A pre-seed fund writes the first institutional check into a company, before there's revenue history, a proven acquisition channel, or a priced round from anyone else. It's a different job from seed and a completely different job from Series A. You're underwriting a person and a problem, at check sizes that only produce meaningful ownership inside a small fund, in the part of the market where price competition is lowest and non-consensus positions still exist.
That's why the stage keeps attracting first-time managers. 85% of VC Lab managers invest at pre-seed or seed. It's also why generic fund advice misfires here. Advice written for a $150MM Series A vehicle says protect ownership, reserve heavily and price against comparables. At pre-seed there are no comparables, ownership gets reset by somebody else's re-underwrite a year later, and reserves compete with the checks that build the portfolio.
What a pre-seed fund actually is, and where the stage boundary sits
Forget the round labels. Founders and investors use "pre-seed" loosely and the label moves with the market. The useful boundary isn't a dollar amount. It's what evidence exists when you commit.
What exists at pre-seed
A team, usually one or two people. A wedge, meaning a specific narrow first use case chosen out of a much larger problem. Sometimes a prototype, a design partner, a waitlist, or a founder's prior relationship with the buyer. Often a strong opinion about why incumbents can't respond. That's the whole evidence file, and you can read it in an afternoon.
What does not exist at pre-seed
Revenue history, meaning more than a few months of the same thing repeating. A repeatable acquisition channel with a known cost. Cohort retention over a meaningful window. A management team beyond the founders. Any external party who has already priced the company. If those exist you're at seed, whatever the round is called in the deck.
The practical test: at seed you're checking whether an observed pattern continues, at pre-seed you're deciding whether a pattern will ever start. Different questions, different diligence, different temperaments. More on the first in our guide to early stage diligence in emerging VC.
Where the boundary sits in practice
In practice the pre-seed round is the first priced or capped instrument a company raises, often a SAFE, at valuations low enough that a small check buys real ownership. It exists because building version one got cheap while the first institutional round got expensive. A pre-seed fund is built to be present at that moment, which is what people mean by a first check fund.
Why a pre-seed fund is structurally a small fund
This isn't a preference. It's arithmetic, and it runs in one direction. It's also why price competition is lowest here, the case we make in the first principles VC's guide to non-consensus deals.
The check size that fits the stage
$25K, $50K and $100K checks still happen constantly at pre-seed. Run one through two vehicles. Inside a $400MM fund, a $100K check is 0.025% of committed capital, and even a 100x on it returns $10MM, or 2.5% of the fund. It cannot move the number.
Inside a $10MM fund with $8MM investable, that identical $100K is 1.25% of deployed capital and the identical 100x returns the whole fund. Same company, same check, entirely different meaning. The difference isn't skill or conviction. It's the denominator.
What a large fund's minimum check does to the stage
Deployment pressure closes the door from the other side. A $400MM fund with four investing partners has to place roughly $100MM per partner. At $100K per check that's a thousand checks per partner, which nobody can source or sit on. At $10MM per check it's ten. Large funds aren't ignoring pre-seed out of snobbery. They can't deploy there at the pace their own structure demands.
So the stage is left, structurally, to smaller vehicles. Roughly 90% of emerging manager commitments land in funds under $15MM. 61% of VC Lab managers are solo GPs, 28% are female, 56% are international, operating across more than 90 countries. That's a population of small, specific, locally rooted vehicles, not scaled-down large firms. The broader case is in small funds win big and our primer on venture capital micro funds.
Price follows from that. At pre-seed, often nobody else has met the founder, so you aren't in an auction, you're in a conversation.
How to start a pre-seed fund: sizing and construction
Fund size is a strategy decision disguised as a fundraising decision. Pick the number that makes your check size, position count and loss rate cohere, then raise that instead of the biggest number you can get.
Worked example: a $10MM pre-seed fund
Start with $10MM committed. A 2% annual management fee over a ten year life is $2MM, leaving $8MM investable. Allocate $4.5MM to initial checks and $3.5MM to reserves.
Thirty initial positions of $150K is $4.5MM. At a $6MM average post-money entry, each check buys 2.5%. That's a real position, and only because the entry price is low.
Now apply a pre-seed loss rate. Fifteen of the thirty return nothing. Ten return between cost and 3x, call it $2.5MM across all ten. Four are genuinely good. One is the fund's outcome.
Follow the ownership on the one that works. 2.5% at entry, diluted 20% at seed, 20% at Series A and 20% at Series B, lands at 1.28%. On a $400MM exit that's $5.1MM, about 0.51x of the fund from your best company. That's where most first-time pre-seed models quietly break.
Now put reserves to work. Deploy $3.5MM into the five or six positions that cleared a real milestone, enough to hold 2% on the top one through Series A. 2% of $400MM is $8MM. Add four good-not-great companies returning $1.5MM each, which is $6MM, plus $2.5MM from the middle ten. Gross is around $16.5MM on $10MM committed, roughly 1.65x.
