Where Founders Actually Lose Deals: A Post-Mortem of 47 Failed Rounds on the Institutional Readiness Grid

The founder line that started this

In April 2022 Fast shut down after raising roughly $125 million led by Stripe against reported revenue of about $600,000. The public artifact I kept coming back to was Domm Holland's line on the way out: "Sometimes trailblazers don't make it all the way to the mountain top. But even in those situations, they pave a way that all others will follow." That is a Story-axis framing. It positions the failure as a category-timing question, a first-mover-tax question, a market-not-ready question. The dossier for this memo puts Fast squarely under Operations, not Story. The one-click-checkout market did not evaporate. Bolt did not evaporate. Shop Pay did not evaporate. What broke inside Fast was burn discipline against traction, including a widely reported million-dollar Chainsmokers performance that ran alongside almost no revenue.

I have read a lot of these post-mortems now, and I kept noticing the same shape. Founders write their own ending as a Story failure because Story failure is the most emotionally survivable version of it. The failure the deck describes is not the failure that killed the company. That gap is what this memo is about.

The dataset

We assembled 47 named failed or distressed venture rounds from 2022 to 2026, weighted toward 2024 to 2026. Sources include Crunchbase, TechCrunch, The Information, PitchBook down-round coverage, founder blog post-mortems, LinkedIn founder posts, and shutdown trackers from Carta, AngelList, SimpleClosure, CB Insights, and postmortem.io. Every named company below traces back to a public URL in the research dossier.

Each round is mapped to a primary failure axis on the Institutional Readiness Grid: Story, Capital, Operations, or Diligence. Where a failure legitimately spans two axes (EasyKnock, Bird Global, Parker) we made a single primary-axis call and preserved the multi-axis nuance in the write-up rather than force false precision. One entry, Yara AI, was excluded from the axis tally because it was a voluntary ethical shutdown, not a market or fundraising failure. That leaves 46 categorized entries.

This is a curated sample, not a random one. Public post-mortems skew toward well-funded, well-covered companies and founders willing to write candidly. Quieter bootstrap-stage shutdowns are undercounted in every public dataset, ours included. We are describing what the documented failures look like, not the entire population.

The headline finding

Story: 30%. Diligence: 28%. Operations: 22%. Capital: 20%.

70% of the failed rounds in this sample died on an axis other than Story. Diligence and Operations combined outweigh Story alone. Capital-structure failure, which is nearly invisible to founders until the round is already dead, is almost as common as Story failure.

The "VC didn't get it" excuse turns out to be the least common actual failure mode once you count instead of remember. Story failure is still the plurality, and no serious institutional investor should dismiss it. What the data does not support is the founder reflex that treats a rejection as a taste disagreement about the narrative. In this sample, Diligence readiness alone is nearly tied with Story, and the two of them clustered together (Diligence + Capital) are the single most-common failure pattern at growth stage.

Founders who misdiagnose the axis that killed the last round fix the wrong thing before the next one. Rewriting the deck does not fix a data room integrity problem. Hiring a designer does not fix a unit-economics problem. A better warm circuit does not fix a governance problem. The self-diagnostic is what changes the outcome, not the pitch.

The four failure modes, with named cases

Story failures (30%)

Story failure at growth stage does not look like a bad deck. It looks like a narrative that was structurally or temporarily wrong at scale. WeWork ran a business whose 10-to-20-year lease obligations funded short-term rentals against $18.6 billion in debt and $15 billion in assets, and the underlying story never squared. Convoy's freight-marketplace narrative broke when the macro tailwind reversed and revenue halved inside 18 months of raising at a $3.8 billion valuation. Hopin raised at a $7.75 billion peak on a pandemic tailwind sized as if it were permanent, then sold its flagship product for $15 million. Olive AI raised $832 million on an AI-automation narrative that customers said was RPA wrapped in marketing. These are not "the VC didn't get it" failures. These are "the story described a business the company was not actually running" failures.

Diligence failures (28%)

Diligence failure is where the pitch was fine and the substantiation behind it was not real or not durable. Builder.ai raised $445 million against a $1.5 billion peak valuation, then filed bankruptcy in 2025 after internal audits slashed revenue projections by 75%, "AI-powered" development turned out to be largely offshore human engineers, and lenders seized $37 million of the remaining $42 million in cash. The company had no CFO since 2023. Charlie Javice's Frank fabricated synthetic user data (paying a data scientist roughly $18,000) to inflate reported users from about 300,000 to about 4 million ahead of a $175 million JPMorgan acquisition, and Javice was convicted in 2025. IRL's board investigation found 95% of its reported 20 million users were bots or automated, and the SEC charged the founder in 2024. Latch restated two years of financials after internal-control deficiencies in revenue recognition, cut 60% of staff, and got delisted. Not every case in this bucket is fraud. theGist, Spotlight Therapeutics, and Zume all failed on diligence without any fraud at all, because the unit economics, clinical data, or core technical claim simply did not survive the next round's scrutiny.

