How do LPs evaluate a VC track record?

The thirteen dimensions, and what each one is actually testing

Use these dimensions to prepare the record. Individual LPs will weigh them differently and may request additional evidence.

• Fund-level performance. The returns investors actually experienced. ILPA's performance template guidance defines fund level as the transactions occurring between the fund and its investors, which is a narrower and more useful definition than most managers use in conversation.
• Investment-level outcomes. What each company did, which ILPA's guidance treats separately as portfolio level, based on the cash flows between the fund and its investments.
• Partner-level attribution. Identify which decisions were yours and provide the records or references that support that account.
• Team continuity. Whether the people who produced the record are the people who will make the next decisions, which is what makes a record predictive rather than historical.
• Decision rights. Whether you decided, recommended, or participated, and whether the record distinguishes the three.
• Sourcing role. Whether you found the company or joined a round somebody else found, which is a different claim about your edge.
• Stage and geography. Whether the record was produced in the market you now propose to invest in, at the stage you now propose to enter.
• Vintage. When the capital went in, since an entry environment is not a skill and allocators adjust for it.
• Realized and unrealized results. Separated, always. ILPA's guidance deliberately does not define realized or unrealized, requiring instead that a fund be clear and consistent in its own definitions and disclose the methodology surrounding them, which puts the burden on the manager to say what they mean.
• Repeatability. Whether the process that produced the outcome is described well enough that somebody could check it happening again.
• Downside experience. What you have lost, and what changed afterwards. A record with no losses in it reads as a short record rather than a good one.
• Benchmark context. What the result is being compared against, and whether that comparison is defensible.
• Reporting quality. Whether the numbers arrive in a form an allocator can reconcile, which is itself evidence about how the firm will operate.

A compelling result against a credible one

These are different properties and allocators separate them early. A compelling result is a number that looks good. A credible result is a number somebody else could reconstruct from the underlying facts, attributed to a person, produced by a process that still exists.

The practical difference shows up in what happens under a follow-up question. A compelling result gets larger when explained, because the explanation adds context that flatters it. A credible result usually gets smaller and firmer, because the explanation adds the qualifications that make it defensible. Allocators have seen both, and the second pattern is the one that survives an investment committee.

This is why candour is a performance feature rather than a courtesy. A manager who says the sample is nine investments across two years, of which two are realized and one is a write-off, has produced something an allocator can work with. A manager who says the portfolio is performing well has produced nothing.

The LP evaluation table

Four columns, one row per question you will actually be asked. Blueprint's own framework, built to be filled in before a first meeting rather than during diligence.

• Question: what did you actually decide? Evidence requested: deal-level records with your role, the date, and the decision. Common weakness: a logo wall with no roles attached. Credible response: a table naming your role per deal, with a co-decision maker who will confirm it.
• Question: how are these numbers calculated? Evidence requested: methodology, dates, and whether figures are gross or net, fund level or portfolio level. Common weakness: gross portfolio numbers presented as if they were investor returns. Credible response: a stated methodology, the same one applied everywhere, with gross and net labelled.
• Question: what is realized? Evidence requested: realized, partially realized, and unrealized separated, with valuation dates and basis for anything unrealized. Common weakness: paper marks presented as outcomes. Credible response: three columns and a note on how marks are set.
• Question: what did you lose? Evidence requested: write-offs, markdowns, and what changed afterwards. Common weakness: a record containing only winners. Credible response: the losses named first, with the process change that followed.
• Question: does this repeat? Evidence requested: the sourcing channel map and the decision process. Common weakness: a network described rather than a channel evidenced. Credible response: a funnel with conversion by channel, including channels that produced nothing.
• Question: is it yours to use? Evidence requested: permission and documentation from the prior firm. Common weakness: an assumption that leaving means taking the record. Credible response: written consent, with the presentation limits it carries stated up front.
• Question: who else was involved? Evidence requested: references, including people you did not nominate. Common weakness: a curated list. Credible response: a list that includes somebody who can describe a difficult moment.

What changes by manager type

The dimensions are constant. Which one carries the weight is not.

• Fund I with no fund record. Attribution and process dominate, because there is no fund-level number to evaluate. The evidence framework for angel, scout, operator, and SPV histories is a separate page in this library.
• Fund II. Fund-level performance enters the conversation for the first time, and the interesting question becomes consistency between what the first fund said it would do and what it did.
• Spinout. Attribution and permission dominate, and the record is often strong and constrained at the same time.
• Solo general partner. Team continuity and key-person exposure attach to every other dimension, since the record and the person are the same asset.
• Operator-led manager. Sourcing role and decision rights carry the weight, because operating credibility is assumed and selection credibility is not.
• Established manager. Benchmark context and repeatability dominate, and the record is evaluated against the manager's own prior funds as much as against peers.

Where jurisdiction changes what may be shown

How a record is presented in fundraising materials is regulated, and the rules differ by where you market.

For advisers subject to United States marketing rules, gross performance may not be presented without net performance shown with at least equal prominence, in a format designed to facilitate comparison, and calculated over the same period. ILPA's guidance restates that requirement and adds the practical consequence for portfolio-level tables: where gross performance is shown for the realized and unrealized portions of a portfolio, net does not have to be shown for those portions, but net portfolio performance for the total portfolio is still required.

Other jurisdictions regulate financial promotions and marketing communications on their own terms. The safe sequence is the same everywhere: settle the methodology, then take the presentation to counsel, then send it.

What a strong evaluation does not prove

Clear methodology, attribution and supporting records make a track record easier to assess. They do not independently verify every result or establish how the next fund will perform.

Nothing on this page states what a good venture return is, and no quartile, median, or benchmark figure appears anywhere in it. Those depend on vintage, stage, geography, and the dataset, and a number here would be read as a standard it is not.

This page is educational and general. It is not legal, tax, accounting, valuation, or investment advice. Performance methodology and its presentation may need review by counsel, an auditor, an administrator, or a valuation professional before materials are used.