Family offices allocate to private equity at rates that have risen consistently over the past decade, and the structural logic behind that migration is not difficult to identify. Public-market liquidity, once understood as a feature, has come to function in many portfolios as a constraint, forcing mark-to-market discipline on capital whose natural duration is long, and compressing the risk premium available to investors willing to hold through illiquidity cycles. According to the UBS Global Family Office Report 2024, private equity now represents the single largest alternative allocation class across surveyed family offices globally, with a median target allocation of approximately 29 percent of total assets. The trend is directionally consistent across single-family and multi-family office structures alike. What varies is not the destination but the architecture of the journey, the set of decisions about vehicle type, manager selection, governance requirements, and distribution mechanics that determine whether a PE allocation delivers on its mandate or merely absorbs capital into an illiquid position with uncertain characteristics.
The purpose of this article is to examine that architecture in detail. It begins with the structural difference between a family office and a pension or endowment, because that difference determines which allocation approaches are available and which create structural mismatches. It then maps the three primary vectors through which PE exposure is obtained, identifies the failure mode most common to under-resourced direct investors, describes the co-investment architecture that resolves the scale problem without importing infrastructure costs, and articulates the manager-selection criteria that separate durable from fragile commitments. It concludes with a direct argument about how tenet-encoded fund structures resolve multiple family office problems simultaneously, a claim that the architecture of The SAVI Capital Model makes tractable by design.
The Family Office Mandate Is Structural
Institutional allocators, pension funds, endowments, sovereign wealth funds, operate under mandates that are externally accountable in ways that family office capital is not. A pension fund's investment policy statement is a legal instrument governing the disposition of beneficiary assets. Its allocation to private equity is calibrated against a liability schedule, a funding ratio, and actuarial assumptions about longevity and contribution rates. The investment horizon is notional, it extends across beneficiary generations, but the accountability cycle is quarterly, the governance structure is a board of trustees, and the political sensitivity of underperformance is real. These constraints produce a risk-management posture that is structurally conservative: diversified by manager, vintage, strategy, and geography, with fee and terms negotiations informed by ILPA Principles and standardized LP protections.
A family office operates under a categorically different mandate. The capital is proprietary, the accountability is internal, the governance is the principal family or investment committee, and the time horizon is genuinely multi-generational rather than notional. The Campden Wealth Global Family Office Report 2024 identifies preservation of family wealth across generations as the primary mandate for the majority of surveyed offices, ahead of absolute return maximization. That preservation mandate does not produce the same allocation architecture as an endowment's return-maximization mandate. It produces a different hierarchy of risk, one in which governance compatibility, distribution predictability, and manager alignment carry weight that a pension fund, insulated by actuarial cushion and political cover, can afford to subordinate to return percentiles.
This structural difference is not a deficiency in family office governance. It is the parameter that determines which private equity approaches are fit for purpose and which generate friction. A family office that allocates to private equity using a framework designed for an endowment, chasing the same managers, in the same vehicles, on the same terms, is applying a structurally mismatched toolkit. The endowment has a dedicated PE team, decades of manager relationships, co-investment infrastructure, and legal resources sufficient to negotiate bespoke terms. The family office that lacks those inputs but imports the endowment's allocation framework absorbs the endowment's exposure without the endowment's negotiating position. Understanding that mismatch is the precondition for designing an allocation that actually fits the mandate.
Three Allocation Vectors
Private equity exposure for a family office arrives through three primary vehicles, each with a distinct risk-return-governance profile and a distinct set of requirements for successful execution. The three are fund commitments, direct investments, and co-investments. They are not mutually exclusive, and most sophisticated family office PE programs use all three in combination, but they are governed by different logics, and conflating them is a common source of portfolio construction error.
A fund commitment is a limited partnership interest in a closed-end vehicle managed by a general partner. The LP commits capital to a fund, the GP deploys that capital into portfolio companies over an investment period, and distributions flow back to LPs as investments are realized. The LP's role is largely passive, due diligence happens at the manager-selection stage, and thereafter the operating decisions belong to the GP. The primary inputs the LP controls are manager selection, vintage-year diversification, and commitment size relative to the fund. The primary risks are manager selection risk, vintage-year concentration risk, blind-pool risk (the LP commits before portfolio companies are identified), and the structural risk that the GP's incentives, once the carry clock is running, diverge from the LP's long-term return interest. According to Preqin's family office private equity research, fund commitments remain the dominant vehicle for family office PE exposure precisely because they impose the least operational demand, they do not require a deal team, a valuation capability, or portfolio company access. They require only manager due diligence and LP monitoring, capabilities that most offices with even modest investment infrastructure already possess.
