Every private capital firm has one. The partner whose relationships are, in practice, a meaningful share of the franchise. They know which allocator is genuinely in market rather than politely taking meetings, which consultant will actually return a call, which intermediary owes them a look at the next process, and which family office changed its allocation policy last spring because somebody mentioned it over dinner.
When that person leaves, a portion of the firm's relationship capital leaves with them. Everyone involved has agreed to treat this as a fact of nature.
It is not a fact of nature. It is a data capture problem that was uneconomic to solve, and it has stopped being uneconomic. That should be good news for most firms. In our experience it is received as something closer to a threat, and the people who react that way are not being irrational.
Relationship capital is a personal asset that the firm has been booking as its own
The economics here have always been slightly dishonest, in both directions. The firm pays for the relationships: the travel budget, the brand that gets the meeting, the track record that makes the conversation worth having, the deal flow that gives the connector something to talk about. It then books those relationships as a firm asset in every fundraising deck it produces.
The individual, meanwhile, holds the asset in a form the firm cannot inspect, cannot value and cannot transfer. That asymmetry is not incidental to how senior people are paid and promoted in this sector. It is one of the main sources of bargaining power a rainmaker has, and it is why a departing partner can credibly threaten to take a book of LPs with them.
Most firms have nominally addressed this with a CRM. The CRM records that a meeting happened, roughly when, and with whom. It almost never records what was said, what was inferred, what was promised, or the read on whether the counterparty was serious. The system captures the metadata and loses the asset.
The CRM is empty because filling it costs the wrong person
We have watched a great many firms attempt to fix this by exhortation. Nobody has succeeded. Manual entry imposes a real cost on the person entering the data and delivers the benefit to someone else, quite possibly to a colleague competing with them for the same carry pool. Dressing that up as a discipline problem misreads it. The behaviour is a rational response to the incentive, and it has been stable for thirty years.
What changes the calculus is that capture no longer requires anyone to type. The raw material already exists inside the firm: email threads, calendar histories, call recordings where policy permits them, meeting notes taken in whatever half-structured form each person prefers, LP portal activity, the fundraising pipeline as it actually moved rather than as it was reported. Systems can now read this material, resolve who is who across it, and construct a picture of the firm's relationships that nobody had to sit down and write.
That removes the excuse. It does not remove the objection, which is a different thing and deserves to be taken seriously.
What a complete relationship record actually tells you is uncomfortable
The naive version of this is a better address book. The useful version is considerably more pointed, because a complete interaction history does not just record who knows whom. It measures the relationship.
Reciprocity becomes visible. So does responsiveness: which contacts reply within a day and which take a fortnight, and whether that has changed over two years. So does concentration, which is the one that tends to cause the argument. On an illustrative mid-market GP with a dozen investment and business development professionals and perhaps 400 meaningful LP, consultant and intermediary contacts, it would not surprise us to find that a third of the genuinely warm relationships route through two individuals. That figure is modelled rather than observed, but the shape of it is familiar to anyone who has run a fundraise and privately worried about what happens if one particular person gets a better offer.
Once that concentration is on a page, it becomes a governance fact rather than a hunch. Succession planning stops being a conversation about titles and becomes a conversation about specific named coverage gaps. Compensation committees acquire evidence about origination that they previously had to take on trust from the person claiming it. Neither of those conversations is comfortable, and both of them are improvements.
The resistance is rational and should be negotiated, not overridden
The senior partner who resists this is usually characterised internally as protecting their patch. Sometimes that is exactly what is happening. But the substantive objections are sound, and a firm that steamrollers them will get a system that is technically complete and practically worthless, because people will simply move the important conversations somewhere it cannot see.
Start with the honest one. Making an individual's network legible transfers value from that individual to the institution. That is the entire point of the exercise. Doing it without adjusting how origination is recognised in compensation is a unilateral change to the terms on which senior people agreed to work, and they will read it exactly that way.
Then the ones that are not about self-interest at all. A relationship is not a data record, and a great deal of what a good connector knows is confidential in a way that has nothing to do with regulation. An LP who tells you over dinner that their CIO is leaving has told a person, not a platform. Contacts have privacy expectations and, in several jurisdictions, enforceable rights over what is held about them. And a system that scores relationships will sometimes be wrong in ways that damage them, because a contact who has gone quiet for six months may be dealing with a health problem rather than cooling on your fund. A firm that acts mechanically on that signal will send exactly the wrong message at exactly the wrong time, and the partner who warned against it will be right in a way that sets the whole programme back a year.
This is a governance decision that happens to require technology
The build is the straightforward part. The decisions that determine whether it works are all policy decisions, and they are better made before anything is switched on than retrofitted after the first complaint.
A firm needs to settle what is captured and what is deliberately not. It needs to decide who can see what, because full transparency across a partnership is a different organisation from one where each person sees their own coverage plus an aggregate view. It needs a position on whether private notes remain private, and we would generally argue they should, because the alternative is that the valuable observations stop being written down anywhere. It needs retention periods and a defensible answer for a contact who asks what is held about them. And it needs to say out loud how origination credit will be assessed now that the evidence exists, rather than allowing everyone to discover the new rules by inference.
Our own bias, for what it is worth, is towards capturing less than is technically possible and being explicit about the boundary. A narrower record that senior people trust and use is worth considerably more than a comprehensive one they route around.
Firms that treat this as an IT project deliver a database. Firms that treat it as a change to how relationship capital is owned, valued and paid for deliver something that people actually feed. The difference is not in the software.
The connector becomes more valuable, not less
The fear sitting underneath all of this is that the best-connected person in the firm is being made replaceable. We think the opposite is closer to the truth. What the system replaces is recall, and recall was never the scarce input. Judgement was: knowing which of 400 warm contacts to call this week, what to say, and what not to say. That judgement gets better when it is applied to a complete picture rather than to whatever the individual happens to remember about a relationship they last touched in March.
What the firm gets is continuity. A junior person walks into a first meeting already knowing the history, the last three commitments, the unresolved question from two years ago and the fact that this investor cares about co-investment rights more than fees. A departure becomes a loss of judgement and warmth, which is serious, rather than a loss of institutional memory, which is avoidable. The best-connected person in your firm becomes a system, and the best connector in your firm becomes the person who knows what to do with it.
