In 2018, roughly 35–40% of funded seed startups made it to Series A within two years. For the 2022–2023 seed cohorts, that number has collapsed to around 15–20%—and in some cuts of the data, closer to 10–15%.
If you’re a seed founder today, the default outcome is no longer “raise A a bit later.” The default outcome is “never raise A at all.”
From 40% to 15%: The Rapid Fall of Seed-to-Series A Success
Historically, the rule of thumb in venture was simple: about a third of funded seed companies would graduate to Series A within ~24 months. That’s the 35–40% graduation rate you’ll see if you look at pre-2020 cohorts across multiple datasets.
Now, multiple independent analyses are converging on a much harsher reality. An update on venture graduation rates from Incisive VC’s 2025 data shows recent seed cohorts graduating to Series A in the mid-teens. Scaleup Finance goes further, arguing that around 85% of seed-stage startups now fail to raise a Series A.
That’s not a small cyclical dip. It’s a structural reset of expectations. A world where 2 out of 3 seeds graduated has become a world where 4 out of 5 don’t.
The two-year window matters here. Most of these analyses define “graduation” as raising a priced Series A within 24 months of the seed round. That’s long enough for a real build-and-sell cycle, but short enough to exclude “zombie” companies that limp along for years on extensions and bridges.
Carta’s internal data (summarized in various investor letters) and Crunchbase’s coverage of the elongated path from seed to Series A both show the same pattern: the curve has shifted right (rounds take longer) and down (fewer companies ever get there). The cohorts that raised seed in 2021–2022 are now far behind where 2016–2018 cohorts were at the same age.
There are caveats. Different sources define “seed” differently, some include pre-seed, some don’t; some treat large “seed+” rounds as A’s, others don’t. And we won’t fully know the fate of the 2023–2024 seed cohorts for another year or two.
But when independent datasets, different methodologies, and multiple investors all land in the same range—15–20% graduation, 80–85% failure to reach A—the direction of travel is clear. The Series A crunch isn’t a Twitter meme; it’s the new base case.
Why Investors Are Raising the Bar—and What That Means for Your Team Size
Investors haven’t just tightened the purse strings; they’ve rewritten the bar for what “A-ready” looks like. A few years ago, a strong narrative, early product-market fit signals, and some revenue momentum could get you into a $8–12M Series A.
Now, many Series A investors quietly treat $1M ARR as the minimum ante, not the win. In SaaS, I’m hearing more “we really want to see $1.5–2M ARR with efficient growth” than “we’ll lean in at $500k ARR with a great story.”
That “efficient” part is doing a lot of work. Investors are screening hard on burn multiples, payback periods, and team discipline. The post-2021 hangover means they’ve seen what happens when you fund headcount growth instead of customer pull, and they’re over it.
There’s also a simple supply-and-demand story. The 2020–2022 boom massively increased the number of funded seed startups. That oversupply gives Series A investors a huge menu of companies to choose from, so they can be brutally selective without feeling FOMO.
When a partner can choose between ten seed companies in the same space, they will default to the one with more revenue, cleaner metrics, and a tighter team. The others don’t “almost” raise; they just quietly never close.
This is showing up in team size data too. Across multiple portfolio snapshots I’ve seen, the average headcount at Series A has dropped to around 15–16 people—down roughly 15–20% from five years ago, when 18–20+ was common.
That’s not because companies are doing less. It’s because they’re expected to do more with fewer people before they’re allowed to scale. The seed-to-A journey that used to fund “find product-market fit and build the team” now looks more like “prove product-market fit, show repeatable acquisition, and only then start building the team.”
Investors are saying this out loud. In Keith Newman’s “founders’ reality check for 2025”, one VC puts it bluntly: “We’re not paying for experiments anymore; we’re paying for working machines.” Another: “If you need 30 people to get to $1M ARR, that’s our problem with the business, not your justification for a bigger round.”
Translated: headcount is now a negative signal if it’s not tightly tied to revenue and learning. The bloated seed team that felt normal in 2021 is a liability in 2025.
Preparing for the 85% Failure Rate: What Founders Must Internalize Now
If 80–85% of seed-funded startups won’t raise a Series A, you can’t treat “we’ll raise an A” as a plan. It’s a hope. Your actual plan has to work in a world where the A never comes.
That requires a different kind of psychological resilience. You’re not just signing up for a hard journey; you’re signing up for a hard journey where the most likely outcome is “we built something real and still couldn’t clear the bar.”
That’s not a reason to quit. It’s a reason to be brutally honest with yourself earlier. Are you building a company that can survive on customer money, or are you building a company that only works if the next round shows up on schedule?
In this environment, “survival metrics” matter more than vanity metrics. Things like runway at current burn, runway at “emergency mode” burn, gross margin, payback on customer acquisition, and how much real pull you’re seeing from users or buyers.
Defensible moats also need to show up earlier. If you’re in a crowded category, “we’ll build a moat later” is a fantasy. Investors are asking, explicitly or not: what gets stronger with every customer you add? Data? Network effects? Workflow lock-in? Switching costs?
On expectations, seed founders in 2025–2026 should assume:
- Raising a Series A will take longer than you think, even if you’re “doing well.”
- Round sizes may be smaller and valuations flatter than your seed deck assumed.
- You may need to show closer to Series B metrics from 2018 to raise a Series A in 2026.
- Bridge and extension rounds will be common, but they’ll come with sharper terms and tougher conversations.
Notice what’s missing here: generic advice like “build relationships early” or “tell a compelling story.” Those are table stakes. They don’t change the math that only 15–20% of your cohort will get through the Series A gate.
The strategic implication is simpler and harsher: design your company so that if you end up in the 80–85% that never raise an A, you still have options. Profitability paths. Smaller but real exits. A business that can shrink to survive instead of exploding on impact.
Facing the Crunch: A Data-Driven Wake-Up Call for Seed Founders
The Series A crunch is not a vibe shift; it’s a measurable collapse in seed-to-Series A conversion. We’ve moved from a world where ~40% of funded seeds graduated within two years to one where ~15–20% do, and 80–85% never raise an A at all.
Investors have responded by raising the bar on traction, efficiency, and team discipline. Leaner teams, higher ARR expectations, and stricter filters on what “working” looks like are now the norm, not the exception.
If you’re building at seed today, you don’t control those macro conditions. You control how honestly you see them, how you design your burn and team against them, and whether your plan survives contact with a world where the next round is unlikely.
The founders who make it through this cycle won’t be the ones with the loudest narrative. They’ll be the ones whose numbers, teams, and psychology were built for a world where the Series A crunch is just how the game works now.
