A Third Of Startup Funding Now Comes From Family Offices. Most Founders Still Pitch Them Like VCs.

Roughly 31% of global startup funding now comes from family offices rather than venture funds. Five years ago that share sat in the low teens. It has close to tripled while almost nobody updated their fundraising process to match.

The mechanism is simple. About 70% of family offices now make direct private investments instead of routing everything through venture funds as limited partners. They write checks in the $500,000 to $5 million range. There is no management fee, no carry on the direct deals, no ten-year clock, and no investment committee memo that has to survive a partnership vote.

Meanwhile venture dry powder has been shrinking. Global uncommitted venture capital peaked at $743.9 billion at the end of 2023 and has since fallen 19% to $600.9 billion.

The problem is that most first-time founders find these investors by accident, at the end of a raise, when a friend of a friend mentions someone. That is an expensive way to meet a third of the market.

A single family office is the private investment arm of one very wealthy family. A multi-family office serves a handful of them. Either way, the capital is the family's own. No outside limited partners, no fund vintage, no pressure to return capital on a schedule.

A venture fund, by contrast, pools money from pensions, endowments and funds-of-funds into a fixed-life vehicle, usually ten years, with a 2% management fee and 20% of the profits. Every structural difference between the two flows from that distinction.

Scale is bigger than most founders assume. Of roughly 2,215 documented single family offices worldwide, about 1,032 invest in venture capital in some form. J.P. Morgan Private Bank's 2026 Global Family Office Report surveyed 333 of them and found an average of $1.1 billion in assets under management per family, with direct investments ranking a top-three priority for 72% of respondents. This is not a fringe pool.

Here is where founders get the strategy wrong. The headline number does not distribute evenly across stages.

At true pre-seed and seed, venture funds still dominate deal volume. Most family offices have no deal-sourcing infrastructure. They do not see hundreds of pre-product pitches a year and they do not want to. Within a family office's own direct-investment allocation, early-stage venture actually shrank from 8% to 6% while growth equity jumped from 19% to 31%, according to BNY Mellon's 2025 study.

Where they are taking real share is later. Family office participation in Series B through pre-IPO rounds rose from 18% of rounds in 2022 to 29% in the first quarter of 2026.

Read that before you build your list. If you are raising a first institutional round, a family office is realistically a participant, not the source. And 83% of these direct deals are co-investments alongside a lead, so they are usually not setting your price.

The single structural edge a family office has is decision speed. A $2 million check from a principal can clear diligence and wire in a matter of weeks because one person, or a small internal committee, decides.

A venture fund needs a partner to champion the deal, write the memo, walk it through the investment committee, run the conflict check, and get the partnership to sign off. Even at funds known for moving quickly, that takes weeks and often months. For a founder with eleven months of runway, that is the difference between closing in October and closing in January.

There is a second, quieter advantage. Permanent capital has no exit clock. A fund raised in 2019 needs liquidity by roughly 2027 whether or not that suits your company. Founders building in hardware, regulated markets or deep infrastructure tend to underrate what that patience is worth until the fund on their cap table starts asking about strategic alternatives.

This is the part that gets skipped in the enthusiasm.

Venture funds reserve 50% to 60% of committed capital for follow-on rounds in their winners, with a partner whose job includes defending the firm's pro rata. That reserve is a real asset on your cap table even though it never appears in a term sheet.

Family offices mostly do not work that way. A Dentons survey found 64% expect to make six or more direct investments over the next twelve months, but those are spread across new deals rather than concentrated as follow-on support. Many single family offices write one check and step back.

So the framing is not which source is better. It is what you need in the next twenty-four months. If the answer includes a bridge, you need at least one investor with reserves and the muscle to deploy them. The most functional cap tables in 2026 tend to have both.

If you are still deciding what your round needs to look like before you decide who to ask, Foundra walks first-time founders through round design and investor targeting in the same workflow, so the target list comes out of the plan rather than the other way around.

Family offices are hard to find on purpose. Many have no website. Most that do will not tell you what they invest in. There is no application form.

