Hacker News new | past | comments | ask | show | jobs | submit login
Y Combinator Failed Startups (failory.com)
415 points by nicocerdeira on Oct 9, 2020 | hide | past | favorite | 184 comments



I don't get the last section, where you just name 4-5 small "failed" startups. For most of them, you don't even have any real insights!

Why pick on them, for no reason? They didn't do anything unethical, and most didn't seem to even raise a Series A.

You have no clue why they failed; you just speculate... and you don't just name the companies, you call out random founders by name. These people put themselves out there and worked hard, and now they're going to get a Google Alert saying they're in a blog post called "Biggest Y Combinator Failed Startups."

Most YC companies don't work out. In my batch (5 years ago), about 80% are out of business now. That's okay! That's just how it works. I'm all for talking about failures, but it's weird you just pick on a few small random YC startups.


I have to agree. It feels like punching below the belt. The only reason why you know that the smaller startups have failed is because the founder has humbly written up a writeup on what went wrong and lessons learned, in the hope that it'll help others. By including them, the article is functionally punishing them for being open about their mistakes.

Maybe for larger companies this is debatably okay, because there's diffusion of responsibility. But to call out a startup with one or two founders who have been shouldering the weight of the startup? Who wrote a post because they wanted to be helpful, and now they're being publicly shamed? Come on


I had the same reaction, but then I was delighted they included these smaller examples. It was especially interesting to hear about the desktop search startup from the summer founder’s program.

There’s an inclination to view this sort of thing as adversarial / picking on the founders, especially because there are a lot of haters for successful people in general. But this piece seemed clinical, not critical.


I think Failory is meant to be an educational tool for founders. I suppose there's lessons to be learned at any scale.


We aimed to cover different sizes of business, industries, raising funds, etc. as our audience is quite broad.


Why call it "Biggest Y Combinator Failed Startups," and just randomly pick on some very small companies that weren't even contacted for their side of the story?

Like, what could anyone learn about from what you wrote about, for example, the vegan milk startup? You just picked on a random small company, and then said generic stuff about "economies of scale".

I love hearing about failures (and am happy to talk about my own!), but there's a difference between exploring what went wrong, and "punching down". Now when you Google their names, "Biggest Y Combinator Failed Startups" will be the result. Unfair.


I agree with gkoberger, you're not considering the impact on individuals for whom a Google search of their name may in the future come up with "Biggest Y Combinator Failed Startups". At least take the founder names out of the section at the end dealing with the random smaller failures.


One of the companies on the list took consumer crowdfunding money from IndieGoGo and failed to deliver, and said IndieGoGo buyers "eventually stopped receiving communication" from the company.

But one shouldn't mention the names of the founders because future crowdfunders being able to look them up for the purpose of due diligence and deciding whether to invest their hard-earned dollars in their next venture would be unethical.

(I'd imagine both the law on, and attitudes to, online privacy would be in a much different place if people in Silicon Valley exercised the same level of empathy for all those pesky end users as they always find in droves for the saintly and well-intentioned startup founders.)


This applies to only 1 of the 5 companies named in the "Other Failed YC Startups" section. What about the founders of the other 4 companies there?


Presumably, since failing in Silicon Valley is widely seen as a good thing, this should be seen as a step on the path to enlightenment.


Some of the people who will view the post will think that. Others will not.


The YC business model (because it's a business, not a charity) is upselling their abilities to investors, and the financial institutions backing those, such that they continue to put up tens of millions in series A, B, etc. rounds for companies that would otherwise struggle to raise money because their business models, tech stacks, and product market fit are questionable at best.

Part of that business model is that in case a company doesn't work out (i.e. almost always), these losses are absorbed such that the investor can spin the investment as a success to their clients (e.g. pension funds). YC is good at this to the extent that even these failures are quite lucrative for everyone involved. The message with this article is that complete failure and loss of investment is quite rare for them but of course it happens. It's basically a "don't worry, we know what we are doing" kind of message.

Any time you hear the words acquihire, what happens is a controlled shut down of a failed business. Mostly these are actually complete failures in terms of the buying company spending money on a company that then gets unceremoniously killed or absorbed into the main company at a loss and is never heard from again. Quite often this is the whole point: kill the company as cheaply as possible with as little loss of face for founders and investors as possible. Mostly, services are shut down or mismanaged, people start leaving almost right away, and most of the "money" is actually just investors swapping one set of shares for another and high-fiving each other. Somebody just lost but it's not them.

Why would companies volunteer to do that? Well for that, you just have to follow the money and basically you'll find that these companies are typically backed by the same groups of investors and financial institutions. It's a form of creative bookkeeping and investors consolidating their assets and risk such that they can continue to up-sell their abilities to manage other people's money (because it is rarely their own cash).

Startup funding is mostly a pyramid scheme with smart investors at the top basically making money from increasingly more gullible investors the further down that pyramid you go. Sometimes it accidentally produces a profitable company that actually makes it to an IPO worth many billions. These unicorns then promptly start acquiring their less lucky siblings so that those too can be spun as a great success. But even some of those unicorns are questionable. E.g. Uber is a great success. So is Wework. But are they profitable and will they ever be and does that even matter? YC is successful in the sense that their network of investors has access to a great number of such unicorns. This means most of the companies they investment in are relatively safe investments no matter how silly the company is.


