Network mapping is the process of building a queryable graph of who-knows-whom across your company's combined network. Relationship scoring quantifies how warm each connection actually is. Network mapping answers "who's connected to the CRO at Acme?" Relationship scoring answers "which of those connections will actually deliver a warm intro?" Together they replace the question "do we have a path?" with the question "which path is worth pursuing?"
What network mapping pulls together
A complete network map combines signal from at least five sources:
| Source | What it surfaces |
|---|---|
| Email metadata | Who's emailing whom, how often, in what direction. Strongest single signal. |
| Calendar events | Who's meeting whom, recency, duration. High-fidelity for working relationships. |
| CRM contacts | Customer-facing relationships, deal involvement, account history. |
| Degree-1 connections at the employee level, plus jobs-and-companies overlap. | |
| Slack / chat | DM patterns, channel co-membership, real-time interaction frequency. |
Some platforms add a sixth source: declared relationships. Investors map their portfolio. Customers map their champions. Board members map their boards. Declared signals are often the most valuable because they capture relationships that don't show up in email — long-running friendships, past colleagues, advisor connections.
The four pillars of a B2B revenue team's network
For a B2B sales team, the relevant network spans four pillars:
1. Team pillar. Your reps, managers, executives, and recently-departed alumni. Pulled from email, calendar, CRM.
2. Investor pillar. Lead and follow-on VCs, their portfolios, your board members, and your angels. Pulled from declared portfolio data plus reciprocal email.
3. Customer pillar. Champions, power users, executive sponsors, ex-customers (job-changers), and customer alumni. Pulled from CRM, product usage, and champion-tracking data.
4. Partner pillar. Resellers, OEMs, integration partners, and co-sell relationships. Pulled from partner CRM and partner manager declarations.
A network map that only includes the team pillar misses 60-80% of the warm-path graph. The customer, investor, and partner pillars are where the highest-value paths typically live.
What relationship scoring does on top
Once you have the graph, each edge (relationship) gets scored. Most platforms use a Connector Score that combines:
- Recency — when was the last interaction
- Frequency — how often they interact
- Depth — substance and duration
- Reciprocity — do both sides initiate
- Relevance — did they work in the same domain, company, or role
The result: every connection in the graph carries a numerical strength. When you search for paths to a target, the platform returns ranked paths instead of all paths.
The before-and-after
Without network mapping + relationship scoring: A rep googles the target, finds them on LinkedIn, asks their manager if they know anyone there, gets a vague answer, sends a cold email. Pipeline doesn't materialize.
With network mapping + relationship scoring: The rep searches the target in the platform. Sees 7 paths, ranked by Connector Score. Top path is through the CFO of an existing customer (who used to work at the target company). The platform's agent drafts the ask in the CFO's voice. The CFO approves with one click. Intro lands. Meeting books.
Same network. Different outcome.
How network mapping is built in practice
Modern platforms build network maps in three ways:
1. OAuth-based ingestion. Users connect Gmail, Google Calendar, LinkedIn, Slack, Salesforce. The platform pulls metadata (not content), de-dupes contacts, and builds the graph automatically.
2. Declared mapping. Customers, investors, and advisors manually add high-value relationships that don't show up in email (close friends, former colleagues, board peers).
3. Inference layer. The platform infers relationships from public data: jobs-and-companies overlap, school overlap, conference attendance, declared connections on LinkedIn.
The best platforms blend all three. Pure OAuth ingestion misses declared signals. Pure declared mapping doesn't scale. Pure inference is noisy.
What network mapping is NOT
Network mapping is often confused with three adjacent concepts:
Contact databases (ZoomInfo, Apollo, Cognism). These give you contact info for people you don't know. Network mapping shows you who in your existing network knows them.
Intent data (6sense, Bombora). These tell you which accounts are in-market. Network mapping tells you who to reach inside those accounts.
CRM (Salesforce, HubSpot). CRMs store the records of your customer relationships. Network mapping shows the graph between them and the rest of the world.
Network mapping sits on top of all three and unlocks their value.
Pitfalls
Privacy and consent. Pulling email metadata requires user OAuth and clear consent. Platforms that scrape without consent are uncomfortable to deploy.
Stale data. A network map built 18 months ago is wrong. Job changes, role changes, and dormant relationships need to be reflected. Platforms that don't refresh quietly mislead users.
One-sided declarations. If a customer says "I know the CRO at Acme," but the CRO doesn't remember the customer, the path is weaker than the declaration suggests. Best platforms cross-check declarations against email signal.
Over-trusting LinkedIn. A LinkedIn connection is a degree-1 graph edge. It says nothing about strength. Treating LinkedIn connections as warm paths is the single most common reason warm-intro programs fail.
How Boomerang does network mapping + relationship scoring
Boomerang builds the four-pillar graph (team, investors, customers, partners) by combining OAuth-based ingestion, declared mapping from each pillar's stakeholders, and inference from public data. Every edge carries a Connector Score that updates daily.
The agent uses the scored graph to:
- Surface which target accounts have warm paths worth activating today
- Rank the top 5 paths to every target person by Connector Score
- Pick the right connector when drafting an intro request
- Identify Super Connectors (the 5% of your network with disproportionate coverage)
- Flag dormant relationships worth re-warming before you need them
Bottom line
Network mapping without relationship scoring is a phone book. Relationship scoring without network mapping is theater. Combined, they're the operating system of relationship-led pipeline.
For the broader category, see warm introduction software and connector score.
