Hacker News new | past | comments | ask | show | jobs | submit login
How YC Has Changed (ycombinator.com)
181 points by katm on Sept 28, 2018 | hide | past | web | favorite | 55 comments

I've been going to Demo Day off and on since 2007, and one big thing I've noticed is a change in who gets accepted. I see a lot more "sure bets" and a lot fewer "long shots". When I go to demo day now, I see a lot of companies that are fairly interesting, founded by an amazing team with a lot of accolades and/or exits or previous experience, and most of the companies have already launched and have revenue.

What I don't see a lot of anymore are two people who just finished college (or maybe they didn't even finish yet) working on something that could be amazing or a spectacular failure, who haven't launched yet or have no revenue yet. Look at Dropbox, reddit, and even your own startup, JTV, which of course morphed into Twitch. All of them founded by guys just out of college, all of them now worth over a billion dollars (and unless I'm mistaken, all of them pre-revenue at demo day). And while AirBnB was started by guys who had a couple of years work experience, they didn't have much experience with running a company or anything other than a few years working in industry (and of course also worth well into the billions).

It feels like for the most part the risk analysis has shifted from "a couple of folks just out of college with an amazing idea" to "a couple of folks with a proven entrepreneurial track record and an amazing idea".

I'm not saying it's good or bad, but it's definitely different.

We still fund a lot of young early stage founders. I think there are two things that make them stand out less on Demo Day. First, batches are much larger and the average age of the batch has increased. Second, these founders are much less likely to be starting consumer startups and therefore they blend in better with the other startups.

Another important change over the past 10 years is that many talented young potential founders are being vacuumed up by the big tech companies once they graduate from college (or before in college internship programs). Essentially, the big companies are pre-hiring their future competition. The combination of these actions and the strangling of distribution channels (messaging, social networking, and email) is preventing consumer startups from getting a foothold. And by and large young founder tend to be attracted to consumer startups.

We are doing a number of things to counteract the recruiting efforts of big companies including YC Early Decision: https://blog.ycombinator.com/early-decision/ and YC Startup School: https://www.startupschool.org.

> And by and large young founder tend to be attracted to consumer startups.

This, and everything else you said related to it, are an excellent point I hadn't thought about, and maybe that's why it seems different. Thanks for pointing that out.

> We still fund a lot of young early stage founders

Examples from this batch?

It's fairly obvious in the application form itself, If I remember. They do take folks without 'traction' but there are too many applications now I feel YC find it difficult to bet blindly.

The obvious solution then would be another YC like or a sub-brand from YC that do not accept startups with existing traction, as they are just looking for a boost instead of real hand holding. I feel YC realized this and they are doing programs like startup schools to accept/help more startups.

But a long term solution IMO is there should be more YC like orgs in every country/city to really nurture and help the ecosystem. The theme of accelerators should also vary based on the location and not just try to replicate the US model. For e.g. In US, it make sense to start product companies that have global vision. While in China, or India it make sense to promote ideas to solve local challenges or B2B products that do not need the media glare that SV companies get/enjoy. Also in India and other south-east asian countries, it might make sense to accelerate companies that can provide quality outsourcing experience to small/medium product companies in the west. I am yet to see a global startup-friendly outsourcing behemoth when there is an opportunity for same.

Also I feel if a company is not accepted in a cohort, they should be given weightage in the next cohort if they are hanging around and have bettered their stats.

Having been through YC in the second batch in 2006 - their first one in California, and again in 2011, the differences were indeed pretty stark. In 2006, there were 8 startups. We knew each other well and we met with PG at his house a few times which obviously was pretty cool. He was also the one actually doing the cooking at the dinners! There was also lots of interaction with Trevor and Jessica and (less so) Robert. In 2011 there was no such type of interaction as they were just trying to figure out how to scale with 64 startups in the batch, outside money being involved, etc.

The other big difference no one has mentioned is the speakers that came to the dinners. They get the highest profile speakers now, but in 2006, the speakers would meet with each startup after their talk. You could demo to them and they'd give you feedback and personal advice. I distinctly remember a talk by Ev Williams about his failures at Blogger for example. Very raw and emotional. Some of the lessons I carry with me today, 12 years later, were from a few of those talks and after dinner conversations. In 2011 (and on) you're not getting close to one of those speakers unless you just so happen to get to sit by them at dinner.

Before YC, startup funding was truly wild west with too many term sheets with bad deals for founders. So kudos to YC for setting standards for founder-friendly term sheet and continuously pushing funding amount up. Some of the things I wish were different:

* Eliminating the whole "batch" concept. Ideas don't tend to born at YC deadlines. There should be a way to continuously accept applications as they arrive.

* Moving away from CA as center: Being in CA costs way too much for startups. I think startups should come to CA for hiring (may be) and then go away to places were things are dirt cheap, distractions are minimal and internet is fast. This alone can extend runway for many startups by an order of magnitude.

* Less discrimination against old veterans: PG had once mentioned that he typically doesn't prefer people who have worked for 10 years at BigCo because a true founder would have left long ago out of frustration. I think this is misguided line of thinking in many dimension. Even if it was somewhat true, it would be hard to rigidly generalize on entire population.

