The audit had been prompted by a generational transition. The second generation was coming into the family office’s investment committee and had asked a straightforward question before their first meeting: could someone walk them through the investment thesis for the six direct startup investments made over the past two years?
The answer, when the investment director assembled it, was uncomfortable. There was no single documented thesis. Each investment had been made on its own terms, sourced through a different relationship, evaluated by a different combination of people, and approved through a process that ranged from a formal IC memo to a WhatsApp exchange between the family patriarch and the founder. Two of the six had information rights. Three did not have a formal shareholders’ agreement. One had been made through a SAFE with no cap. None had a documented monitoring cadence. The total committed capital was USD 4.2 million across the six investments.
The investment director spent four weeks retroactively documenting what the office believed it owned, what rights it had, and what the current operating status of each company was. In two cases, the answer to the last question required a phone call to the founder to find out. One of those calls revealed the company had pivoted substantially from the model the family office had backed.
This is not an unusual story. According to Citi’s 2025 Global Family Office Report, 70% of family offices are now engaged in direct investing, and the governance infrastructure supporting that activity varies enormously. The governance deficit is not a resource problem. It is a category problem: direct startup investment is treated as a relationship activity rather than an asset class.
Three Investment Structures Compared, a Minimum Viable Governance Framework, and the Specific Gaps to Address
Direct deals, fund commitments, and fund-of-fund structures: a comparison
Before designing a governance framework for direct investment, it helps to be clear about what direct investment requires relative to the alternatives. The choice between structures is not purely a return question. It is a governance question.
| Structure | Expected Return Profile | Governance Requirement | Liquidity Profile | Team Capacity Required |
| Direct investment (startup or growth company) | Highest potential return; highest variance; binary outcomes common; TVPI at 5–10 years post-investment; DPI often delayed or zero | Full IC process; documented thesis; legal review; portfolio monitoring; exit governance; information rights negotiation | Fully illiquid for 5–12 years; secondary market access possible but thin at early stage | Meaningful: sourcing, diligence, legal review, post-investment monitoring, relationship management with founder and co-investors |
| Fund commitment (LP in VC or PE fund) | Market rate for the strategy; diversification across portfolio; manager selection is the primary alpha driver | Manager due diligence; ongoing fund-level monitoring; ILPA-aligned reporting review; capital call and distribution management | Illiquid for fund term (8–12 years typical); secondary market more liquid than direct positions | Moderate: manager selection, periodic reporting review, capital call management, rebalancing decisions |
| Fund-of-fund structure | Diversified exposure across managers and vintages; lower variance; fee drag from two layers of management fee and carry | Manager selection and monitoring at the FoF level; review of FoF reporting; limited visibility into underlying companies | Illiquid for FoF term; typically longest liquidity profile | Low: manager selection of FoF manager; periodic review of FoF reporting |
UBS data shows that direct allocations now account for over 40% of the typical family office private equity sleeve, a sharp increase from a decade ago, driven partly by fee advantage and partly by the desire for operational involvement and alignment. PwC’s Family Office Deals Study found that 69% of family office investments in H1 2025 were club deals structured with co-investors rather than sole deals, which reflects a sensible risk management posture but does not substitute for internal governance. Participating in a club deal led by a GP with a strong process does not mean the family office’s own governance is adequate. It means the investment is better than it would be without the GP. The family office’s own governance requirements are the same regardless of whether the deal is led by an external manager or sourced independently.
A minimum viable governance framework for two to ten direct investments per year
The following framework is designed for a family office with an investment team of two to four people making two to ten direct startup or early growth-stage investments per year. It is not a full institutional framework. It is the minimum standard that separates direct investment as a managed asset class from direct investment as an informal relationship activity.
Decision authority. Every direct investment should require approval from a named investment committee with at least two non-founder family office principals plus one independent member. The independent member does not need to be a professional investor: it can be an experienced operator, a trusted family adviser, or a retired executive with relevant sector knowledge. The independent member’s function is to provide a voice that is not subject to the relationship dynamic that drove the deal source. Single-decision-maker direct investment, regardless of the amount, is a governance failure.