Direct investments are equity or equity-linked positions taken by the family office itself, without the intermediation of a fund. The family office sources the opportunity, conducts its own diligence, structures the investment, and manages the ongoing relationship with the portfolio company. The potential advantages are significant: no management fee on deployed capital, no carried interest on realized gains, full visibility into the investment, and the ability to customize terms. The analytical case for direct investing, as documented in research from the Harvard Business School and corroborated by Family Office Exchange data, rests on the fee savings compounding into meaningful return advantage over long holding periods. The practical prerequisites, however, are substantial, deal flow, sector expertise, legal structuring capability, and board representation capacity. Those prerequisites represent a fixed cost that only justifies itself at deployment volumes that most family offices do not reach.
Co-investments occupy the architecture between fund commitments and direct investments. A co-investment occurs when a GP offers LP investors the opportunity to invest alongside a fund in a specific portfolio company, outside the fund structure, typically at reduced or zero management fee and with limited or no carried interest on the co-invest tranche. The LP exercises judgment about which co-invest opportunities to accept, and therefore exercises selection skill, but does so within a deal already sourced, structured, and underwritten by the GP. For a discussion of the growth equity structural logic that underpins the kinds of co-invest opportunities most suitable to long-horizon family capital, see The Growth Equity Thesis. The co-investment vector, properly structured, allows a family office to increase its exposure to a manager's highest-conviction positions, compress its blended fee load, and develop deal-evaluation capability without requiring the full infrastructure of a direct-investing platform.
The Direct-Investment Trap
The appeal of eliminating the manager fee layer is straightforward. A family office paying a 1.5 to 2.0 percent management fee and 20 percent carried interest to a fund manager is paying, on a net-of-fees basis, for access to deal flow, structuring expertise, sector knowledge, and ongoing portfolio monitoring that the office would otherwise need to internalize. If the office can replicate those inputs internally, the fee elimination flows directly to net return. The NBER Working Paper 28083 on private equity returns to limited partners documents that the average PE fund, net of fees, has historically delivered a public-market equivalent of approximately 1.2x over the past three decades, a premium that is real but thinner than gross return figures suggest. The implication is that fee compression, where genuinely achievable, is a meaningful lever.
The trap is that fee compression through direct investing requires a capability set that has its own cost structure. A family office with a three-person investment team, an annual deal review capacity of perhaps fifteen to twenty targets, no proprietary sourcing network, and legal counsel billing at commercial rates cannot replicate the inputs a specialized PE firm delivers, and cannot spread those fixed costs across the deployment volume required to make the unit economics work. The result is a common pattern in single-family office direct-investing programs: concentration in a small number of deals, disproportionate exposure to deals that arrived through personal network rather than competitive sourcing, inadequate post-investment monitoring relative to what the positions require, and a realized return profile that the Campden Wealth data consistently shows trails well-selected fund commitments over comparable holding periods.
The direct-investment trap is not a failure of intelligence or ambition. It is a failure of unit economics. The cost of building institutional direct-investing capability, the team, the deal flow infrastructure, the legal platform, the monitoring systems, is roughly fixed regardless of the capital deployed. That fixed cost amortizes only above a deployment threshold that most family offices, even large ones, do not reach. Below that threshold, the apparent fee savings are consumed by the cost of producing deal flow and conducting diligence that a specialized manager has already embedded in the management fee. The net result is not fee elimination. It is fee substitution, frequently at a higher total cost and a lower quality of opportunity set.
The Co-Investment Architecture
The resolution to the direct-investment trap is not to abandon the ambition of active selection. It is to find a vehicle architecture that allows active selection to operate without requiring a fully internalized institutional platform. The co-investment structure accomplishes this by redistributing the fixed costs of sourcing and structuring to the GP while preserving the LP's ability to exercise judgment at the selection level.