Operations failures (22%)

Operations failure is the least glamorous bucket and the one that is easiest to hide inside a Story narrative on the way out. Fast is the archetype. Bench shut down abruptly in late 2024, locked thousands of small businesses out of their own accounting and tax data with no transition plan, and the shutdown itself was the defining operational event. Canoo and Lilium each raised over a billion dollars against physical manufacturing and regulatory certification challenges that capital alone was not going to solve. Noogata landed PepsiCo and Colgate logos, watched them all sit in pilot mode without converting, and could not raise the next round. Outdoor Voices burned through funding on mismanagement and extravagant spending before closing all 15 stores in 2024. Every one of these is a discipline problem, not a market problem.

Capital failures (20%)

Capital failure is round shape, timing, valuation, and investor mismatch. Bolt lost $302 million on $27 million of revenue in 2023 amid founder-and-investor legal disputes, two board putsches, and three CEOs in the same window that the reported valuation collapsed from $11 billion toward roughly $300 million secondary. Tally, a16z-backed with $172 million in, ran out of cash mid-operation and stopped making promised credit-card payments for users. Divvy Homes' rent-to-own model was directly exposed to rate risk, and the entire business sold to Brookfield/Maymont for roughly $1 billion after peaking near $2 billion. Pandion raised into a pandemic e-commerce boom and launched into the freight recession that had already begun. Veev's capital raise was cancelled at the last minute and the company shut down without runway to continue. Capital-axis failures are the ones founders almost never see coming, because the round shape looks fine on the day it closes and only looks broken 18 months later.

The stage-shift pattern

The failure axis moves with the round. Bucketing the 46 categorized entries by last-known stage:

Thin substance survives the early rounds because trust is cheap at pre-seed and seed. There is no operating history to interrogate, so a compelling founder and a plausible wedge are enough. That same thin substance dies later, because Series A is the first paper stage where diligence gets serious, and by Series B the story is assumed and the numbers carry the round.

This is the pattern behind what SimpleClosure called the shift from "failed ideas to failed business models," with Series A shutdowns jumping from about 6% to about 14% of all closures year over year, roughly a 2.5x increase. The correction is moving up the stack. Companies that would have died at seed are now dying at Series A, and the axis that kills them is different. Readiness is not a bar a founder clears once. Each axis has a round at which it becomes binding, and founders who survived a seed by nailing Story arrive at Series A undertrained on Diligence.

Contrast cases: rounds that succeeded despite weakness on one or two axes

The grid is a diagnostic, not a filter. Airbnb was passed on more than 20 times by investors who could not get past the safety and adoption story before Sequoia funded them at $600,000 in 2009. Story readiness was, by the market's own repeated verdict, weak. Operations readiness (execution on trust and safety systems, and pure founder persistence) carried the company through the weak axis. Melanie Perkins was rejected by more than 100 investors on Canva because she was in Perth without the technical-co-founder profile the room expected, pitching into a category Adobe was assumed to own. A MaiTai retreat connection produced the $3 million seed. Canva is now valued at $42 billion with eight consecutive years of profitability. Slack came out of Tiny Speck, whose game Glitch was a complete Story-axis failure and got shut down in 2012. The team, the discipline, and the cash management were intact enough to rebuild around the internal tool that became Slack, later sold to Salesforce for $27.7 billion. Figma launched into a category where InVision was the well-capitalized incumbent at a $1.9 billion valuation, and won on Operations (browser-based real-time multiplayer that InVision could not match), driving InVision to discontinue its product in 2024.

Each of these was demonstrably weak on one axis and won because another axis was overwhelmingly strong. That is not evidence the grid is wrong. It is evidence that a founder can survive a weak axis if the compensating axis is genuinely dominant, not merely present.

What founders should actually take from this

When a round dies, run the grid honestly on the company before writing the post-mortem. Ask which of the four axes the investor was actually rejecting. In our reading of these 47 cases, the axis that killed the round almost always correlates with the fix that would have unlocked the next raise. Story-driven rejection asks for narrative reframing and better proof points around wedge and defensibility. Diligence-driven rejection asks for data room integrity and operating history the next investor can stress. Operations-driven rejection asks for burn discipline and a repeatable go-to-market motion beyond the founder-led hustle. Capital-driven rejection asks for a different round shape, different investors, or a different valuation.

The reflex that reaches for Story-blame first is the single most common expensive mistake we saw in this dataset. It is emotionally the cheapest ending to write and analytically the least accurate one most of the time. Diligence readiness in particular is the axis founders undertrain for, because it does not feel like a story problem or an execution problem. It feels like paperwork. It is not paperwork. It is the discipline the company either built in from day one or did not.

Through-line

Memo 001 laid out the Institutional Readiness Grid as the four-axis diagnostic we use at Aureum Capital when we look at a company. Memo 002 covered why the operational-diligence bar has moved in the last two years and why that shift explains the visible spike in Diligence and Operations failures in this dataset. This memo is the empirical companion. When we say Diligence readiness has become the axis that most often kills growth rounds, we are describing the 47 companies above, sourced and named, not a thesis in the abstract.

Grow Today, our founder services platform, exists to run the grid on companies before the market does. The single most useful pre-raise exercise a founder can do is the one that follows a round dying: put the last twelve months of the company against the four axes, name the axis that broke first, and fix that one before touching the deck.