Worth noting that the proximate cause of death is often times largely uninformative as to why the startup actually failed.

Most failed startups fit into the following timeline: was burning more money than it was making => failed at raising more money => did a round of layoffs / cost cuts to get economics under control => couldn't right the ship and shut down / did a fire sale or acquihire

But, the above timeline doesn't teach you much about why the company really failed. The real answer almost always involves a multitude of contributing factors, and requires intimate knowledge of the startup's business.

For the most part, the only reliable source for this type of information is one of the founders, board members, or (sometimes but not always) c-level execs. And, even then, they have to be willing to be vulnerable (which is very rare).

If I could give some advice to Failory, it would be to think about how you can incentivize founders to share their stories. As it stands, they have little to gain and potentially a lot to lose (e.g. being labeled one of YC's "biggest failures")


One of the best talks I had was a two hour discussion of this very topic with John Walecka. John had invested in Freegate and we were discussing risks and why things failed.

He helped me understand both the concept of 'stacking risk' (where the startup is taking on more unknowns than it should) and 'over burning' where a startup attempts to hire itself into shortening its schedule which not only fails but exhausts needed capital.

To this day I never hire an engineer if I cannot write down EXACTLY what they will be working on and how it will help the schedule. Even if they are a "super star", if I don't have something for them to do that will check off things that are currently on the path to the next milestone, no offer.

Whenever I've watched over hiring from inside or outside it tends to end badly.


> Whenever I've watched over hiring from inside or outside it tends to end badly.

I can relate with this. I first saw this phenomenon play out twice within a large FAANG company where the leadership wanted to ship an immensely complex product fast by throwing people and so over hired. Both products failed miserably.

And the I saw a repeat of this at the entire company level. The extent of over hire was immense, fueled by cheap investor money and their push to grow-at-all-cost. The company is now just barely managing to survive, after a few rounds of layoffs and pivots.

Over hiring is a huge red flag from the long term success perspective. However, as an individual, if you get in early on in the cycle and play it well you will rise very fast in the org chart. I’ve seen a few do that consistently. Good for them I guess.


The two are connected. Over hiring is done on purpose to game promotions. In startups for funding. Depending on the timing and where you are in the hierarchy it can go either way. People switch every 2 years now so it’d doesn’t matter how the team does in the long term.


> But, the above timeline doesn't teach you much about why the company really failed.

I for one would like to see this table with the founders' average email response times.

https://news.ycombinator.com/item?id=19375483


I'm sure the trend generalizes, but I'd be wary of cargo-culting it because a number of simple and plausible explanations put the causality arrow firmly in the other direction. Failure isn't fun to share, failure implies a need to re-think and strategize, failure means you can't delegate enough of your responsibilities to specialize in appearing quick and decisive to your investors, failure means you aren't in the startup circle anymore and aren't looking to cultivate your relationship with Sam Altman, and so on. Any one of those effects could singlehandedly create the observed trend without implying that quick response times lead to healthy businesses.


That's actually pretty awesome.

Of course, it's more of a symptom that is an emergent property than a root cause. Like saying every top tennis player uses $$$ rackets - so if I use $$$ rackets, I'll be a top tennis player.

If people knew Sam Altman judged companies more positively by their response times which resulted in more funding, that's a metric just begging to be gamed without actually making the business better.


As someone who is generally a "slow responder", I'll guess at the root cause. My guess is that being a slow email responder is correlated with (a) higher propensity to procrastinate, (b) general indecisiveness and (c) lower self-esteem, all of which are not hard to see why they would negatively impact business success. Of course, I could just be projecting, but when I am slow to respond to an email, it's usually because I want to overthink the response, or that something about writing the response is generally uncomfortable so I practice avoidant behavior (same thing feeds into procrastination). That root cause trait has proven very inhibitory for me achieving my goals.


Oh, that is 100% me. I have this anxiety related to external (non-co-workers) people. When I know it's a negative response email, it makes my heart and body tense up. I know it's purely an internal response, a personal perception/projection, giving imaginative power away, etc etc. I try to woo-sa (meditate) through it, but it's still very real for me.

Side note, for a throwaway account - you are impressively dedicated to it.


Sometimes jumping right into every single thing is not the best, ideal decisions do not all come the same way, and self-esteem has its ups and downs.

Independently, opportunity can be a real wild card.

Regardless, I think an investor can expect a prospect to be putting full time effort into building a relationship that can lead to a growth partnership, and not wanting to miss a day.

Once you get into ongoing communication you may not want to miss an hour.

If there is good allowance for urgency you're still going to need to go full professional and be able to pick up the phone.

Sometimes around-the-clock with backup.

When you can start out that way there's no need for email at all.

Might also be best at a high point of self-esteem, decisiveness, and undelayed action.

