Family offices manage $5.9T globally (Campden Wealth 2024) and over 35% now make direct startup or VC investments. The ones doing it well are not running funds — they are using a layered stack of structures that each solve a different problem.
The mistake most family offices make is treating tech investing like a binary: either hire a team and run a fund, or skip it entirely. That's wrong. There are seven distinct access points with meaningfully different fee structures, control rights, concentration risk, and liquidity profiles. Here's how each one works.
The 7 Structures for Family Office Tech Investing
LP Positions in VC Funds
Best for: Diversified exposure with zero sourcing overhead
The foundation of any family office tech allocation. A $2M–$5M LP check into a top-quartile fund gives you exposure to 20–30 companies, a seat at annual LP meetings, and optionality on co-investments. Top-quartile VC funds return 3.0x+ TVPI on a 10-year J-curve. The cost is 2% annual management fees and 20% carry on profits — meaningful drag, but justified if manager selection is sharp.
Co-Investments Alongside VC Funds
Best for: Doubling down on conviction deals at zero fee drag
Co-investment rights are the single best fee arbitrage in venture. GPs offer co-invest to large LPs to build relationship stickiness and fill round allocations quickly. You invest directly into the company alongside the fund — no management fee, no carry. The catch: you need to be able to move fast (7–10 days), and you need enough internal capacity to diligence the deal independently rather than just rubber-stamping the GP's thesis.
Special Purpose Vehicles (SPVs)
Best for: High-conviction single-company bets with legal simplicity
An SPV is a single-purpose fund for one deal. Operators, scouts, and angels syndicate SPVs through platforms like AngelList and Assure. Minimum checks can be as low as $25K, making SPVs ideal for testing deal-by-deal investing before committing to a fund relationship. SPV carry is typically 10–20% with no or minimal management fees. Ideal for family offices that want direct ownership in a specific company without the complexity of a bilateral term sheet.
Secondary Market Purchases
Best for: Buying into proven companies at a discount, without waiting for a new round
VC secondaries let you buy LP fund stakes or direct company shares from sellers who need liquidity. LP stakes in 2022–2024 vintage VC funds have traded at 10–25% discounts to NAV — meaningful alpha if the underlying portfolio recovers. Direct secondary shares in late-stage private companies (Stripe, SpaceX, Anduril) trade on platforms like Forge and Hiive. Duration is typically 2–5 years vs. 10+ years for primary fund investing.
Direct Angel / Seed Investments
Best for: Highest upside, highest concentration, highest diligence burden
Writing a $250K–$2M check directly into a seed or Series A company gives you uncapped upside and full control over terms. The math can be extraordinary: a $500K seed check into a company that raises at a $50M pre-money and exits at $2B returns 20x before fees. But the hit rate is low (1–2% of seed companies reach unicorn status), and it requires sourcing infrastructure — which is why most family offices only do this selectively alongside a broader LP portfolio.
Fund-of-Funds
Best for: Maximum diversification with outsourced manager selection
Fund-of-funds vehicles (like HarbourVest, Adams Street, or Greenspring) pool capital across 15–30 VC funds. For family offices without the staff to evaluate individual GPs, a FoF is a defensible way to get market-rate VC exposure. The fee layering (1–1.5% on top of underlying 2/20) is real — expect total fee load of 3–4% annually — but for family offices that lack time or expertise for manager selection, the diversification and sourcing are worth the cost.
Evergreen / Interval Funds
Best for: VC economics without a decade-long capital lockup
Evergreen structures like Harbourvest's HVPE or Lexington's interval fund allow family offices to invest in private equity and venture with quarterly liquidity windows. They hold diversified portfolios of fund stakes and co-investments, priced at NAV with quarterly redemption rights. For family offices with liquidity needs or investment committee constraints around long-duration lockups, evergreen is the best available compromise between VC returns and private credit liquidity.
How Sophisticated Family Offices Stack These Structures
A $500M family office with a 15% tech allocation ($75M) might deploy capital across structures like this:
| Structure | Allocation | % of Tech Book |
|---|---|---|
| LP positions in 4–6 VC funds | $35M | 47% |
| Co-investments (top deals from LPs) | $15M | 20% |
| Direct seed/A deals (high conviction) | $10M | 13% |
| Secondaries (LP stakes + direct shares) | $8M | 11% |
| SPVs (opportunistic single deals) | $5M | 7% |
| Evergreen vehicle (liquidity buffer) | $2M | 3% |
| Total Tech Allocation | $75M | 100% |
The co-invest layer is particularly important — it captures upside from the best deals in an LP portfolio without the full fee drag. A family office that writes a $1M LP check and gets $500K co-invest rights in the top performer effectively halves its carry expense on that company.
The Real Barrier: Manager Access, Not Capital
The limiting factor in family office tech investing is rarely capital. It's relationship access. Top-quartile VC funds — the ones actually driving venture returns — are routinely oversubscribed. Sequoia, a16z, and Benchmark don't need your capital. Getting into their funds requires existing relationships, institutional references, or a record of being a value-add LP.
How to get into top funds
- ›Be an existing LP at a sister fund
- ›Come through a referral from a portfolio founder
- ›Offer strategic value (distribution, domain expertise)
- ›Start smaller with emerging managers, build up
Red flags in manager selection
- ›Fund size 3–4x larger than prior fund (return compression)
- ›GP track record concentrated in 1–2 outlier deals
- ›Vintage year misalignment with your liquidity needs
- ›No co-invest rights or information rights for LPs
When Running Your Own Fund Actually Makes Sense
For a small number of family offices, building a proprietary venture vehicle is the right call. That threshold is roughly:
- ✓$1B+ AUM with a dedicated investment team (3+ people)
- ✓A genuine sourcing edge — founder relationships, sector expertise, or deal flow from operating company
- ✓Desire to manage external LP capital (creating a carry business inside the family office)
- ✓Long-term commitment: fund cycles are 10–12 years and require consistent annual deployment
Below that threshold, the math rarely works. The overhead of running a fund — legal ($150K–$300K setup), annual admin ($50K–$100K), compliance, and LP reporting — consumes a disproportionate share of management fees for small vehicles. A $50M fund at 2% management fee generates $1M/year in fees, which barely covers a two-person investment team. Most family offices are better off as sophisticated LPs and co-investors until they have a compelling reason to take on fund infrastructure.
Track how institutional funds are performing on the VC Performance Dashboard and benchmark your portfolio against top-quartile returns at Fund Benchmarking.
The best family offices don't run funds — they stack access.
LP relationships → co-invest rights → selective direct deals. That stack captures most of the upside with a fraction of the infrastructure cost of running a fund.
Explore fund performance benchmarks and VC return data on the Funds Database at Value Add VC. Originally published in the Trace Cohen newsletter.