The YC Playbook for Seed Fundraising: Best Practices, Metrics, and the 2020–2026 Market Landscape

Drawing on lessons from Geoff Ralston, Michael Seibel, Dalton Caldwell, and Gustaf Alströmer — plus six years of market data

Raising a Seed to Series A

Introduction

Seed fundraising is one of the most misunderstood phases of building a startup. Founders romanticize it, fear it, and often run at it for the wrong reasons. Y Combinator has, over the years, distilled a remarkably consistent philosophy: raise less, raise later, and only when your fundamentals genuinely give you leverage.

But that philosophy doesn’t exist in a vacuum. Between 2020 and 2026, the early-stage funding market has lived through a boom, a bust, and a reinvention. The “right” amount to raise, the “right” valuation, and the “right” metrics to show have all shifted dramatically. This article synthesizes YC’s playbook with the market's actual trajectory and its current state.

Part I: The Market Landscape — Seed and Series A from 2020 to 2026

2020–2021: The ZIRP Boom

The COVID-era zero-interest-rate environment unleashed an unprecedented flood of capital into venture. Seed rounds that had historically clocked in around $1–2M ballooned into the $3–5M range. Series A median valuations spiked into the $60–80M+ pre-money range, often raised with little or no revenue. SaaS multiples hit 50–100x ARR. Tourist capital from hedge funds, crossover investors, and SPACs flooded the late stage and pushed pricing downward into earlier stages.

What investors wanted to see: A great team, a big TAM, and a credible narrative. Traction was a nice-to-have, not a requirement.

2022–2023: The Reset

When interest rates rose, the music stopped — fast. SVB’s collapse in March 2023 amplified the shock. Late-stage rounds essentially froze. Down rounds and shutdowns spiked. Series A median valuations fell 30–50% from the 2021 peak, and round sizes compressed. Tourist capital exited the market.

This is the era that produced All-In’s State of Series A’s episode: a median Series A of $7M raised on a $40M pre-money — a 17–26% YoY drop, and what Bill Gurley called “a return to normalcy.”

What investors wanted to see: A pivot. Revenue, retention, burn discipline, and a clear path to profitability. The “growth at all costs” mantra was retired.

2024–2025: Bifurcation

The market split in half. AI startups raised at dot-com-era multiples — sometimes $500M rounds on $2–3M in revenue. Everything else faced a much harder bar. The seed-to-Series A graduation rate fell from a historical ~50% to ~38%, and the time between rounds stretched out. LPs began pulling commitments back by 50–100%, starving the next generation of funds.

What investors wanted to see: Either a defensible AI thesis with breakout traction, or for non-AI companies, real revenue ($1M+ ARR for seed, $2M+ for A) with strong unit economics.

2026: The New Normal

Today’s market is best described as structurally bigger checks, structurally fewer companies funded.

Seed rounds in 2026:

Median U.S. seed: ~$3M — 3x the 2018 level (Crunchbase)

Upper quartile: $5.6M, top deals $8–10M (Crunchbase)

April 2026 alone: 222 seed deals, $2.5B total, median check $3.5M (State of Venture)

Seed-to-Series A conversion: ~38% (Value Add VC)

Series A rounds in 2026:

MetricSeries AMedian round size$10–$15MMedian pre-money valuation$49MMedian ARR at raise$2.8MTop-quartile ARR at raise$6.9M

ARR expectations at fundraising have risen ~20% year-over-year (TheFundCFO). And the AI distortion is real: in March 2026, AI companies captured 60.1% of all U.S. venture capital (AlleyWatch). OpenAI’s $122B raise alone was 43% of Q1 2026’s global venture total (CB Insights).

Meanwhile, the VC industry itself is consolidating. First-time fund formation collapsed to 101 funds in 2025, the lowest since 2011 (NVCA). The top 10 funds now capture 32.9% of all VC capital raised, up from 13% in 2021 (NVCA).

What this means for founders: Bigger rounds at seed give you more runway — but they also raise the bar at Series A and ratchet up dilution if you miss. Investors are writing fewer, larger, higher-conviction checks. The market punishes anyone in between hype and traction.

Part II: What Investors Want to See in 2026

The bar is now concrete and quantitative. Stripped of cycle-specific noise, here’s what seed and Series A investors are evaluating today.