Other times when those are not all completely within reach, related actions should be in progress to more than compensate, in preparation for less inhibited times.


This reminds me of a comment by a YC partner:

https://news.ycombinator.com/item?id=20225078


Most startups fail. I don’t think most founders have the acumen to really know the full story of their own failure.


There is a difference between knowing the difference and admitting it.

When you run a business from start to scale, you likely know the main failure point, and it is probably a few simple variables.


Well the answer is probably in 95% of the cases 'not enough sales', but it wont tell you anything. Also I believe even in hindsight you dont know what was really the problem (in the sense that you know what could have been done to make it succesful)


Not sure I agree with that. Most astute founders can tell you pretty accurately why they didn't achieve enough sales to offset costs, they're just unlikely to do so for strangers because it usually means admitting at least some amount of self-fault. I know I can give a shortlist for my own failed businesses that's a lot more detailed than "not enough sales".


Yes, but is the list really true (in an objective sense). I mean why didnt you act on it, or why was it only know afterwards. (Of course the list might contain a lot of unchangable items, but this would mean the startup was impossible)


One of the real bad ones is _more sales wouldn't have helped anyway_.


Yes, I realized some answer might be: we were too slow another company took the cake. But it is difficult to say what other people can learn from it (be swift is probably what everybody knows)


Probably be much more informative to talk to non-c-level employees who do not have an incentive to modify the story.


Perhaps. There is the danger that those employees aren't party to all the relevant information, so their version of the story might be inaccurate too.


no explanation is needed for joe founder failing to achieve a $1B outcome. Better to ask why the few succeed and the answer is usually one that could not have even been foreseen.


Not really. That’s a definition of survivorship bias. The answer many times comes down to luck.

https://theconversation.com/huge-success-in-business-is-larg...


My experience (N=3) is startups "fail" mostly because the answers to "can this work" and "will people buy it" tend to be "yes", "maybe" and "no".


Also...it's possible nobody knows.

You might have some obvious situations that were big issues, but I'm not sure reason for a business failure (sometimes even success) is always obvious.


As a relative outsider, I would hope being labeled as one of the biggest failures would be a mark of experience among a set that claims or claimed to uphold the “fail fast” mentality and treats failure like the learning experience it is.


In Walk Street if you have a $50 million blow-up your career is shot. If you have a $500 million blowup you get a second shot because you must have been somebody to be trusted with that much money.


"fail fast" refers to experiments, not the whole damn company.


A new company is almost by definition an experiment.


Especially when peoples’ lives depend on it.


Looks like a universal pattern across different domains: an immediate cause of human death is usually different from the long-term cause of their health deterioration. Similarly, an immediate cause of crash is typically different from the root cause of the problem. But is there really such a thing as "a root cause"? I guess we can always look deeper.


How can you make a list of failed yc companies and not mention homejoy?

$66m raised, bad operations, bas unit economics, lawsuits and an abrupt shutdown.

All of that after being the calley darling for so long.


I used HomeJoy once, a student showed up and said "Wow your house looks pretty clean already, I'm not sure what I would even do". My apartment was tidy, but it wasn't clean (at least by my standards)


This actually highlights a big problem we had - aligning expectations of what "clean" meant between a customer and cleaner. Everyone has a different definition!


So your point is that you are unable to articulate what your standards are?


Don't forget this legendary post from the CEO https://news.ycombinator.com/item?id=8794956


Man that seems kinda sad.

If someone is in a situation where they really don't have anything else to do on Christmas, ok. But hard to imagine they don't have some friends or family missing them...


Keep in mind this business finds gig workers to clean houses. That is its mission and passion.


Yeah I was a little bit surprised by all the passion talk.

Granted I could see being passionate about building a business but ... yeah, to an extent it was kinda weird to read in that way too.


Good news, since then I've found my caps locks key.


I hope you have also found some perspective as to why the post I linked to is an absurd thing for a CEO to write.


unlikely


Maybe because the CEO is now a YC exec


After stealing HomeJoy's customer DB for another start-up, she is now a YC partner...

Move fast and break things?


She railroaded our YC interview for having a similar model (albeit in a different vertical) simply because she failed, so naturally we would also fail. Aaron Harris, another failed entrepreneur turned YC partner, has routinely done the same I've heard.

We're now 2 years later pushing $2M ARR profitably and about to raise our Series A. YC not taking 7% of our company was the best thing that ever happened to us.

YCombinator is nothing like what it used to be. The majority of the partners are useless as venture partners.


There's a phenomenon I've noticed in job interviews where the interviewer will in a sense torture the interviewee if it's a domain that the interviewer has expertise in, asking needlessly complex questions as if to grandstand and show off. It becomes theatrical and I can only attribute that to some level of sadism- or narcissism-like traits. I am not impugning Adora's motives here just commenting on my own experience.


I had an interviewer do that once, and it was clearly just to show that he knew more obscure technical info that I did, even though it was at a much lower level than was required for the job. The other interviewer even sighed and rolled their eyes at him, so I'm guessing it isn't the first time he'd done it.


