In June 2021, SoftBank led a round that valued Klarna at $45.6 billion, making the Swedish payments company the most valuable private fintech in Europe. Thirteen months later, Klarna raised $800 million at a $6.7 billion valuation. That’s an 85% cut, and one of the largest down rounds European tech has recorded. The Swedish press treated it as a funeral. Employees who had joined for the equity watched their paper wealth evaporate over a single announcement, and the company spent most of 2022 answering questions about whether it would survive.
It did. In September 2025, Klarna listed on the New York Stock Exchange, raising up to $1.27 billion at a valuation near $14 billion. Roughly double the down-round price, still a long way from the peak, and very much alive.
A down round is when a startup raises new funding at a lower valuation than its previous round, meaning new investors buy shares at a cheaper price than the company was last worth on paper. The mechanics are simple. What the mechanics do to your cap table, your option pool, and the twelve people who joined because they believed in the number on the term sheet is where most founders get blindsided.
Last updated: August 2026
Quick answers
What is a down round in startup funding?
A down round is a financing where a startup sells new shares at a lower price per share than it did in its previous round. The company’s pre-money valuation comes in below the last round’s post-money valuation. Existing shareholders lose paper value, and everyone on the cap table takes more dilution per dollar raised.
How does a down round affect employee stock options?
Options with a strike price above the new share price go underwater, meaning it costs more to exercise them than the shares are worth. The company’s 409A valuation typically drops too, so new grants get cheaper strike prices while older grants stay stuck. Boards fix this by repricing, swapping options for RSUs, or issuing fresh grants.
Is a down round always bad for a startup?
No. A down round that buys 24 months of runway at a price the company can grow into beats a flat round that runs out in nine months. Klarna cut its valuation 85% in 2022 and listed on the NYSE at roughly $14 billion in September 2025. The damage comes from the terms attached, not the number itself.
What is a down round in startup funding
A down round happens when the pre-money valuation of a new financing lands below the post-money valuation of the round before it. The share price is the thing that matters, not the headline number. If your Series A closed at $4.00 per share and your Series B prices at $2.50, you’re in a down round regardless of how much capital you raise.
Run the math on a simple case. A company raises $10 million at a $40 million pre-money valuation, closing its Series A at a $50 million post. Eighteen months later, growth has stalled, and the best available term sheet offers $12 million at a $30 million pre-money. The new post-money is $42 million, below the $50 million the company was worth on paper. Investors in that Series B now own 28.6% of the company for $12 million. In the earlier round, $10 million bought 20%.
The cost of a down round is not the lower number. It’s the extra ownership you hand over for the same dollars. Every point of additional dilution comes out of the founders and the employee option pool, because new investors negotiate their percentage first and the rest of the cap table absorbs what’s left.
One more distinction worth getting right. A cut to your 409A valuation is not automatically a down round. Instacart lowered its internal 409A valuation from roughly $39 billion to $24 billion in March 2022 without raising a priced round at all, then cut again to about $13 billion that October. No new investor bought in at the lower price. The company repriced its own common stock so it could grant employees cheaper options during a brutal hiring market. Those are different events with different consequences, and founders conflate them constantly.
Down round vs up round vs flat round
The three outcomes differ on one variable: how the new share price compares to the old one. Everything else follows from that comparison.
| Round type | Share price vs last round | Anti-dilution triggered | Effect on employee options | Typical signal |
|---|---|---|---|---|
| Up round | Higher | No | Existing grants gain value; new grants get pricier strikes | Metrics beat the last round’s plan |
| Flat round | Same | No | Neutral; grants hold their strike price | Progress made, but not enough to reprice upward |
| Down round | Lower | Yes, if prior rounds carry protection | Older grants go underwater; 409A usually resets lower | Missed plan, market reset, or an overpriced prior round |
| Structured round | Nominally flat or up | Sometimes, via ratchets | Common stock value falls even when the headline holds | A down round wearing an up round’s clothes |
That last row deserves attention. Some founders protect the headline valuation by accepting heavier liquidation preferences, participation rights, or an IPO ratchet instead of a lower price. The press release reads better. The common stock is worth less. Carta’s data shows liquidation preferences and participation rights sitting near multi-year lows in Q1 2026, which means founders currently have room to refuse those trades.
