A relationship is not a connection. Two people who were at the same company eight years ago, in different offices, are connected on paper. Two people who met six times last quarter are not a guess. Relationship signals are how software tells the difference, and the number and quality of them is what separates a warm path from a long shot.
Boomerang monitors more than 80 relationship signals across first-party systems and public sources, for every Super Connector in a company's network. Here is what they are, where they come from, and how they turn into a path a rep can use.
Start free and see your own paths the same day.
What counts as a relationship signal
Any observable fact that makes it more likely one person would take a call from another. They group naturally into five families, and most have several variants, because the same fact means different things at different recency and depth.
| Family | Example signals | Where it comes from |
|---|---|---|
| Shared work history | Work overlap at the same company with overlapping dates; past employers in common; board seats held together | Career histories, board and filing records |
| Direct interaction | Past meetings; email and calendar history; CRM history on the account; previous introductions made to the target | Calendars (read-only), mailbox if connected, CRM, Boomerang's own request history |
| Shared institutions | Same university, program or cohort; alumni networks | Education and profile data |
| Shared output | Co-authored papers and publications; shared patents; the same event stages and panels; public testimonials | Patent and publication records, event listings, public web |
| Public engagement | Social engagement with the target or the target account; public mentions and advocacy; portfolio overlap; mutual connections | Public social activity, investor portfolios, network data |
LinkedIn connections and mailbox access are optional enhancers, never prerequisites. A connector who connects neither still appears in the graph, through the signals the company already holds and public sources.
How signals become a path
A single signal rarely proves much. A path is graded on how many signals agree, how recent they are, and how direct they are. Boomerang grades every path into one of three strengths:
| Strength | What it means |
|---|---|
| Strong | A clear, active relationship: they worked together recently, met several times, or have a direct connection plus supporting signals. |
| Likely | Solid evidence, but older, lighter or based on fewer signals. Most requests through Likely paths still land. |
| Long Shot | A relationship may exist, but the signal is weak, such as the same company many years ago with no clear overlap. Worth trying only when nothing better exists. |
Every path also shows its relationship context: the raw signal behind it, such as "Work overlap: Dell Technologies, Jan 2016 to Jul 2021". The rep sees it when choosing a path, and the connector sees the same context when the request arrives, which makes it far easier for them to remember the relationship and decide.
Strength is about the connector's relationship with the target, not about the connector. A famous, well-connected executive can still be a Long Shot for a specific buyer.
From a strong path to the right ask
Knowing someone well is not the same as being willing to make the introduction. Boomerang separates the two:
- Path score is capability times willingness, per connector, per deal. Capability is what the signals say about the relationship. Willingness is whether the connector is likely to act on it for this particular ask. For the scoring inputs in more detail, see connector score.
- The internal hop is the person inside your company with the strongest tie to that connector, measured in meetings held and threads exchanged. A board member asked by the CEO says yes more often than a board member asked by an SDR they have never met, so the ask goes out from whoever is most likely to get a yes.
Why first-party signals matter more than public ones
Public signals find candidates. First-party signals confirm them. Calendar history in particular is one of the strongest signals available, because a meeting is a choice both people made. Boomerang reads roughly the last 24 to 36 months of meeting history, with read-only permissions, and uses it to grade who actually knows a prospect best.
This is where most false positives die. A work overlap says two people could know each other. Four meetings in the last year says they do.
Inference versus consent
Some tools build relationships purely by inference: same company, same period, similar seniority, therefore a path. That shows value in a demo before anyone connects anything, and it is a fair approach to finding candidates. Boomerang uses public signals the same way.
The difference is what happens next. Proximity is not a relationship, and nobody can vouch for a stranger. So in Boomerang, inferred signals are graded against first-party evidence, and no path is used until the person who holds it has seen the ask and agreed. You get the day-one map, and you only ever spend relationships that are real.
What connectors are shown, and what they are not
- A connector's whole network is never exposed as a directory. Paths surface when someone needs them, scoped to what that person is allowed to see.
- Calendar and mailbox permissions are read-only. Nothing is written to Google Workspace or Microsoft 365.
- Connectors set their own engagement preferences, including minimum deal value, opportunity stage and seniority, and every request is filtered against them before it reaches them.
- Connectors are never billed.
Why the number of signals matters
Fewer signals means more of your real relationships are invisible. A tool that only reads LinkedIn misses the board member who never accepts connection requests. A tool that only reads email misses the co-founder of the target's last company. The breadth of signals is what decides how many of the doors your company can already open actually show up on screen.
For how this fits the wider category, see relationship intelligence and AI agent for warm introductions.
Frequently asked questions
What are relationship signals?
They are the evidence that two people actually know each other: overlapping employment, meetings on a calendar, shared board seats, co-authored work, shared universities, and public engagement with each other. Platforms combine them to decide whether a warm path exists and how strong it is.
How many relationship signals does Boomerang track?
More than 80. They can be grouped into five families: shared work history, direct interaction, shared institutions, shared output such as patents and publications, and public engagement. They are monitored continuously for every Super Connector.
What are the strongest relationship signals?
Recent, direct interaction. Meeting history is one of the strongest signals, because a meeting is a choice both people made. A work overlap says two people could know each other; several meetings in the last year says they do.
Do connectors have to connect LinkedIn or email for their paths to appear?
No. LinkedIn and mailbox connections are optional enhancers. Connectors are mapped from signals the company already holds and from public sources, so none of them has to upload, install or agree to anything for their paths to appear.
What do Strong, Likely and Long Shot mean?
They grade the connector's relationship with the target. Strong is a clear, active relationship. Likely is solid but older or lighter evidence. Long Shot is a weak signal worth trying only when nothing better exists.
Is inferred relationship data enough to make an introduction?
It is enough to find a candidate, not to make the ask. Boomerang grades inferred signals against first-party evidence such as meeting history, and no introduction goes out until the connector has seen the request and agreed to send it.
