What LPs are already evaluating that a track record cannot answer

The reasonable objection

Scoring a venture firm is presumptuous on its face. Firms are built by people who know their market better than any model does, returns depend on factors no configuration captures, and a number attached to a partnership invites exactly the kind of false precision this industry already has too much of.

The answer is not that the model knows better. It is that a structured evaluation of how a firm is built is already happening, on the other side of the table, and a General Partner is usually meeting it unprepared.

LPs are already running a structured evaluation

V7 Labs, writing on LP due diligence, describes institutional allocators evaluating managers across three organized workstreams: investment performance, organizational capacity, and operational infrastructure. The process typically runs four to twelve weeks, and endowments can take six to eighteen months.

That is not a conversation about a track record. It is a structured assessment of a firm as an institution, conducted over months, by people who do it professionally and repeatedly. Whatever one thinks about scoring firms, firms are being assessed against a framework already. The question is only whether the General Partner has seen the framework before the meeting.

The question a configuration reading is built for

The most useful thing in the V7 Labs piece is a question that is almost purely structural, and it is one no track record can answer.

Their reasoning is that scale changes the problem rather than its size: deal sizes rise, competitive dynamics change, and the sector coverage needed to source enough qualified investments at a larger fund is different in kind rather than in degree. LPs respond by asking how the strategy adapts at scale, what the maximum investment size will be, and how that compares with the largest deal in the previous fund.

Every one of those is a question about configuration rather than about performance. It is asking whether the cheque, the ownership, the portfolio count, the scope, and the team still describe one workable firm at the new size. That is precisely what the three engines read, and it is why a returns history, however good, does not answer it.

The arithmetic LPs already run

The second piece of evidence is smaller and more telling. GoingVC reports that Michael Kim's 60x rule has become standard LP shorthand: a $100 million fund targeting 5 percent ownership needs to reach roughly $6 billion in cumulative exit value to return 3x.

It is a back-of-envelope calculation an allocator can run in a meeting from three numbers a General Partner has already stated. And it is a configuration reading in every respect: fund size against ownership target against a return objective, checked for whether they describe something achievable.

The same source reports Samir Kaji cautioning managers that fund size should be set by what a firm can credibly win rather than by what it thinks it should raise, and observing that a $40 million focused seed fund can be more credible than a $150 million generalist one. That is a judgment about coherence between size, scope, and capability, made without reference to any track record at all.

What this argument does not claim

It does not claim the Architecture Score replicates what an LP computes. It does not. The 60x rule is one relationship between three variables; the model reads many more, weights them differently, and arrives at something an allocator would not recognize as their own arithmetic.

Nor does it claim that a good Architecture Score will satisfy an LP. Diligence covers references, operational infrastructure, fund administration, team dynamics, and a dozen things a configuration cannot express, and a firm can read well here and fail diligence comprehensively.

The claim is narrower. Firms are already evaluated structurally. Structural evaluation is legible in advance, unlike returns. And a firm that has read its own configuration before someone else does has the option to fix what it finds or to explain it, which is a better position than discovering it in week six of a diligence process.

The narrower version of the claim

The model evaluates how a firm is built. It does not predict how the firm will perform, does not rate it, and does not decide whether it is worth backing.

What it offers is that the structural half of the question, the half an LP will ask and a track record cannot answer, is available to a General Partner before the conversation rather than during it. Why no one can honestly tell you what your fund will return covers the reasoning behind that boundary in full.

Sources

• LP Due Diligence on GPsV7 Labs, August 2026The three diligence workstreams, the timelines, and the direct treatment of whether a strategy transfers at a larger fund size.
• The 12 Signals LPs Use to Spot Top Emerging ManagersGoingVCMichael Kim's 60x rule as standard LP shorthand, and Samir Kaji on right-sizing a fund to what a manager can credibly win.