> * Less discrimination against old veterans: PG had once mentioned that he typically doesn't prefer people who have worked for 10 years at BigCo because a true founder would have left long ago out of frustration. I think this is misguided line of thinking ...

Great point - I have thought of doing a startup. I am even "worse" in terms of experience, 20 years at big companies, in various leadership roles the last 10 years - no I'm not a vp or president.

But I could even self fund my company for a few years. Being in Seattle I probably have less connection with startups than in SF, but I have even worked for two! I did do some early stage funding for a friend's company that eventually failed. As an experienced person who has been fortunate, I could actually work for free for a few years, and could provide early funding for my company. I can actually tolerate risk - how come this isn't an area of attraction for vc funding?

> In 2007, if you had a B2B startup, you had to reach out cold to potential customers. Now there are over 1000 YC companies you can sell your service to and all of their contact information is in Bookface. As a result, YC alumni companies are often over 50% of your new sales during YC.

This alone would be an interesting angle if you consider the equity given to YC purely marketing and sales spend.

Network Effects, cool!

I get this vibe of FB back when you could connect to people just in your classes. I mean, if corporations are people, then they should have their own version of FB, right? =P

Hey it's Friday and I always wanted to ask: what do the people at YC think about bringing back some kind of smaller-scale investment for a larger number of applicants?

There was a time in my life after I graduated in college in '99, after the dot bomb, before the housing bomb, before YC, when I could have made just about any app/website in 6 months for $10,000. I wasn't driven by money, I just needed basic living expenses for enough runway to get ahead before someone snagged me back into working another day job to make ends meet. To me, microlending to this vast pool of untapped potential was the basic premise of YC and what I found so revolutionary about it.

Would you consider breaking one of your $150,000 investments into 10 or 15 smaller investments? Why or why not? Thanks!

That is what we are doing in Startup School - https://blog.ycombinator.com/announcing-startup-school-2018/

I was half regretting doing last years startup school (as opposed to this one) -_-

At the same time, those AWS credits went a long way for me -- I'm sure that grant could go very far for others. Kudos for doing it!

Honestly, the most valuable aspect of startup school is the feedback from your peers. Half of my group I still stay in contact with (even intro'd them to a few customers). I guess, it's similar to YC; the networking effects.

I do think the money would be better upfront, as opposed to at the end. However, I guess there's no bar on entry and it's "free" so, I can see the reason to wait.

I suspect it costs them ~10,000$ to hand out XX,000$. If nothing else they want some due diligence to avoid harming their brand. On top of that 10,000¥ in 2019 does not go as far.

> I suspect it costs them ~10,000$ to hand out XX,000$. If nothing else they want some due diligence to avoid harming their brand. On top of that 10,000¥ in 2019 does not go as far.

10,000 yen is only $87.96, so I expect not. :P

It's the world that has changed. Launching into a relatively open world wide web is fundamentally different to breaking into the consolidated platforms of today. Startup culture was at its peak in the PG days today it's closer to a nadir.

> Startup culture was at its peak in the PG days today it's closer to a nadir.

Yes, totally. I became interested in startups in 2009. I was a CS major and there was a tiny group of other students who shared my interests - we had a weekly "internet entrepreneur society" meetup where we'd discuss Paul Graham, Eric Ries, "Getting Real" (37Signal's book), and so on, along with the various projects we were working on and ideas we had.

By the time I graduated in 2012 the field was already starting to change. In the first two years out of university I worked with 3 different co-founders on various ideas, none of which went anywhere (frustratingly). Two of those co-founders had no genuine interest in technology or innovation -- I believe that they only became interested in startups after they became the hot thing to do. They wanted prestige and they wanted to raise funding for the sake of it. They weren't excited by the potential of the new platforms that had just appeared.

(After that I ended up working for a medium-sized tech startup for a while, got disillusioned and burnt out, took some time off, and then spent the last 3 years building myself back up. I still want to take a real shot at starting a startup, but the industry has changed so much now. The fun and excitement has gone).

I think there were several factors that changed startup culture:

1. The technology of the web matured. We are now in the Gartner cycle "plateau of productivity". Once upon a time you could upload some PHP scripts and a little bit of AJAX and have an interactive app that was admittedly crappy, but also genuinely novel and interesting. Now you need to know a complex backend framework, the modern JS ecosystem, automation/build tools, etc, just to compete. That's fine for a 10-man dev team, for 2 founders out of college it's a daunting amount of work.

2. Startups became a much more popular and prestigious career path. "The Social Network" movie (2011) made startups much more glamorous. Tons of articles were boosting "disruptive innovation". By the time I graduated, suddenly all the kids who'd been polishing their IB/consulting applications decided they wanted to be entrepreneurs, maybe after 2 years at Goldman/Bain/etc. Some of these guys were genuinely talented and cool. Many were arrogant and douchey, had a shallow understanding of tech, and treated their technical cofounders as code monkeys (yes, I had a few bad experiences).