Can anecdotally confirm this when I've been on the interviewer side, alongside coworkers who I am/was also close friends with. I'd agree with you on the traits.

I always felt really gross afterwards because I felt like the interviewee was wronged, but it just wasn't worth the potential conflict with my coworkers/friends if I did anything about it, either during or after. Walking on eggshells and all that.


Some mutant form of bikeshedding.


Seems like we made a mistake, and congratulations on your success!

If you're open to it, I'd love if you could email me the name of your company (jared@ycombinator.com). We are extremely interested in learning from mistakes like this.


Why would I do that?


Wouldn't you expect nxmnxm99 to find that tone insulting?


It's true if you have a capital-intensive business, I will ask some of the toughest questions. I do this to figure out where you are in your journey of learning, not to find reasons to not accept you. In fact, we all take super-extra precaution to not over-learn from our own failures.

And to be clear, I've invested in plenty of startups with similar models. But I understand your perception of how things went.


> In fact, we all take super-extra precaution to not over-learn from our own failures.

I'm curious to hear more. How do you go about this?


I don't like the phrase too much but short answer is intellectual honesty. For example, understanding what went wrong that was probably because of your decisions versus the market, customers, external forces, etc.

There are far too many examples of founders succeeding with an idea after many people before them failed with same idea. I think any good investor is cognizant of this.


In my interview with her for my B2B startup, she said (paraphrasing) "I never had this problem when running Homejoy, so why would others? I don't get it." Years later, our problem space continues growing like crazy.


That’s just what happens when worthwhile initiatives get too big and bloated. I’m sure there was a sweet spot YC was in years ago but I guess it’s no longer that.


I understand there's probably a desire to stay at least pseudo-anonymous, given that your account was created 42 minutes ago as of this posting, but I'd like to hear more of this story. Feel free to post here or contact me directly via the info in my profile.


Jumping on this because I love underdog stories.

What did the co. look like a year before YC application and a year after?


YC existed to monetize the reputation Paul Graham earned by writing two very good Lisp books.

YC today: no Paul Graham, and too many right-wingers.


While seemingly harsh criticism, I think you may be right. YC of the old (PG, Jessica, etc) was much more selective and smaller inner circle. However, even when PG was at the helm he said they passed on many insanely lucrative opportunities.

Picking startups is hard, like picking stocks. You aren’t gonna pick every winner, but frankly 2M in ARR after two years is not exactly a home-run for YC. So, perhaps them passing was the right call for YC.

Congrats though, sounds like you’ve built something useful and you can be proud of that. Plus, taking VC money ain’t all it’s cracked up to be. If you can bootstrap to get to a level that pay’s yourself $200-300k a year, that’s a win.


> frankly 2M in ARR after two years is not exactly a home-run for YC

It's been awhile since I looked at this deeply, but I thought the path of a good startup is to raise a seed with an 18 month runway, grow to 1m ARR, and then raise a series A. Assuming that's true, growing to 2m ARR in 2 years is in the ballpark.


$2M ARR is such a small part of the story though. You really have to look at things like cost of acquisition, churn, gross profit, along with understanding the market itself. Certain markets can be easy to create quick ARR either through enormous CAC spends and/or large amounts of churn, all of which means poor long term growth potential.


I guess people can brand themselves as "successful" after what is by all objective standards a spectacular failure.

I have no idea how impressive Adora is and I cannot speak to her qualifications, but HomeJoy should be a textbook example of an SV failure.


[flagged]


How is my comment sexist?


It's just you.


Ok, the situation seems to be: brother and sister cofounded startup; startup failed; brother acquired failed startup’s assets via legal proceedings.

Which part of this is illegal, or — I’ll meet you halfway — unethical? It seems like a straightforward business transaction.

If I’m missing something nefarious, I’d like to educate myself to avoid it. What do you see as the problem?

(Apologies if my facts are incorrect; this is info from elsewhere in the thread, so maybe you know something that hasn’t been said yet.)


I'm not familiar with the story, but this is what I found: https://www.businessinsider.com/aaron-cheung-brings-homejoy-...

> The weird tale begins with an email that John Salzarulo received Tuesday afternoon. A Los Angeles based user, Salzarulo received an email from Cheung that "$20 cleaning is back!" thanks to its local partner.

> "I wanted to reach out personally today to invite you to join a private house cleaning trial with our Los Angeles partner, Fly Maids," Cheung wrote, not disclosing his connection to the company.

> When Salzarulo clicked the email link, the Fly Maids' site logged him into his Homejoy account, which still had his credit card number and notes about where to find the trash can.

I don't know if it's illegal, but I definitely think it's unethical.


Hmmm.

Thank you for finding that! This opens up a fascinating discussion about ethics.

So, to start from a purely capitalistic viewpoint, it seems like you are free to use your property that you own however you wish, subject to the law. That raises questions like: in this situation, is it legal for the CC numbers to be stored in that way? From the customer’s POV, they authorized HomeJoy to store their CC info, not Fly Maids. But that leads to the question of: those CC numbers are stored somewhere (or the authorization token) and those assets were a part of the sale.