Why the 11.4% down-round rate misleads founders
The headline down-round rate is the lowest it’s been since 2020, and it tells you almost nothing about your own risk. Carta’s State of Private Markets report for Q1 2026, published May 29, 2026, put the rate at 11.4% of priced rounds, down from a 22% peak in 2023. Startups on the platform raised $30.4 billion in the quarter. Dilution fell. Terms moved toward founders.
Then read the second number in the same report. More than 60% of all venture capital raised on Carta in Q1 2026 went to AI companies, the highest share the dataset has ever recorded. Foundational model companies alone took 14.2% of total capital. An AI foundational model startup raising a Series A saw a median valuation near $300 million. A non-AI startup at the same stage saw $55 million.
Those aren’t the same market. When one segment absorbs most of the capital at prices five times higher than everyone else, it drags the blended down-round average down with it. If you’re not building in AI, the 11.4% figure describes a market you aren’t raising in.
The stage split makes it sharper. Down rounds reached 24% of US growth-stage financings in H1 2025, against a 4% baseline in 2021. Growth-stage companies are the ones carrying 2021 vintage valuations into a repricing environment, and there are only so many ways that reconciles. Seed companies priced in 2024 or 2025 have almost no gap to close. A Series C company that raised at 40x forward revenue in late 2021 has a very large one.
Ask a more useful question than “are down rounds common right now.” Ask what multiple your last round implied, whether current comparables support it, and how many months of runway you have to close the gap. GJ’s coverage of headline raises like Baseten’s $1.5 billion at up to a $13 billion valuation and Hark’s $700 million Series A at $6 billion shows what the top of the market looks like. Bezos-backed rounds like Prometheus at a $41 billion valuation sit in the same tier. Most companies are not raising in that market.
What actually triggers a down round
Down rounds come from one of four places, and only one of them is about the company being bad at its job.
The prior round was overpriced. This is the most common cause in the current cycle. A company raised in 2021 at a multiple that made sense when capital was free, then had to grow into a number the market no longer supports. Nothing went wrong operationally. The pricing was wrong from the start.
The company missed its plan. Revenue came in under forecast, churn ran higher than modeled, or the product took nine months longer to ship than the deck promised. Investors underwrite the next round against the last one’s milestones. Miss enough of them and the price moves.
The comparable set repriced. Public multiples in your category compressed, and private valuations follow public ones with a lag. Klarna’s cut in 2022 tracked a broad fintech selloff more than any single Klarna failure. The company’s valuation history reads like a chart of investor sentiment toward buy-now-pay-later. The same dynamic runs in reverse for AI infrastructure right now, where companies like Together AI repriced upward on category sentiment rather than a step change in fundamentals.
Runway ran out before the milestone landed. A company that needs money in six weeks does not get to negotiate. This is the trigger founders control most directly and manage worst. Every fundraising conversation you start with under six months of cash is a conversation where the other side sets the price.
The pattern worth internalizing: three of those four causes are about timing and market conditions, not execution. Lumping every down round into “the company failed” is how boards make bad decisions about people who are actually doing fine work. GJ’s look at the biggest startup failures of 2024 found that the companies that actually died burned capital against milestones they never came close to hitting, which is a different problem from a repriced round.
How does a down round affect employee stock options
Every option granted at a strike price above the new share price goes underwater the moment the round closes, and stays underwater until the company grows back past that price. An underwater option costs more to exercise than the resulting share is worth. It has no exercise value. Employees holding them are working for a salary plus a lottery ticket with a negative expected payout.
Two things happen mechanically. First, the 409A valuation, the board-approved fair market value of common stock, usually resets downward after a priced down round. Second, new grants issued after that reset carry cheaper strike prices than old ones. So the engineer who joined last quarter gets a better deal than the engineer who joined two years ago and helped build the thing. That asymmetry is what drives attrition, not the valuation headline.
Boards have four standard responses:
- Reprice existing options. Cancel the underwater grants and reissue at the new lower strike. This needs board approval, and it carries tax consequences under IRC Section 409A that a lawyer has to structure. Cooley’s guidance on private company repricing walks through the mechanics. Existing investors often push back, because repricing hands value to employees at their expense.
- Exchange options for RSUs. Restricted stock units have no strike price, so they carry value at any share price above zero. The tradeoff is tax treatment and a different vesting conversation.
- Issue fresh grants at the new price. The simplest option and the most dilutive. It leaves old grants underwater but gives people a reason to stay. Instacart’s 409A cuts in 2022 were explicitly framed around this, letting the company grant more shares at a lower price during a hiring war with DoorDash and Amazon.