IC documentation requirements. Every investment requires a written IC memo before approval. The memo should contain: investment thesis in one paragraph (what the company does and why it is worth backing); market and competitive context; team assessment; use of proceeds; current and projected financials or metrics; risk factors including the three most likely reasons the investment fails; proposed terms; and the IC’s recommendation. This does not need to be a long document. A one-to-two page IC memo that addresses these eight elements is more governance-sound than a ten-page deck that addresses none of them. The memo should be filed and retained.
Portfolio monitoring cadence. Each direct investment should have a named relationship owner within the family office who is responsible for receiving and reviewing investor updates, attending board or observer meetings if applicable, and providing a quarterly one-paragraph status note to the IC. The monitoring standard should align with the information rights negotiated at investment. A family office that consistently accepts no information rights in its term negotiations has no mechanism for monitoring. Information rights, at minimum the right to receive quarterly management accounts and an annual audit, should be a standard request in every direct investment term sheet.
Exit governance. Every direct investment should have a documented exit scenario: under what conditions would the family office accept an offer, support a secondary sale of its position, or exercise drag-along rights? Exit governance is not required to be precise at the point of investment, but the absence of any documented exit framework means that every exit decision is made without a mandate, which produces inconsistent outcomes and governance disputes at exactly the moment when they are most costly.
The specific governance gaps most commonly found in family office direct investment programmes
Farrer & Co’s 2025 analysis of family office investment structures notes that family offices are increasingly professionalising, with dedicated investment, legal, and compliance functions, but that the gap between the sophistication of the investment being made and the governance structure supporting it remains a challenge particularly for direct deals in venture and early growth stages.
The five most common specific gaps are: no documented investment thesis at the portfolio level (investments are made deal by deal without reference to a stated objective or allocation target); no consistent legal review at entry (some deals go through counsel and some do not, depending on round size or relationship pressure); no information rights or only nominal ones (family offices frequently accept minority positions with no formal information rights because requesting them felt presumptuous or created friction with the founder); no documented exit framework per investment (the family office has no stated position on when or at what multiple it would sell); and no IC review cadence for existing positions (investments are made and then not formally reviewed until an event occurs that forces attention).
Citi’s 2025 survey found that family offices are retreating from seed and early-stage investing in favour of growth-stage and secondaries, partly as a result of the poor governance outcomes that early-stage investments without adequate process produced during the 2021 to 2023 period. The retreat reflects a rational response to a governance problem, but it is not the only possible response. The alternative is to maintain early-stage exposure and fix the governance.
ILPA’s 2025 reporting standards, discussed in the P3-01 context, are calibrated to institutional fund managers rather than family office direct investors. However, the underlying principle, that LPs should receive standardised, consistent, comparable information about the investments they have made, applies directly to a family office monitoring its direct portfolio. Adopting a monitoring protocol that requires each direct investment to provide quarterly updates in a consistent format, covering revenue or ARR, cash position, headcount, and a brief narrative from the founder, is the direct investment equivalent of ILPA-aligned reporting. It is achievable without a fund administrator and without a dedicated investor relations function on the company side.
The Implication
Governance investment in direct venture exposure is a return driver, not an administrative cost. The mechanism is specific.
Family offices with documented IC processes make better entry decisions because the written memo requirement forces a structured analysis that informal approval does not. Family offices with information rights receive the investor updates that allow them to support their best companies and identify problems early enough to act. Family offices with exit frameworks exercise those frameworks rather than defaulting to inertia. Investment discipline in direct markets increasingly requires a consistent diligence process, financial, legal, and operational review to a repeatable standard, in order to maintain governance clarity as direct exposure grows.
The governance investment required to achieve this standard for a two-to-ten-deal-per-year programme is not large. A one-to-two-page IC memo template, a legal review requirement at a defined investment threshold, a named relationship owner per investment, and a quarterly monitoring touchpoint are the operational requirements. Together they convert direct startup investment from a relationship-driven activity with uncertain outcomes into a managed asset class with consistent process and legible performance.
The family office that treats direct startup investment as an asset class will, over a ten-year horizon, produce better outcomes than the one that treats it as a relationship activity. Not because the deals are better. Because the decisions are.