I don’t know. It’s a massive advantage to have your customers in a position of “just click this button to give us money” rather than pestering for CC details.

It’s a little odd, to be sure, but... it seems like unless it’s illegal, it might not be unethical to take advantage of that opportunity. It depends how you feel about capitalism, I suppose. If there was nothing illegal here, which seems perhaps likely, then it seems valid.

From another point of view, it sounds like he was just trying very hard to succeed, and in some sense Fly Maids was the continuation of his previous endeavor. So I sort of understand why it might have felt natural to reach out to the customers you were already doing business with.

But again, all of this has two important assumptions: (a) he legally owned all assets, and (b) used those assets to the letter of the law. If those are mistaken then someone with more experience should definitely call it out.

"When we contacted customers, we didn’t tell them we were Homejoy relaunching because we wanted to gauge reception to our new model without the influence of Homejoy’s brand," Cheung allegedly wrote. "As a result, we scared many customers, who expected the worst had happened to their data. We should have told customers upfront who we were, what we were testing, and used original content."

I dunno. This seems pretty reasonable, honestly. It kind of alarms me that you see this as clearly unethical, because I could see myself making this same mistake, in a different life. If you feel like explaining more of the reasoning regarding the ethics, I’d personally find it interesting to listen.


Not everything is an asset. You do not legally own the credit card numbers. At most you own the right to use them to fulfill a contract. Just like if I share my screen with an IT support company that does not give them the right to suddenly transfer my screen to an advertising company, except if I have been explicitly told that would happen (and no, terms and conditions are not enough for that, because it is way out of the expected behavior of the contract, just like having terms that make the user pay a billion dollar on each visit would not make it work either).


If card data isn’t property the way that customer data is, and isn’t transferable during an acquisition, then it’s hard to see how Fly Maid wasn’t in violation of the law here.

It seems like this should be readily answerable: in the event of an aquisition, does the acquirer have the legal right to use the card authorization token from the customers of the acquired startup? Note that Fly Maid did not charge them without their consent; they merely made the option available without them having to enter any CC info.

The reason I’m pressing this is because we’re talking about a YC alum + illegal behavior, which to my knowledge might even be a first.

Hopefully a lawyer might chime in with clarification. If Fly Maid was not authorized to utilize any of HomeJoy customers’ CC info during the course of business, regardless of acquisition, then this seems pretty clear cut.


I used to work alongside Adora at YC and she works super hard for the startups she is working with and is very helpful — totally justified for her to be there.


so you said it and we should buy it? Why? Because you had successful experience? What about people whose data is stolen?


In the hedge fund world there is a mindset that if you're going to fail, you must fail BIG (without breaking the law). Then you get a sort of undeserved glamor associated with the failure that can actually lead to new opportunities. For example, the case of Brian Hunter, if he failed small he would've been long forgotten.


Sounds like WSB.


I think it's a pretty universal phenomenon. You want to be either outstandingly good (best outcome) or extremely offensive (second best outcome). Just don't be another drone that is indistinguishable from the million of other drones, that only leads to being forgotten.


The person who was accused of "stealing customer data" was the other co-founder - not Adora (CEO & now YC partner).

If you actually look into the details of the matter, what happened was that the co-founder acquired the failed company as it was put through a bankruptcy process.

People who weren't privy to the process thought it was stealing when it really wasn't different from any other acquisition.


How can you say all that and not mention the fact that they are sibling? You make it sound like 2 separate person that have no relationship to each other. It looks like you have something to hide by just ignoring that big fact. https://en.wikipedia.org/wiki/Homejoy


Don't you think it's possible that the person you're responding to has a different understanding of what a sibling relationship means? I think it's a little rude to inject your feelings about what a sibling relationship must be and resort to innuendo.


> The person who was accused of "stealing customer data" was the co-founder, not the CEO & now YC partner.

So, her brother?


You're right that it may not have been "stealing" in the strict legal sense, but it was unethical and deceptive. More details here:

https://news.ycombinator.com/item?id=10466888


My comment doesn't use the word CEO so I'm not sure what point you are making.


Ok sure but what was the implication you were trying to make?


Is it true that you worked with your brother behind-the-scenes so he could buy your failing startup?

For reference: https://en.wikipedia.org/wiki/Homejoy#Controversies


A 3rd party called Sherwood was brought on to wind down the company. This is common thing for a company to do when there are lots of assets to sell and loose ends to tie up. Everything from office furniture to domains to etc. They made all the decisions to sell what to who.

If anyone wants a startup idea, it's to make the inverse of Stripe Atlas. Closing down a company properly is a very long and complicated process.


This account doesn’t explain why your brother proceeded to misappropriated the startups database, and then terminated the rebranded startup once the wrongdoing was uncovered.

http://www.businessinsider.sg/aaron-cheung-brings-homejoy-cu...

I'm skeptical a neutral 3rd-party would advocate selling credit card data to a person planning an illegal scheme. Especially when that person is your own brother.


It seems you've made up your mind on what the narrative should be. So I'll just stop here for now.


i dont know Adora and i'm sure she has useful things to teach people. however i really wonder if this clouds every interaction with people going through YC. it's kind of the huge elephant in the room.