A well-designed co-investment program for a family office has several defining characteristics. First, it is attached to a GP relationship deep enough that the family office receives genuine co-invest access, not the overflow positions that primary fund capacity could not absorb, but the high-conviction positions where the GP is actively choosing which LPs participate alongside. The quality of co-invest access is a direct function of the quality of the LP relationship, which means that co-investment architecture begins with primary fund selection, not with co-invest screening as a standalone exercise. Second, the co-invest evaluation process is conducted at speed compatible with deal timelines, typically days to a few weeks, which requires that the family office has either internalized the sector expertise necessary for rapid independent assessment or has a GP relationship transparent enough that the office's diligence is additive to rather than duplicative of the GP's underwriting. Third, the co-investment tranche is sized in proportion to the office's conviction and liquidity, not as a reflexive percentage of commitment size, which requires a process for conviction-based position sizing rather than formulaic allocation.
The aggregate result of a well-executed co-invest program is a blended fee load, across primary fund commitments and co-investments, that is materially lower than fund-only exposure while the quality of the underlying positions is equal to or, in high-conviction co-invest tranches, higher than the average fund position. Bain's Global Private Equity Report has documented the growing share of deal value flowing through co-invest structures, reflecting that both GPs and LPs have converged on the architecture as superior to the alternatives for long-duration capital. The family office that builds this co-investment architecture around one or two GP relationships of genuine depth captures most of the fee advantage of direct investing with a fraction of the infrastructure cost.
The Manager-Selection Filter
Given the primacy of GP selection in both fund commitment and co-investment contexts, the criteria by which a family office screens managers are foundational to the success of the entire PE allocation. The standard institutional screens, track record, team stability, strategy differentiation, deal flow quality, exit capability, are necessary but not sufficient for family office purposes. A family office's mandate, as described above, includes considerations that most institutional allocators can afford to rank below raw return percentiles. Those considerations resolve into four persistent screening dimensions that recur in every rigorous family office manager evaluation: alignment, governance, distribution mechanics, and long-horizon orientation.
Alignment, in this context, means more than GP commitment to the fund. It means structural congruence between how the GP earns its return and how the LP earns its return. A carried interest structure in which the GP is indifferent between a realized gain at year four and a realized gain at year eight, because both yield the same carry calculation, is structurally misaligned with an LP whose cost of capital and distribution planning are time-sensitive. Alignment at the structural level means the GP's incentive to maximize value is not truncated by an exit clock driven by fund-life conventions rather than investment maturity. Families whose capital is genuinely multi-generational have a particular sensitivity to this misalignment, because the public-equity alternative to a PE commitment is liquid; the opportunity cost of being locked into a GP whose exit behavior is driven by fund mechanics rather than value is paid in foregone optionality, not merely foregone return.
Governance screening for a family office investor goes beyond ILPA Principles compliance and reporting quality. According to the analysis in Governance Compatibility as Investment Prerequisite, the relevant question is whether the GP's governance architecture at the portfolio company level is compatible with the LP's own values and stewardship obligations. A family office with a multi-generational mandate and a reputational stake in the outcomes of its portfolio, as distinct from an anonymous institutional LP whose identity is invisible to the portfolio company, has a material interest in how its capital behaves inside the companies it owns. A GP that acquires companies and governs them in ways the family would not endorse, even while delivering adequate returns, is creating a governance liability that the financial return does not offset.
Distribution mechanics determine whether the PE allocation functions as a component of the family's liquidity architecture or as a liability against it. A family office that commits capital to a fund expecting distributions to fund consumption or next-generation capital formation events has a different requirement from a university endowment that is indifferent between distributing and reinvesting. The GP's historical distribution cadence, not just the DPI at fund maturity, but the timing and predictability of interim distributions, is a material variable for family office LPs that is routinely underweighted in institutional manager-selection frameworks. Long-horizon orientation, finally, is the quality that distinguishes a GP whose investment thesis is durably asset-building from one whose operational model is optimized for the exit, which is a distinction that the Cambridge Associates PE benchmark data, disaggregated by strategy and holding period, supports as a meaningful predictor of net-of-fees LP outcomes across vintage years and market cycles.
Why Tenet-Encoded Funds Solve Multiple Family Office Problems Simultaneously
The Four Tenets of The SAVI Capital Model, equitable profit-sharing, compensation ratio caps, verifiable stewardship, and a social-impact obligation above a 5x return threshold, are often read as an ethical framework applied to investment activity. That reading is accurate but incomplete. The Four Tenets are simultaneously an architectural specification that addresses, at the structural level, the four manager-selection criteria that family offices identify as most persistently difficult to satisfy.
