Hacker News Reader

Comment by kamikazeturtles | Hacker News Reader

It seems the first few batches of YC were VERY successful. But, the number of successful YC companies I can think of from the last 10 years of batches is very minimal despite them investing in thousands of companies.

Makes you wonder, does the funding landscape need a new YC level disruption because YC seems to have peeked, return wise (not hype wise).

Maybe AI might be their saving grace if everything works out

View on HN

I'm always surprised people just seem to discount timing.

The mid/late 00s came with the rise of "Web 2.0" and mobile that was a gigantic technological shakeup, and YC was definitely in "the right place at the right time", and their Web 1.0 pedigree and expertise transitioned perfectly to these new technologies. Importantly, these technologies required relatively little capital for new startups because all the infrastructure in terms on the Internet, GPS, mobile, etc. had already been built.

I'd just argue that in the past 10 years there simply hasn't been nearly the "greenfield" tech landscape that existed in the mid 00s. AI obviously has been a tech sea change, but so far the contours of that change just look much harder to monetize. The frontier models are becoming more and more commoditized, and everyone else is just trying to figure out how not to get shoehorned as "just an OpenAI/Anthropic wrapper". Looks like there may be some winners that are really putting all pieces together in an attractive package (Cursor and Glean come to mind), and it's fair to wonder why YC doesn't have one of these "AI app winners" (and maybe they do, or will, I don't know), but in general, I think there is just nothing like the opportunity there was in the late 00s/early 10s to gobble up a big part of the new tech landscape.

There are many factors (YC needs to allocate more capital, not founder-led anymore, needs to fill more batches...) etc.

But another one is that the first batch has had 20 years to grow and compound while the rest has had less time to become massive.

I'm not saying YC is perfect now (though it's still great). But I think a simple factor could just be time for many portcos.

Or that most of the largest opportunities have been filled and newer companies are aiming at much smaller niches, or to be alternatives to well-established competitors.

Also, in 2005 there wasn't that much of an acquisition ecosystem. I feel like fewer founders back then actually thought about getting acquired and instead shot for building big companies.

You had to go big or your thing would fail, whereas nowadays you can fill a tiny niche and get rich by being acquired by the incumbent you're a complement to.

The era of "any promising startup gets swallowed by big tech" seems over with antitrust, but M&A is still happening.

Agreed, to compare you would have to not just look at current magnitude but rather compare the historical trajectories of growth

Companies take time to become successful, so the verdict is still out on most of the companies that started in the last 10 years. Time will tell.

Though there are already quite a few successrul YC companies from the last few years. Full list of top YC companies: https://www.ycombinator.com/companies (filter by "top companies")

I believe median time to unicorn status is now about 7 years, so this is not as uncertain as you think. In other words, 50% of YC's 7-year old unicorns have already showed up. The number is higher for older companies.

Dropbox and Airbnb were superhot companies within 2 or 3 years after YC. My first Airbnb booking was in 2013, and that I felt like a late adopter.

Yes, but average time to unicorn status has been increasing since the time of Dropbox and Airbnb. I assume you can draw more inferences from companies that are earlier-stage, too.

If you read their tweets, YC's message is that their companies are growing faster than ever

Just looking at how most have done since going public , most VC backed companies including YC are just exiting and finding the “bigger fool”. YC has done phenomenal for itself. For the public that bought at IPO - not so much.

https://www.marketsentiment.co/p/the-yc-report

If you invested equally in all YC companies on their IPO day, you would now be down 49% compared to the +58% return of the S&P500! 14 of the 17 companies lost money for the investor, and 3 companies lost more than 99% of their IPO value. Only one company ended up beating the S&P 500 benchmark

> Maybe AI might be their saving grace if everything works out

Most of their recent communication is going all in on AI. Saying stuff like this batch prefers hiring devs that use AI, and 95% of their codebases are all AI generated.

Of course? On average, companies that are still standing after ~15 to 20 years are going to be a lot larger than the companies who started more recently.

A) Very true!

