Most B2B prospecting tools treat your contact list as a flat database: a list of names, companies, and email addresses that you search by applying filters.
A vectorized network graph treats your contact list as something fundamentally different: a relational intelligence layer where the connections between people, the strength of those connections, and the paths between them are themselves the valuable data.
This distinction sounds technical, but its implications for prospecting are significant.
What a Vectorized Network Graph Is
In a standard contact database, each person is a record with attributes: name, company, title, email. The database can tell you who works at a given company, but it can't tell you who knows whom, how well they know each other, or what path exists between two people who don't know each other directly.
A network graph changes this by representing relationships as edges between nodes. Each person in your network is a node. Each relationship — email thread, meeting, introduction — is an edge with a weight based on interaction strength.
When this graph is vectorized, each node and edge is represented as a mathematical vector, enabling:
- Similarity search: finding people whose relationship patterns suggest they'd be valuable connectors for a given target
- Path finding: identifying the shortest or strongest path between you and any target prospect
- Relationship strength scoring: quantifying how strong a given connection is based on communication frequency, recency, bidirectionality, and meeting depth
- Second-degree mapping: automatically understanding who your contacts know, not just who you know
How AskScout Uses Network Graphs
When you connect your Gmail and Google Calendar to AskScout, the system builds a vectorized network graph from your communication metadata. Each contact becomes a node. Each email exchange, meeting invitation, and introduction thread becomes a weighted edge.
When you search for a target prospect, AskScout runs a path-finding algorithm against your network graph to identify:
- Whether the prospect is in your network directly
- Which of your connections has the strongest relationship with the prospect
- What the weight of that connector relationship is — how likely they are to make an introduction if asked
- Whether any second-degree paths exist through team members' networks
The output is a ranked list of introduction paths for each prospect, ordered by relationship strength and likelihood of successful facilitation.
Why This Scales in Ways Manual Research Doesn't
The manual version of this process — thinking through your network to find connections to a specific prospect — might take 30–60 minutes per target. It's also highly error-prone: you'll forget about a colleague who worked at the prospect's company two jobs ago, or miss an advisor who met the prospect at a conference.
Graph-based analysis runs this search across your entire relationship history in seconds, never forgets a connection, and surfaces paths you'd never find manually. At scale, this means the difference between warm intro coverage on 20–30% of your target list (what most founders achieve manually) versus 60–80% (what systematic graph analysis achieves).