Equitable profit-sharing at the 50/50 distribution architecture resolves the alignment problem. When the GP and LP share realized returns at economic parity, as distinct from the conventional two-and-twenty arrangement in which the GP extracts a disproportionate share of value at the carry layer, the GP's incentive to maximize value is not truncated by a structure that pays the same carry regardless of holding period. The SAVI Capital Partners mandate encodes this distribution architecture as a structural condition, not a discretionary policy. For a family office evaluating a GP through the alignment lens, the structural encoding of equitable profit-sharing is a materially different kind of commitment than a GP's verbal assurance of long-horizon orientation.
Compensation ratio caps address the governance question from the inside of the portfolio company rather than from the LP monitoring layer. A GP that encodes compensation ratio limits into the operating agreement of portfolio companies is embedding a governance standard that persists regardless of which LP happens to be on the investor list in any given fund vintage. This is a structurally different governance posture from a GP that governs portfolio companies according to conventional market practices and reports compliance to LPs periodically. The family office that has a reputational stake in the behavior of its capital inside portfolio companies benefits from a governance architecture that does not depend on LP vigilance to enforce, it is encoded into the legal and operational structure of the investment itself. Verifiable stewardship, as the third tenet, makes this governance architecture legible rather than asserted: the stewardship standard is defined, the measurement methodology is specified, and the reporting framework produces data against which compliance can be independently assessed rather than taken on GP representation.
The social-impact obligation of the fourth tenet, triggered above a 5x return threshold, addresses a structural question that has become increasingly relevant to family office governance committees: whether capital at scale has obligations beyond financial return to the LP. The mechanism is described in detail in The Tenet 4 Mechanism. For a family office operating under a multi-generational stewardship mandate, a mandate that by definition extends beyond the current generation's financial optimization, the encoding of a social-impact obligation above the return threshold is an architectural alignment with the family's own institutional values rather than a charitable concession from financial return. It is, structurally, a solution to the governance compatibility problem that the family office's own mandate creates.
The two investment platforms through which The SAVI Group deploys this architecture, SAVI Capital Partners for conventional PE structures and Alitheia Capital Partners for tokenized exposure, are accessible to Qualified Purchasers through the firm's investment platforms. The common architecture across both platforms is the Four Tenets, applied structurally rather than aspirationally, which is the property that makes the platforms suitable to family office capital whose mandate includes governance compatibility as a prerequisite, not as a secondary filter applied after financial criteria have been satisfied.
Bringing the Family Capital to the Right Door
The analysis in this article converges on a single practical conclusion: the family office that approaches private equity allocation as a framework question, structuring the vectors, selecting the managers, and matching the governance architecture to the mandate, will consistently outperform the family office that approaches PE allocation as a product question, selecting from whatever managers happen to be marketing at the moment. The framework question is harder and slower. It requires the office to be explicit about its mandate, articulate about its governance requirements, and disciplined about the manager-selection criteria that distinguish a durable institutional relationship from a transactional LP-GP arrangement.
For family offices whose allocation framework has reached the stage at which manager-selection criteria are well-defined and governance compatibility is a primary screen, the institutional inquiry process is the appropriate next step. The Family Office Exchange and Campden Wealth research consistently document that the family offices achieving the most consistent PE outcomes are those with the most structured approach to initial manager engagement, one that begins with a governance and alignment conversation before a financial data-room conversation, rather than the reverse. That sequence is not procedural. It is the practical application of the principle that a misaligned but financially attractive GP is a worse long-horizon outcome than a well-aligned GP whose returns are competitive rather than exceptional.
The SAVI Group has served Qualified Purchasers since 2002, and the institutional inquiry process the firm has developed over that period is designed for exactly this stage of the family office allocation conversation: the stage at which a family knows what it is looking for, has the mandate clarity to evaluate governance architecture seriously, and is seeking a manager relationship rather than a fund product. Family offices that have reached that stage are invited to begin the inquiry process at the firm's inquiry portal. The conversation begins there and proceeds at the pace the allocation decision requires, which is, in the architecture of long-horizon capital, whatever pace the decision itself warrants.