B) peaked* :)

C) Can you think of any tech companies from the last 10 years that were successful? The answer is ultimately "they exist, but the victory condition is to be bought by MANGA, so you don't remember them even if you heard about them in the first place."

EDIT: to add more context, this is a great post: https://jaredheyman.medium.com/on-the-176-annual-return-of-a... The TL;DR is that Stripe & AirBnB (and to a lesser extent DropBox, DoorDash, and Instacart) were absolutely massive wins that they have yet to (and shouldn't expect to!) replicate; otherwise, they're doing just fine in terms of ROI.

Yeah it's a good question to ask:

Was the startup success rate in 2005 overall higher or was it actually due to YC selecting so well.

Also, success is relatively defined. A $50m exit might be something that gets washed away after a TechCrunch article and a few social media posts.

But that's still life-changing money for the founders. It's not changing the world writ large, but changes the world for a few people.

= Besides, there are many successes we never talk about. Wiz is worth $32B (at least Google was willing to pay that), but rarely talked about because they're a "boring" cybersecurity infrastructure company.

yeah. That YC hasn't produced another Google or made a Facebook should in no way mean it's an utter failure.

> Can you think of any tech companies from the last 10 years that were successful?

OpenAI and Anthropic come to mind

10 years, not weeks.

Neither are profitable.

That’s… fair, I guess. Both are majority controlled by MANGA, but they are indeed new companies. That’s kinda illuminating: it takes the breakthrough of the century to establish a new name these days, and even then you’ll probably end up selling slices of yourself to the giants.

> tech companies from the last 10 years that were successful?

Discord, Anduril, CoreWeave, Veho, MoneyBox, Peak, Marshmallow.

Discord is cool! Anduril is both evil and overvalued. The rest… I mean, I think the names speak for themselves ;)

what do their names say to you?

It's because every market inefficiency eventually disappears as entrepreneurs capitalize on it. It's like there's an optimal architecture for the economy at a given technology level, and the job of the entrepreneur is to find that architecture and refactor the economy, piece by piece. And if they're successful, they get to take everything of value in the old, obsolete parts of the economy that get destroyed.

When YC was formed the Internet was young; it had just dramatically changed the equilibrium point for the economy, but people didn't know it yet. YC's investment thesis was that many more startups could be founded than were being founded, so they set about funding them. And they were right, and so their early investments found fertile new markets that were ripe for disruption.

Now founding a startup is mainstream, so wannabe entrepreneurs have picked clean most of the big profitable market niches. And that's why all the later YC startups tend to be smaller and less successful. The big successes already happened. I could see AI spawning a few more successful startups - it is a technological change, after all - but not to the scale of the Internet. (Honestly, I see solar and genetic engineering as spawning more. Freeman Dyson wrote a prophetic book in 1999 entitled "The sun, the genome, and the Internet" about the 3 technologies that would define the 21st century, and IMHO he was totally on point.)

And when the new equilibrium point involves a huge amount of centralization and off-shoring and therefore wealth concentration and loss of livelihood, the proletariat becomes angry.

In that context it's a fiendishly genius move to atomize and distract the "obsolete parts of the economy" and have them vote for their own disempowerment.

It's maddeningly obvious and feels unrelentingly deterministic. If the plutocrats fail to properly cement their position however, we already know what happens next.

The alternative is that it takes x > 10 years to become very successful. But that seems somewhat doubtful.

Well Dropbox (ycs first big success?) was founded in 2007 and only went public in 2018

Reddit (yc 2005) only went public last year

I know "going public" doesn't define success, but it does show that companies can still be growing after 10+ years

You might find this blogpost eye opening: https://www.lennysnewsletter.com/p/pulling-back-the-curtain-...

Wow, there's so few solo founders, and so few non-US companies. I bet nowadays, there's very few non-AI companies. So if you aren't a US-based AI-startup with multiple founders, perhaps ycombinator is not for you.

If you have a cofounder it forces you to get in the habit of delegating really early on. Prolific solo founders like pieter levels just do everything themselves, which is great for their ego but not enough to build a venture scale business.