- Buy back underwater options. Rare, cash-intensive, and mostly used for departing employees. It converts a dead asset into a small real one.
Whatever the board picks, do it fast. The gap between the round closing and the retention fix landing is when people update their resumes. Startups built around equity-heavy comp feel this hardest, which is part of why creator-economy and marketplace companies like Phia lean on cash and revenue share instead of option-only packages.
What anti-dilution provisions do to founder equity
Anti-dilution clauses decide who absorbs the pain of a down round, and the version in your last term sheet matters more than the valuation in your next one. These provisions adjust the conversion price of preferred stock when new shares issue below what earlier investors paid, effectively giving them more common shares on conversion.
Two formulas dominate. Full ratchet resets the earlier investor’s conversion price all the way down to the new round’s price, no matter how few shares get issued. If an investor paid $4.00 and you price a small round at $1.00, their entire position converts as though they’d paid $1.00. It’s brutal and it’s rare, appearing in fewer than 5% of venture deals. Weighted average adjusts the conversion price partially, factoring in how many new shares actually issued relative to the total outstanding. It splits the damage between founders and prior investors instead of dumping it on the common stock. Broad-based weighted average is the market standard and the one to fight for.
Then there’s pay-to-play, which has become common again in this cycle. A pay-to-play provision strips anti-dilution protection, and often other preferred rights, from any existing investor who declines to participate pro-rata in the new round. It forces your cap table to choose: put in more money or lose your protections. Founders generally want it, because it converts passive investors into either fresh capital or cleaner common stock.
Read the protective provisions before you read the valuation. A $30 million pre-money with broad-based weighted average and pay-to-play is a better outcome for founders and employees than a $40 million pre-money with a full ratchet and a 2x participating preference. The second one prices higher and pays out worse in every exit scenario short of a blowout.
Is a down round always bad for a startup
No, and treating it as a death sentence causes more damage than the round itself. A down round that buys 24 months of runway at a price the company can realistically grow past is a good trade. A flat round that buys nine months and leaves you fundraising again next spring is not, whatever the press release says.
Klarna is the clearest proof available. The company went from $45.6 billion in June 2021 to $6.7 billion in July 2022, absorbed a full year of “is Klarna dying” coverage, cut costs, and listed on the NYSE in September 2025 at a valuation near $14 billion. That’s roughly double the down-round mark and a fraction of the peak. Both facts are true. The 2021 number was a bubble artifact; the 2022 number was a floor the company built from.
The upside nobody mentions: a reset 409A gives you the cheapest hiring currency you’ll ever have. New grants at a low strike price are worth more to a candidate than the same percentage at a high one, because the spread is where their money is. Companies that recruit aggressively in the twelve months after a down round often build their best teams, since strong operators know how to read a cap table and can see the entry point.
The real risks are narrower than the panic suggests. Terms that survive the round, like participating preferences and ratchets, follow you into every future exit. Key-person attrition in the first 90 days costs institutional knowledge you can’t rehire. And a board that concludes the CEO mispriced the last round sometimes starts a conversation about leadership. Manage those three things and the valuation number is mostly an ego cost.
How to tell your team without losing half of it
Announce the round, the runway, and the equity fix in the same meeting, or don’t announce it yet. The worst version is telling people the valuation dropped and promising to “figure out equity soon.” That’s an invitation to speculate for six weeks, and speculation always lands worse than the truth.
What works, based on how companies that retained their teams through 2022 and 2023 handled it:
- Lead with runway, not valuation. “This round funds us through Q4 2028” is the number people actually need. The valuation is context.
- Say the word underwater out loud. Your engineers already checked the strike price against the new 409A. Pretending otherwise costs you credibility you’ll need later.
- Bring the board decision, not the board discussion. Repricing, RSU exchange, or fresh grants, decided and dated. If the board hasn’t decided, give a date when it will and hold that date.
- Name what changed in the plan. If the round came with a headcount freeze or a shifted roadmap, say it in the same meeting. People handle bad news once. They handle it badly in installments.
- Do 1:1s with your top ten within a week. Group announcements manage information. Individual conversations manage retention, and the people most likely to leave are the ones with the most options and the best alternatives.
A down round tests whether your team trusts you to tell them the truth when the news is bad. Handle it cleanly and it’s a story you tell later, the way founders who survived the last cycle tell theirs. Handle it badly and you’ll spend the runway you just bought backfilling the people who left. The down round is a price. The exodus is the actual loss.