At Seed (2026)

A working product with early users — the era of pre-product seed checks is largely gone for non-AI companies

Early signs of growth — week-over-week or month-over-month traction, even at small absolute numbers

A founder profile that screams unfair advantage — domain depth, technical edge, prior operating credibility

A clear AI angle, or a clear reason you don’t need one — investors will ask either way

A believable path to $2–3M ARR within 18–24 months — because that’s now the Series A bar

At Series A (2026)

$2.8M+ median ARR — and top-quartile founders show $6.9M+ (Zeni/Carta)

3x+ YoY growth at minimum, with retention metrics to match

Net revenue retention of 110%+ for SaaS, or strong cohort retention curves for consumer

CAC payback under 18 months, ideally under 12

Burn multiple under 2x — a metric that barely existed in 2020 and is now standard diligence

A real go-to-market motion, not just inbound demand

This is the lens through which YC’s principles should be read. The advice hasn’t changed — but the bar against which it’s measured has risen dramatically.

Part III: The YC Playbook — Principles That Survive Every Cycle

1. Start With the Right Mindset: The Market Is Not Rational

Geoff Ralston opens YC’s Fundraising Fundamentals with a warning every founder needs to internalize: the venture market “seems like kind of an open market but it’s not rational, it seldom fair”. The Sand Hill Road mythology — founders walking out the door with term sheets — is “the exception rather than the rule”.

You will hear “no” a lot. You will hear smart people explain, in detail, why your startup won’t work. The job of the founder is to remain “tough and resilient” and, above all else, to keep believing. This is even more true in 2026, where deal volume is down and investor conviction thresholds are up.

2. The Cardinal Rule: Raise When You Don’t Need To

If there is one piece of advice YC repeats more than any other, it’s this: “The best time to raise money is when you don’t need it.” When you’re not desperate, investors see opportunity. When you are, “VCs can smell that a mile away”.

Dalton Caldwell and Michael Seibel reinforce the point with two of the most famous examples in tech. Google had massive traction running as google.stanford.edu before raising its first dollar — and because of that leverage, “the first money they raised was on very good terms, and every additional round they raised after that they had incredible leverage”. Facebook was reportedly profitable even as a tiny college network — and that’s a major reason Mark Zuckerberg still owns such a meaningful share of the company today.

In a 2026 market where the top 10 funds control a third of all capital, leverage matters more than ever. The terms of your first round shape the cap table for every round after.

3. The Single Most Important Metric: Growth

When asked what actually makes fundraising work, Dalton Caldwell is blunt:

“The easiest way to fundraise is to indeed have a good metric that’s growing. When I ran a startup that wasn’t growing, I spoke to 140 investors and only got two angel checks. Now I’m working on a startup that is growing — almost every well-known VC is trying to figure out how to talk to us. It took us one week to raise our seed round.”

A common myth is that it’s smarter to raise before you have hard numbers — that ambiguity preserves valuation. YC pushes back hard. Once you have revenue, “you will be judged on the revenue in all like 97% of the time”. And an MVP with real customers gives you more leverage 97% of the time.

This advice has only gotten more important. In 2020, ambiguity was an asset. In 2026, with Series A medians at $2.8M ARR, ambiguity is a red flag.

There’s a deeper warning here too: founders who, deep down, believe their product is weak often rush to raise “before the world figures it out”. Investors sense it.

4. How Much to Raise

Ralston’s rule of thumb: raise as if this is the last money you’ll ever get. Specifically, raise enough to either reach profitability or hit clear milestones that will unlock the next round. The standard YC framework is 18 months of runway.

Dalton and Michael reinforce it :

“You will find a way to spend all of the money in your bank. Stay lean and get your fundamentals right.”

Brian Chesky, in a now-famous YC batch talk, reframed it: money is more like food than oxygen. People die from too much food, not just too little. Overcapitalized founders dilute more, innovate less, and lose discipline.

This warning is especially relevant in 2026, where seed checks are 2–3x what they were five years ago. Bigger rounds tempt bigger spending — and bigger spending raises the bar for Series A in a market where only 38% of seeds will ever raise an A.

5. Operational Metrics After the Raise

In How Much Should You Spend After Fundraising?, Gustaf Alströmer translates philosophy into specific operating numbers:

Day 1 of close: Set “a very clear set of milestones with good metrics.”

Plan for a 24-month total window from seed close to the next raise

Begin fundraising at ~8 months of runway remaining — that gives you 16 months to actually hit your milestones

Once revenue is flowing, don’t spend more than revenue coming in on hiring or marketing

YC’s frugality hack: Move half your seed money into a separate bank account and pretend it doesn’t exist for the first year

The brutal context: “Most of the companies that raise a seed round will not be able to raise a Series A. And most that raise a Series A will not be able to raise a Series B. The bar just gets higher.”