That’s a heavy duty accusation. Source? Attached is the wired story. https://www.wired.com/2015/10/why-homejoy-failed/


It's clearly not a very complete list. I wish they had presented a more complete list, filterable by funding round, amount raised, year founded, year shut down, and maybe one or 2 other things.


Maybe something like this (https://www.failory.com/cemetery), but only for YC startups?


Will add it!


$40m but who's counting :)

I've seen the $66m figure quoted a lot but not sure where it comes from.


[dead]


You can't post slurs like that here. I've banned this account.


Anyone else thought that the list of failed startups starts with NerdPilots? That ad is a little misplaced.


I didn't realize it wasn't first on yhe list until I read your post! I though to myself, of course, why would YC fund a contract development shop, such a wrong model for VC no wonder they must have failed.


Ha!! This makes my day - I thought they were on the list of failed startups too, thought the way fact same thing as you and didn’t even realize they weren’t a YC company until I got here. :)

Thanks friend - I needed that chuckle today!


"Sponsored by" didn't do it for you


Obviously not - otherwise I wouldn’t have posted. The ‘sponsored by’ font isn’t terribly visible.


They did fund Gigster which is what the business model was in essence


Yeah, the ads were pretty intrusive in this article. Still got us all to engage, though! Some content marketer is happy today.


Right, we all associate them as a failed startup. Great marketing!


Can we talk about NerdPilots CSS mastery ? https://ibb.co/vZHS7xT (on firefox, laptop)


Yup, I was "oh they failed fast!" and then realized it was an ad.


Interesting that two of these (three if you count Unicorn Rides which is a note within Boosted) are hardware companies and both were a result of the recent tariff war with China.

I've always known startups are hard but it seems that hardware startups, even when they appear to be on a rocketship trajectory, are nigh impossible.


Hardware is often a lot more capital intensive, harder to get to an MVP, and, these days, often requires software in addition to hardware.


And that’s the reasons why all the capital sources avoid hardware startups as hard as they can. It’s like betting for a 10x slower horse than the rest.


This article says Atrium ran out of money, but I feel like Justin Kan’s blog post said they gave tens of millions back to investors when they shut down.


Yeah, I commented that the list is incomplete and at times straight up wrong. This is one example. Atrium absolutely did not run out of money.


Can't help feeling like this article is doing their sponsor a disservice.

> Here are some of their biggest failures, sorted by funding. [...] NerdPilots

My first thought was "I've never heard of NerdPilots, but they must have been awful" before realising that it was an ad.


This list seems to be a biased list. I thought the list was first in descending order (funding), but it's not. Some super early stage startups were picked for some reason in the bottom section?

To clarify: by numbers, the vast majority of startups fail, and that's part of trying new things until one of them works great. I know multiple founder friends that had 1-3 *big failures before they built a great company after that.



Missing from that list is Grin Scooters / Grow Mobility (e-scooters for LATAM) which is probably one of the biggest failures out of all YC companies in terms of investment money raised.

They raised $70M as Grin and then $150M after the merger with Yellow. I think at Demo Day they were trying to raise at a $100M cap with just an idea / pre-launch. After raising an insane amount of money in a matter of moths, they expanded extremely quickly to dozens of cities in several countries in LATAM while having bad unit economics ("land-grab"). There are likely interesting learning lessons there.


> suspicions uBiome was billing its client’s medical insurance providers multiple times without the prior knowledge or consent of their clients

> Growth-hacking on the wild-west which is the internet is one thing, but when you try to use similar tactics in traditional, heavily regulated industries in developed, heavily regulated markets - that’s a recipe for disaster.

Was this growth-hacking or outright fraud? Are those the same thing?


Outright insurance fraud.

https://www.cnbc.com/2019/05/06/ubiome-suspends-clinical-ope...

>In the months before the FBI search, the company would often bill insurance multiple times for its tests without patient consent, insiders told CNBC. In an interview, one customer, Marc Harris, told CNBC that he sent in two samples, but the $2,970 test was billed to his insurance five times.


It happened to me. Billed my insurance double and sometimes triple for each test.


To all those who failed (apart from the weird sounding fraud with the FBI), well done for actually climbing into the arena and fighting. Better luck next time.

To the rest of us. Schadenfreude is its own reward.


That's not why Meta failed. Meta failed because their product was garbage.


Ahahah, I dropped in here to share this little tidbit too. Looks like you beat me to it. I figure it's been long enough that I can share some war stories.

I was in the car w/ the founders on the way to Hacker Dojo when they got the news that they'd been accepted into YC. At this point much of their tech demo had been built by some kid they'd hooked up with at the Dojo. One time I remember one of the founders was on the phone with his mom trying to get more time out of him. I never saw the technology deliver on any promises but I was probably like 22 at the time and I was stoked to get to play with depth cameras and AR goggles at all. I helped them film some tech demos and build some jigs to take footage through the googles. It's through Meta that I met Steves Mann and Feiner. I had a few great conversations with them at the time, and later it also made me realize that brilliant people who you deeply respect can make some really dumb bets too.

