Ask any founder how they met their lead investor and you will almost never hear "cold email." You will hear a chain: a former colleague, who introduced an angel, who forwarded a deck to a partner they had co-invested with twice.
The industry answer to this reality has been lists — databases of names with filters. But a name is not an asset. The asset is the path: the sequence of real relationships that lets your story arrive with borrowed trust already attached.
Paths have measurable properties. Length: fewer hops preserve more trust. Strength: a six-year working relationship transfers more credibility than a conference handshake. Recency: dormant edges leak. Direction: an intro from someone the partner has backed lands differently than one from someone who once emailed them.
Consider two founders reaching the same partner. The first has a two-hop path through a board member who co-invested with that partner on a prior deal — a strong, recent, mutually-invested edge. The second has a three-hop path ending in someone who met the partner once at a conference three years ago. Both paths technically "connect," but only one of them is worth spending a favor on, and the difference is invisible unless you can see the whole graph, not just the last name in the chain.
This is where most databases quietly fail. They can tell you a person exists between you and a partner. They cannot tell you whether that person would actually pick up the phone, or whether the relationship has enough remaining trust to spend on your behalf. A name without those properties is a hypothesis, not a path.
A name is not an asset. The asset is the path — and a path you cannot explain is a path you cannot walk.
Timing compounds all of this. A path that was strong eighteen months ago can decay — the connector changed jobs, the partner moved funds, the relationship went quiet. Deals On re-walks the graph nightly precisely because trust is not a static property; it is closer to a perishable good, and a warm-path recommendation from six months ago can be actively wrong today.
There is a second-order effect worth naming: knowing your path in advance changes how you write the ask. A founder routing through a board member frames the note differently than one routing through a former teammate — the register, the length, even which metric leads. Deals On surfaces the path early enough that the intro draft can be calibrated to the actual relationship carrying it, not written in the generic voice of a cold list.
Deals On models all four properties — length, strength, recency, direction — as first-class signal, not an afterthought bolted onto a name-and-filter database. When we recommend an investor, we are really recommending a route, and we show the evidence for every hop, because a path you cannot explain is a path you cannot walk.