In 2026, this isn’t theory — it’s the 38% conversion rate in hard numbers.

6. What Investors Are Actually Buying

Ralston distills the investor’s mental model down to two things:

You. “Investors invest in you. Ask yourself: if you were an investor, would you invest in you?”

Your story. Does it describe a genuinely large opportunity? Is there a compelling product and traction? Is the storyteller impressive?

The best investors aim to find founders before the metrics are obvious — “you’re expensive by then”. As Ralston puts it: “The very best investors are Airbnb before they’re ever B&B — before they have traction”.

In a 2026 market where capital is consolidating at the top, this dynamic intensifies. The best partners at the best firms are still hunting for pre-traction insight — but they’re harder to reach, and they need a sharper story to bite.

7. The Instrument: Why YC Recommends the Post-Money SAFE

YC’s clear preference at seed is neither priced equity nor convertible notes — it’s the Post-Money SAFE:

3–5 pages long

No lawyers required; total cost can be a few hundred dollars

Transparent on dilution: “$1M invested on a $4M post-money SAFE = you own 25%”

Convertible notes are harder to reason about because dilution depends on how much additional money you raise alongside them. Equity rounds are “slow, almost always expensive,” require lawyers, and bring preferred-stock provisions you may not want at the seed stage.

Critical advice that founders too often skip: read every word of every document you sign. “There are many stories of nightmares with people agreeing to things they didn’t know they were agreeing to”.

8. The Pitch Meeting: Tactics That Work

From Ralston’s playbook:

Research the investor first. Know what they’ve invested in and what they care about [6]

Capture attention in the first 1–2 minutes. The simplest version of your story is usually the strongest

Bring a demo — even a wooden prototype is better than slides

Listen more than you talk. “A good sign that an investor meeting is going well is when they talk at least as much or more than you do”

Never leave without a conclusion — a check, a firm no, or clearly defined next steps

Decks matter less than you think at seed. Ralston confesses: “As an angel investor myself, I almost never even look at the deck. I just want to look at the founder and hear their story.”

Your pitch improves with every meeting. Like a golf swing, it needs feedback reps to develop

9. The Valuation Trap

Setting valuation is one of the most dangerous parts of a seed round. Pitch too high and you can kill the round outright — and worse, lowering it later doesn’t make you a bargain, it makes you look weak. Pitch too low and you over-dilute permanently.

Ralston’s grounding principle: “The most important thing is to get the money in the bank and get back to work.”

VCs are mathematically motivated to own a target percentage of your company [8]. Understand their model, not just yours. And in negotiation: “you can delay. Say you need to talk to your co-founder. Don’t go toe-to-toe with the pros”.

In 2026, with Uncork’s Andy McLoughlin publicly stating their average check has grown from $2.5M to $4.5M while still targeting 10%+ ownership (Crunchbase), the math of dilution is unavoidable. Bigger checks at the same ownership target means higher valuations — but only for the founders who clear the bar.

10. The Underlying Mistake: Confusing Investors for Customers

Michael Seibel’s most pointed observation is psychological. Many first-time founders unconsciously treat investors like teachers — trying to get an “A” (money) by pleasing the authority figure. That mental model corrupts everything downstream.

The honest audit, in Seibel’s words:

“How much of your waking hours last week were spent talking to customers and building product? If it’s 80–90%, you’re probably doing it right. If it’s 20%, you’re probably doing something very very wrong.”

Fundraising is a means, not an end. Customers fund companies; investors fund rounds. Founders who reverse that hierarchy tend to build companies that can raise but can’t survive — and in 2026’s market, “can’t survive” comes faster than ever.

Closing: The YC Synthesis, in a 2026 Market

The principles haven’t changed. The bar has.

In 2020, you could raise on a deck and a dream. In 2023, you needed a story. In 2026, you need a working product, real growth, and a credible path to $2.8M ARR — or an AI thesis sharp enough to bend the rules.

If you compress YC’s seed-stage philosophy into a single paragraph for today’s market, it reads roughly like this:

Build something people want. Get it growing before you fundraise. Raise the minimum you need on a clean Post-Money SAFE when you have leverage. Treat half your bank balance as if it doesn’t exist. Set milestones that get you to $2–3M ARR within 18–24 months. Start your next raise with eight months of runway left. And spend 80%+ of your time with customers, not investors.

The market will keep cycling. The principles won’t.*

Learn from YCombinator — https://www.alaniconnect.com/room/y-combinator