The best way to describe Meta is "store brand Magic Leap". Huge gap between what they had and what they were selling, and never enough focus to put in the hard work to narrow it, or the humility to admit that they overpromised. They did eventually pay me, though I think some friends of mine were not so lucky. I would say it was mostly incompetence and not malice. I think the founders truly believed many of the lies they told and it was easy to get caught up in that energy. Working with them and hanging out in their general orbit at ChezJJ and the Hacker Dojo was probably the most HBO's Silicon Valley experience of my life and I'm deeply grateful for it.


I will have to agree. The tarrif war accelerated their demise; but there product was far from ready. Infact the CEO of Meta said at some point said that employees will be required to shift from monitors to using Meta and they werent able to reach that goal because their tech (or AR in general) wasnt there


Has any AR product not been garbage?


The HoloLens 1 does what it claims to do. It is uncomfortable, the FOV is narrow, but it's usable and there aren't any big, negative surprises. You can use it and say, "yeah, I get it, I can see where this is going".

The Magic Leap doesn't do what they claimed it would do. But it does do what the HoloLens 1 does, slightly better in some areas, slightly worse in others.

The Meta2 was fundamentally broken. Nothing on it worked correctly. Everything on it was a hack job.


I can say Pokemon Go was a great success. I never played it because I'm not into AR, but I like this vertical of positive social/gaming experiences.


I guess it falls under the very vague term of "AR product", depending on how you define it, but it's not an AR visualization product. Rendering pokemon on top of a surface is a cute little gimmick but if that feature wasn't there it wouldn't really matter.


but none of the CEOs say that you have to switch from using monitors at the company to AR glasses


This list feels wrong and incomplete. The startups at the end are completely random. If you included the full list of failed YC startups regardless of funding it would be hundreds deep. I don't understand what I just read.


No list of YC failures is complete without mentioning Loopt.


I remember seeing that Buttermilk company arguing with people on their Facebook ads.


How's Y Combinator doing these days?

Many smaller base hits, a few home runs, or striking out? I don't follow the VC world much.


Just this week:

- Momentus: $1.2B IPO (https://www.cnbc.com/2020/10/07/momentus-the-latest-space-st...)

- Segment: acquired for $3.2B (https://news.ycombinator.com/item?id=24735414)

- MessageBird: reached a $3B valuation (https://techcrunch.com/2020/10/08/messagebird-series-c/)

More generally:

- 15-20 YC companies are now worth $1B or more.

- 100ish YC companies are now worth $100M - $1B. A good number of those (and others currently worth less than that) are growing very rapidly and will likely reach unicorn status in the next year.

So it seems like one can reliably expect 2-3 unicorns to come out of each YC batch. Those 2-3 unicorns likely more than make up for the costs associated with running each batch (including all of the ~$150K investments). Seems quite good to me.

Admittedly, there hasn't yet been a Google/Amazon/Apple/Microsoft-scale company that has come out of YC yet. I think it's only a matter of time until that happens -- anecdotally it seems like at least a handful out of the recent batches could become massive home runs, as well as other more established companies like Stripe that seem to be growing quite well.


I'd definitely consider Stripe in the running.


I am an outsider to this site. Found it trying to find something like reddit was around Mt Gox timeline. Posted my Startup to this forum. Never got much traction. 6 months later though Acre.com started much better funded than I. Is Acre a YC startup? Now i just follow on and read. aka old scool reddit lurking


I really don’t think Nerdpilots wants that sponsorship spot. Thought they were the first example of a failure


Kind of surprised not to see Homejoy on this list. After failing to understand three sided marketplaces, posting a widely mocked Christmas blog post, Homejoy flamed out, and founder Adora Cheung landed as a partner at YC... for some reason.

Kudos on failing up.


It's very unfortunate The Buttermilk Company closed down. I was looking for them recently and found the founder's posts about how stressful it was to run. Does anyone know any similar products out there?


Just buy MTR or Haldirams ready to eat if you want instant Indian food. More authentic and tastes nice.

I haven't tried them as I'm not from the states. I'm Indian though and found the buttermilk company products to be highly priced and didn't find anything novel when compared to Haldirams or MTR. At their price you could probably get food from a takeaway.


The main thing I was looking for was something that tasted pretty good and was shelf stable and easy to transport. Haldirams and MTR look great! Thanks for the pointers.


Shameless plug: I research exactly this, but from a different perspective - namely the personality traits behind productive/unproductive decisions that ultimately lead to failure or success. If anyone interested: https://www.dropbox.com/s/w31inj3xf4aiu66/Successful%20Start...


A really good idea. Not at all the same social/economic climate as it was at the time but I couldn't help remembering https://en.wikipedia.org/wiki/Fucked_Company which really lent some perspective to industry when it was so hard to understand where the hell all these dotcoms came from and why not all of us were getting the millions.


How is Docker a success from YC? Pretty much the opposite no?


Has it officialy failed yet?