Four approaches that work:

Follow the operator. Most family wealth came from an operating business. A family that sold a logistics company understands logistics and is a poor fit for your consumer app. Identify families whose original wealth maps to your category and you have both a filter and an opening line.

Use the co-investment pattern. When a venture fund commits to your round, ask which family offices it has co-invested with in the past year. Funds keep these relationships and will frequently make the introduction because it fills their round.

Start with multi-family offices. They have staff, mandates and something resembling a process, which makes them reachable while you build direct relationships.

Show up where the principals are. Conferences in their operating sector, not startup conferences.

A venture partner is optimizing for fund returns across a portfolio where most companies fail and one carries the fund. That produces a specific pitch appetite: enormous outcomes, power-law framing, market size above everything.

A family office principal is often optimizing for something else. Capital preservation with upside. Relevance to a sector they understand. Sometimes the pleasure of backing an operator who reminds them of themselves.

Lead with the business, not the multiple. Unit economics, path to profitability and downside protection carry more weight than a $50 billion market slide. Many principals built companies that had to make money.

Be specific. How much, what it buys, what milestone it reaches, and what happens if the next round does not materialize. And do not oversell momentum. Fake scarcity is a common tactic with venture funds who expect it. A principal who has been in business for forty years usually finds it insulting.

Because family offices do not raise outside capital, founders sometimes skip vetting them. That is a mistake with a long tail.

Ask how many direct deals they have closed in the past twenty-four months. A principal with zero is not disqualifying, but you need to know before you plan a close around their timeline.

Ask whether they intend to lead or follow, and what they will do if the company needs a bridge in eighteen months. Get a real answer, not a nod.

Ask who signs. Some family offices have a professional CIO with delegated authority. Others require the patriarch or matriarch to approve personally, which adds months and can evaporate if that person's attention moves.

Ask about information expectations. Some want a monthly update. Some want quarterly calls and board observer rights that no $500,000 check justifies.

Three situations where you should keep looking.

You need signaling. A round led by a known seed fund creates a legible story for the next investor. At Series A, "who led your seed" gets asked in the first ten minutes.

You need the next check. If your plan requires a bridge or an inside round, prioritize investors with reserves and a mandate to use them.

The money is the only thing on offer and you need more. First-time founders benefit from an investor who has watched fifty companies make the same hiring mistake. Many family offices are financial participants only, which is fine as long as you know it going in.

For context on why this matters now: in the week ending September 4, the largest US rounds were dominated by AI infrastructure, with Crusoe raising $3 billion and Fluidstack $1.5 billion. In a market concentrating that heavily at the top, a second and less crowded source of capital is how a round that would otherwise stall gets to a close.

How do I know if an investor is a family office? Ask directly in the first call: "Are you investing personal or family capital, or from a fund with outside LPs?" Nobody is offended by the question, and the answer changes how you run the rest of the process.

Will a family office on my cap table hurt my Series A? No, as long as your round has a credible lead. A round funded entirely by unnamed entities with no institutional name invites questions about why no fund would price it.

Do they use standard documents? Increasingly yes at the early stage, where SAFEs are common, but not universally. Read the paper carefully and use a lawyer who has seen family office deals.

How long does the process take? Weeks rather than months when the principal is engaged. It can also go silent for a month with no explanation. Plan for both.

Should I approach them before or after I have a lead? After, in most cases. Since the large majority of these deals are co-investments, having a lead and a price makes you far easier to say yes to.

1. 31% of Deals: Family Office vs VC for Startups (Value Add VC, July 23, 2026)
2. 2026 Global Family Office Report (J.P. Morgan Private Bank)
3. How to Raise from Family Offices at Pre-Seed: Round Design, Terms, and Investor Strategy for 2026 (VC Lens)
4. 1,000+ Family Offices Investing in Venture Capital (Family Office Hub, June 2026)
5. The Week's 10 Biggest Funding Rounds: Crusoe And Fluidstack Lead Multibillion-Dollar AI Infrastructure Haul (Crunchbase News, September 4, 2026)