That's the honest answer. A well-run thirty position pre-seed fund with one $400MM outcome returns about 1.65x gross. Reaching 3x takes a larger outcome, two of the four good ones going large, or a lower entry price. Buy the same companies at a $3MM average post and every ownership figure doubles, turning that top position from $8MM into $16MM.
Loss rate sets your position count
Everything above flows from the loss rate. At pre-seed, half the portfolio going to zero is a normal year, so these portfolios run wider than later-stage ones. Twelve positions isn't a portfolio here, it's a coin flip on the whole fund. Past roughly forty, a solo manager is indexing rather than investing. Twenty five to thirty five is where most small pre-seed vehicles land. More on pacing in emerging manager portfolio construction and fund size pitfalls in emerging VC.
Reserves behave differently at pre-seed
A Series A fund reserves to defend a position it already understands. A pre-seed fund reserves into a situation where the next round is a genuine re-underwrite by somebody else, with fresh eyes and more data than you had. Sometimes they decide the company is worth far more than you paid and your follow-on is expensive but obviously correct. Sometimes they decline, and your follow-on becomes the only capital in a round no institution wanted.
So pre-seed reserves should be slower and more concentrated than standard advice suggests. Release them only against an external signal you didn't generate yourself, and model the case where you deploy less than the full reserve. A fund forced to spend on schedule spends on companies that didn't earn it.
Ownership: defend it or concede it deliberately
There are two coherent postures and no third. Either you defend ownership with reserves on a few positions and accept a narrower portfolio, or you concede dilution deliberately, run wider, and rely on entry price instead of follow-on muscle. Both work. What doesn't work is claiming the first in your LP deck and doing the second in practice, which is how a fund ends up with thirty thin positions and nothing left when the good one raises. The power law in VC is unforgiving here, and venture fund economics covers the fee and carry side underneath it.
Pre-seed investing when there's no data to underwrite
"No traction" doesn't mean "no evidence." It means the evidence isn't in a spreadsheet. Four things reliably substitute for traction.
Founder market fit
Not passion. Specific, hard-won, unfair familiarity with the problem. The founder who ran operations at the company that would buy this knows which budget line it comes from, who blocks the purchase, and what the workaround costs today. Generic founders describe the market. Founder-market-fit founders describe a Tuesday.
The fund side has moved the same way. Generalist funds fell from 22% of new funds in 2020 to 5% in Q1 2026, while specialists went from 78% to 95%. LPs are backing managers who know one thing deeply, and those managers are best at recognizing founders who do.
The quality of the wedge
Most pre-seed companies pair a large ambition with a narrow first product. The narrow part is what you're underwriting. A good wedge ships in months, hurts enough that someone pays before it's polished, and earns a genuine right to the next thing. A bad wedge stays small, or it's a feature a larger product absorbs. Ask what the second product is and why the first one earns it.
Speed of learning
The highest-signal variable at this stage and the easiest to observe. Meet the founder, then meet them again three weeks later. What changed? A team that ran six customer conversations, killed an assumption and rewrote the positioning has shown the one capability that determines survival. A team that spent three weeks polishing the same deck has shown something too. Building conviction over two months costs almost nothing when nobody else is competing for the round.
The sourcing relationship that got you there early
How you met matters as much as what you saw. If a founder came to you before they were fundraising, because of something you'd written, built or done, that's information about your position in the market and about their judgment. If you saw the deal alongside two hundred other funds, you're a price taker with less data than the people running the process. A pre-seed fund without a proprietary reason for founders to arrive early is a seed fund paying pre-seed prices for what's left. The mechanics are in deal sourcing in emerging VC.
Raising a pre-seed fund from LPs
Most LP conversations for a first pre-seed vehicle reach the same question inside twenty minutes: why do you need a fund at all, why not just angel invest?
The honest answer to the angel investing question
It isn't deal access, because a good angel has that too. It's three things a checkbook can't do.
First, consistency of deployment. An angel writes checks when they personally have cash and conviction at the same time, so the portfolio gets shaped by their liquidity rather than by the market. A fund deploys against a construction model, the only way to get wide enough to survive a pre-seed loss rate. Thirty positions of $150K is $4.5MM, and almost nobody writes that personally, over three years, through a bad quarter.
Second, reserves. An angel rarely has capital ready for the follow-on that matters, so their best position dilutes exactly when it's working.
Third, the fund is the product. A committed vehicle lets you tell a founder yes in a week and gives LPs diversified exposure to a stage they can't reach one company at a time. That's what they're buying. Micro LP versus angel covers both sides, and so does the LP Institute.