"The Trump administration’s trade war with China resulted in a failed funding round led by a Chinese investor, which put the company in dire straits financially. "

Trump is a convenient scapegoat here but in no way related to why they failed. Meta was headed to guardianship not long after the Meta2 was released.

Overall Meta was just a poor experience with basically zero traction. They changed their hardware approach from the Meta1 to the 2 and it created a completely different experience while on the side the Hololens was blowing everyone out of the water and they weren't in any position to compete. Add to that the complete collapse of Magic Leap and the bubble burst for AR this round.

Reinforcing again - Augmented Reality is probably the hardest market/technology to succeed or even just stay alive in.


Many startups are on their deathbed before finally succeeding. A famous example is Fred Smith, FedEx founder, gambling the company's last $5k, in order to come up with $24k for a deadline[1]. Had the cards come up differently, or a political wind had made gambling illegal, FedEx would have failed, and someone would have been able to say that logistics was "a hard market to succeed or even just stay alive in", and even be correct! The casino (or lack thereof) would have been a "convenient scapegoat". But it doesn't change the fact that having the chance to have these kinds of lucky turnarounds seem to play a pivotal role in a lot of success stories.

[1] https://www.businessinsider.com/fedex-saved-from-bankruptcy-...


Quick note: you got the pronoun wrong for the founder of The Buttermilk Company when referencing her Medium post.


Current title is misleading: Y Combinator Failed Startups.

Why did it change? This is a discussion of the biggest failures out of thousands of startups, not that YC is horrible or something.


It's not misleading, though I can see why

It's not YC "failed" startups, it's (a list of) YC "failed startups".


Something like “Failed YC startups” or “YC startup failures” would be a less ambiguous title.


This is my first time hearing about uBiome. Oh well, at least I have one of their t-shirts from some startup trade show thingie I went to. It's pretty cool looking.


Imagine every failed startup would open-source their stuff so others could stand on their shoulders if they want to give that idea another shot.


I wonder who the founders were for Plurchase. It's kind of weird that they're the only one without this info listed.


I wonder how many could succeed if they started at a different time.


Which reminds me : only 13 days to when the interview invitations will be sent.


airware?


"The Trump administration’s trade war with China resulted in a failed funding round led by a Chinese investor..."

Is there another president that's so callous about small medium sized businesses?


I have this question on my mind and if someone could explain this sincerely I would be very grateful.

Time after time, YC startups seem to engage in practices which are unethical or dark or downright scammy. Be it Homejoy, uBiome, AirBnB (updating consent without asking). They seem to employ aggresive marketing tactics. They work under assumption that "law does not apply to us". Although this is true for ultra rich people, why regular seed funded do it? Is growth focus this much necessary to the to do outright fraud or in engage in shady tactics? They are willing to kill other people's business and livelihood for their growth. I would call this "preying tactics"

Paul is very successful person and I read about him a lot. Why companies going through YC do this? Or is this just Business-as-usual (BAU) in Americas? Kill or be killed?

Thanks for responding!


[flagged]


Would you please stop posting crap comments to HN?


Personally I find the call to actions rather tasteful on that site


I like how stealing is called growth-hacking now!


It looks like TheMelt is down to 7 grilled cheese stores.

https://www.themelt.com/locations


To be fair, Covid is a nuclear bomb dropped on the entire restaurant sector.

However:

1) Putting a grilled cheese sandwich restaurant in ferociously expensive real estate known for upscale, health-conscious consumers can't be smart (Irvine Spectrum Center?--LOL)

2) The food just wasn't that good. Sorry. They deserve to go out of business.


TheMelt wasn't in Y Combinator, but it is an interesting story.


I'm sorry for linking the two. I have had this wrong idea stuck in my head since 2011. I should have validated that fact.


What's the story?



Why do they need 7 locations? They should downsize to one location with an army of toaster-drones.

Just make some specially-designed plates with a QR code target printed on them and the drones won't even need to land, they can just drop the food right on your plate after toasting the sandwich into a more aerodynamic shape...


Did they go through YC? I can't find anything online about that, but I had no clue they raised $10M from Sequoia. Surprising.


Ah yes The Melt Index. You can track silicon valley startup activity by how many tech workers are grabbing a quick grilled cheese sandwich while their code compiles.


Does Theranos count?

I'd say they achieved even less than most start-ups.


No, they don’t count, because they were not a Y Combinator startup.


Why would Theranos count? They were not YC.


Theranos wasn't a YC company


I have this question on my mind and if someone could explain this sincerely I would be very grateful.

Time after time, YC startups seem to engage in practices which are unethical or dark or downright scammy. Be it Homejoy, uBiome, or many others. They seem to employ aggresive marketing tactics. They work under assumption that "law does not apply to us". Although this is true for ultra rich people, why regular seed funded do it? Is growth focus this much necessary to the to do outright fraud or in engage in shady tactics? They are willing to kill other people's business and livelihood for their growth. I would call this "preying tactics"

Paul is very successful person and I read about him a lot. Why companies going through YC do this? Or is this just Business-as-usual (BAU) in Americas? Kill or be killed?

Thanks for responding!





Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: