requires users to connect their accounts.
• Combined intelligence: When inferred and self-reported connections overlap, Cabal's confidence in the connection and its strength score increases significantly. The combination is more powerful than either source alone.

What Tools Work Without Setup?

Most relationship intelligence and networking tools require significant setup before delivering value. Cabal's inferred connections are the primary exception in this category.

• Cabal: Inferred connections from public data. Value from the first query, no setup required. AI-queryable through chat, Claude, ChatGPT (MCP), Slack, and API.
• LinkedIn: Requires an account and manual connection building. Only shows your own direct connections.
• CRM-based tools: Require CRM integration, data cleanup, and ongoing maintenance. Value is limited to data already in the CRM.
• Email-based tools: Require email account connection and weeks of scanning history. Value is limited to people you've emailed.

How Teams Compound Value on Top of Inferred Connections

Inferred connections provide the baseline. When teams layer in their own data, the intelligence compounds exponentially.

• Step 1: Inferred connections deliver immediate value — warm paths visible from the first query
• Step 2: One team member connects LinkedIn — connection data gets richer with verified first-degree connections
• Step 3: More team members connect email and calendar — relationship strength scores become more accurate with interaction data
• Step 4: The full team is connected — the organization's collective relationship intelligence is dramatically more powerful than any individual's network

Each step compounds on the previous one. The network effect means a 10-person team with all members connected has exponentially more intelligence than 10 individuals using the tool separately.

Frequently Asked Questions

How do inferred connections work?

Inferred connections analyze public data — shared work history, co-investments, board memberships, and public interactions — to map who knows who without requiring any data upload. Users get relationship intelligence from their first query with zero setup.

What tools work without setup for relationship intelligence?

Cabal delivers relationship intelligence from inferred connections without any setup. Most other tools require CRM integration, email connection, or manual data import before providing value. Cabal's inferred connections mean users see warm intro paths from their first query.

How accurate are inferred connections compared to self-reported ones?

Inferred connections provide broad coverage across public signals. Self-reported connections (LinkedIn, email, calendar) provide deeper, more accurate data including private relationships. When both overlap, confidence and accuracy increase significantly.

What data sources are used for inferred connections?

Inferred connections analyze shared employment history, co-investment relationships, board and advisory roles, conference appearances, public interactions, and educational overlap. Multiple overlapping signals increase connection strength scores.

Do inferred connections improve when I connect my own data?

Yes. When users connect LinkedIn, email, and calendar, the intelligence gets dramatically richer. Inferred connections provide the baseline, and self-reported data compounds on top of it. Each additional team member who connects multiplies the value exponentially.