What the LP arithmetic looks like
The average LP check into a first-time fund is $159K. Against a $10MM target that's about 63 commitments, against $5MM about 31. Real registers cluster tighter, with a few anchors and a tail, so most small pre-seed funds close with 25 to 45 LPs. Checks in the $150K to $250K band convert to signed LPAs at 1.2x to 2.4x the rate of other bands. The very large check that never closes is the most expensive object in a first fundraise.
Timing has favored new managers. Across a sample of 1,000+ PACTs, 1,000+ LPAs and 900+ funds, February through May 2026 ranked among the top five fundraising months in four years, each running 1.2x to 2.2x the same month in 2025. Managers under 40 are now 38% of the population, up from 25%. The data sits in the VC Lab research hub, and close mechanics in decoding first closes in emerging VC.
What LPs diligence in a first-time pre-seed manager
Track record helps but rarely exists in citable form. LPs test three things instead: how a founder finds you before anyone else, what your model does when half the portfolio fails, and whether your back office will hold. Their side is mapped in the LP due diligence checklist.
The honest hard parts of running a pre-seed fund
The portfolio looks bad for years before it looks good
Failures show up fast and successes show up slowly. Companies die in year two and get large in year eight, so from year two to year five your LP report is a list of shutdowns next to companies that haven't raised yet. Markups don't fix it, since a marked-up pre-seed position is an unrealized number set by one other investor. Say this in the first LP meeting and put it in writing. The managers who get hurt are the ones who let LPs find out in year three.
Long time to DPI
Cash back to LPs takes longer from pre-seed than anywhere else in venture, because you're earlier by definition and secondaries here are thin and heavily discounted. Plan for a full ten year life. If your own finances need distributions inside five years, pre-seed is the wrong instrument.
Signalling risk when you can't follow on
This one is specific to small funds and it's real. When you skip the seed round, sophisticated investors notice, and some read it as information about the company rather than about your reserve capacity. The fix isn't pretending you have deeper pockets. State up front, with founders and with the next round, that your fund invests once at pre-seed and reserves by written policy. A consistent policy carries almost no signal. An inconsistent one carries plenty.
The operational load lands on one person
With 61% of managers running solo, the same person sources, diligences, closes, reports, files and answers LP questions. Administration and compliance don't scale down just because the fund did, and those hours come straight out of sourcing, the only activity that creates return. Decile Hub runs 1,000+ firms and Decile Partners handles the back office, which is the point of using infrastructure built for small funds rather than assembling it yourself.
Frequently asked questions
What is a pre-seed fund?
A pre-seed fund, or pre-seed venture fund, writes the first institutional check into a company, before there's revenue history or a proven acquisition channel. Typical checks run from $25K to $150K into rounds priced low enough that a small check buys real ownership. Because those sizes are immaterial inside a large fund, pre-seed funds are almost always small, frequently under $15MM.
How much money do you need to start a pre-seed fund?
Less than most people assume. Sub $10MM vehicles are common and functional, and roughly 90% of emerging manager commitments go to funds under $15MM. What matters isn't the total, it's whether the total supports a coherent construction. A $5MM fund can write 25 to 30 checks of $100K to $150K with modest reserves.
What's the difference between pre-seed and seed investing?
The difference is what evidence exists when you commit. At seed you're checking whether an observed pattern continues, so there's usually revenue, retention data, or a channel with a known cost. At pre-seed you're deciding whether the pattern will ever start, working from a team, a wedge and sometimes a prototype. That changes the diligence, the entry price, the loss rate and the position count.
How do you evaluate a pre-seed startup with no traction?
Four things substitute for traction. Founder market fit, meaning specific unfair familiarity with the problem. The quality of the wedge, meaning whether the narrow first product ships fast, sells before it's polished, and earns a right to the next one. Speed of learning, observed by meeting the team twice a few weeks apart. And how the deal reached you, since arriving before a competitive process is itself evidence.
Why not just angel invest instead of raising a pre-seed fund?
Angels deploy against their own liquidity, so portfolios get shaped by personal cash flow rather than a construction model, and few reach the thirty positions a pre-seed loss rate demands. Angels also rarely hold reserves, so their best company dilutes them away exactly when it's working. A fund gives you consistent deployment, structured reserves, and gives LPs diversified access to a stage they can't reach one company at a time.
Start your pre-seed fund
Nobody talks you out of a large fund. Plenty of people will talk you out of a small one, usually people whose economics depend on managing a large one. The arithmetic above is the answer: at pre-seed, a small vehicle isn't a compromised version of a big firm, it's the only instrument that makes the stage's check sizes mean anything. Stop apologizing for the size. Treat it as the product.
VC Lab is the accelerator for first-time and emerging managers launching exactly this vehicle, with 20 cohorts run, 950+ VC firms accelerated and 85% of participating managers investing at pre-seed or seed. It covers thesis, LP strategy and fund formation end to end. If you want to start writing checks while you raise, Start Fund gets a first vehicle live fast so your track record builds before the close. Apply to VC Lab and build the fund that fits the stage you actually invest at.