Solo founders can have employees.

They have a disproven ability to find people and rally them around a common cause though. Whether it's due to their personality or the idea or some other thing like timing, the inability to find a co-founder could be considered signal, for some.

These are two well-known biases of YC. I believe the reasons are the following:

* YC founders are usually very young, and young people can be capricious. They don't want a solo founder just deciding to do something without a second founder having some leverage to protect YC.

* YC is in the US and has a very pro-US perspective, so they invest in what they know and like.

Non US companies — but many non-US founders. It’s much better to be a U.S. startup than in most other places.

https://capbase.com/why-european-startups-register-in-us/

YC has a known bias towards multiple founders. They accept solo founders, but solos have to stand out more or maybe already started their business. YC even provides a co-founder matching system (of unknown efficacy).

PG said "Empirically it seems to be hard to start a startup with just one founder. Most of the big successes have two or three. And the relationship between the founders has to be strong. They must genuinely like one another, and work well together. Startups do to the relationship between the founders what a dog does to a sock: if it can be pulled apart, it will be."

Also see: Does YC fund solo founders: https://www.ycombinator.com/library/7P-does-yc-fund-solo-fou...

I think all the recent AI slop dominated batches will be their downfall in 2-3 years time when they all become junk and the AI-hype is over.

Ya that is my fear. Just like the whole crypto dominated batches. My fear is Garry Tan is leading YC and he falls for every single "the current thing" in the book. I remember when he was pushing for so much of the crypto stuff and all of that evaporated to nothing.

Yep, since Garry took over YC has fallen off drastically in my opinion

> he falls for every single "the current thing" in the book. I remember when he was pushing for so much of the crypto stuff and all of that evaporated to nothing

Isn't this what an early-stage VC firm is supposed to be doing? Lots of investments in companies working in whatever the hot new tech may be. Most companies lose, some break even, one wins so big you forget about the rest.

High risk, high reward, high volume is always how I thought of a firm like YC. And one expects a lot of investments to evaporate to nothing for such an industry.

I do think YC has become chasers of the latest trend rather than creating the trend. That is why most of the successes are skewed early like Reddit and a bunch of other companies which defined the internet.

Now they have gone from crypto grifting companies and AI hyped companies and it really seems that the a-level players are missing.

Or maybe it's just a matter of smaller teams and focus and care.

I think now I see YC more like a job for the partners than a passion.

I might be wrong but there is definitely something to it that YC has not had a success for a very long time.

It's really sad because YC is one of the best things to happen to the tech industry.

Well it’s a function of those that apply and are accepted, and pitches are going to reflect the current hype. YC occasionally puts out a call for applications in areas they think are interesting but perhaps they haven’t seen many compelling pitches for. Have they recently put out a call for specific areas of interest?

Yeah I think they do it every batch or at least every year. https://www.ycombinator.com/rfs

Thanks, I see that one is full of AI, but I notice they are guiding toward specific areas, probably so they don’t get a thousand vague AI pitches

Except now their call for startups are focusing on tech solutions (use LLMS to do x), instead of focusing on problems that need solving. They're now doing exactly what they've railed against, solutions in search of problems

> That is why most of the successes are skewed early like Reddit and a bunch of other companies which defined the internet.

Reddit rode the wave with zero innovation. Really what is Reddit? Message boards predated Reddit, which was predated by Usenet, predated by BBS. Somewhere in there were Yahoo! chat forums.

All they did was colocate message boards in one place, and allowed freewheeling moderation. That's it. They got lucky, made millions with zero new ideas. People connected the same way on AOL 15 years before.

Like most Silicon Valley innovations, you just reuse an existing idea or resell polished turds as new.

and before that, there were literal community message boards. corkboard at the grocery store. people gossiped before there was Facebook. the only real innovation was going from verbal communication to the written word, but now with TikTok, we're back at verbal communication, just with a lot of extra steps. there's no other innovation that's happened in the world since writing, isn't there?

Reply on news.ycombinator.com