Many interesting startup ideas (Apple, Microsoft, Facebook, Google) came from young technologists working on ideas they thought were cool, yet were niche and weird at the time. Once the IB kids started entering tech en masse the culture shifted. (Granted the Winklevoss twins tried to recruit Zuckerberg back in 2004. So this is not completely new. But I think the techie/suit ratio changed dramatically in the early 10s).

3. Lots of cheap capital. This meant more ideas funded, ideas that got funded raising more money, and more market saturation in web, mobile, everywhere. In some ways this helped entrepreneurs, but it also made fundraising more necessary.

4. A controversial one: all the "sexism in tech" controversies that kicked off a few years ago. I paid close attention and I remember that this exploded in 2013. You had things like a Github founder being ousted (along with Github apologising for their "meritocracy" carpet), people being fired for tweets, etc. Sexism is wrong and there were probably many invisible barriers to female founders. However the big social justice activists (Shanley Kane, etc) were not concerned with technology, they were motivated by ideology, and they created a very hostile atmosphere. Make an edgy joke on a blog post intended for about 100 friends to read? Suddenly 50,000 people on Twitter have read it and are calling for your head. PG made a poorly-worded comment during a long interview -- suddenly every media outlet quotes it out of context, takes it as proof of his sexism, and it becomes untenable for him to remain in charge of YC. (I know this wasn't the only reason he stepped down, but it was a factor).

5. Regulation. GDPR is the biggest one, though there are others. It is not a startup killer, startups can still operate, but it makes it much harder. The idea of putting a prototype online, getting some early users, and rapidly iterating based on feedback is now much more dangerous. You need to invest heavily upfront to become compliant (even then your legal status will be unclear), and if you pivot, or make drastic changes, you bear the risk of becoming non-compliant. The EU legislators assume that everyone is a big company with a stable business model, not a 3-man team experimenting with various ideas. These regulatory changes went hand-in-hand with increasing political hostility to tech entrepreneurs. Say you manage to succeed despite all the obstacles put in the place of tech entrepreneurs in 2018, and become the next Mark Zuckerberg -- you can expect to be treated like the current Mark Zuckerberg, that is, have everyone hate you and try to get you fired.

Surprised Stripe didn't explicitly make the list of notable startups from over the years. Bigger than even Dropbox by valuation (obviously not apples to apples vs. public market cap) and arguably has had a bigger positive impact on the world than any other YC startup by actively increasing the GDP of the internet.

It's probably because Stripe didn't go through the traditional YC program.

"What’s interesting is that unlike 99% of YC companies, Stripe didn’t actually go through the standard startup bootcamp process. Patrick had already been through it once and didn’t feel the need to do it again. The amount YC invested was in the $20k-$30k range." [1]

[1]: https://www.startupgrind.com/blog/the-collison-brothers-and-...

Interesting. I didn't realize that was the case. Makes sense and thanks for the additional context.

> YC startup by actively increasing the GDP of the internet

Sorry, but what does this actually mean? Before Stripe existed, you most certainly could process a payment online, so why would Stripe's existence increase the overall market for goods and services online?

Sorry for missing this thread. Stripe's mission is to "increase the GDP of the internet." It's the main headline on their About Us page. https://stripe.com/about

I take it to mean that by making setting up payments infrastructure easier for developers, more projects actually get started. Not to promote too much, but my brother's side project, NanaGram.co, would probably not exist without Stripe.

We actually used a pre-Stripe legacy payments gateway for our first startup (Rentabilities, now dead) and it was painful. I doubt he would have gone through that pain again to setup payments without Stripe. Yes, there are other providers that are good enough now like Braintree, etc, but I'd argue that Stripe forced the industry by raising the bar for developer UX.

I take it to mean that Stripe is easier to use than its competitors, which means new companies start that wouldn’t have, resulting in revenue through Internet sales that wouldn’t have happened online yet without those companies existing.

Obviously the premises can be debated, but developers do like Stripe’s API and products, and it seems plausible that some of those developers have been successful.

While Stripe is extremely developer-friendly, the alternatives aren’t that bad. They’re annoying to set up, but certainly not prohibitive to a determined founder starting out. Today, options like Braintree are comparable in terms of dev-friendliness.

Yes, but I’d credit Stripe for increasing the overall dev-friendliness of processors by introducing competition. And there were several companies started in the years before their competition caught up. I built two applications in 2012 using Stripe. One wouldn’t have existed otherwise, and the other would have cost my client a lot more money.

I’d also credit Atlas with new revenue. I don’t think that program has a peer yet.

Seems like kind of a weird thing to really take note of, but it wasn't just a general list, both parts had criteria that exclude Stripe ("household name" [i.e. consumer brand] and "funded in the last five years," respectively).

> [..] bigger positive impact on the world than any other YC startup by actively increasing the GDP of the internet

How is selling more stuff on the internet better for the planet? If anything online sales with all the additional packaging and logistics is worse than traditional retail (at least when looking at ecological factors).

I'll be hanging out in the comments this morning if anyone would like to discuss.

YC 2005-2014 = Doing things that don't scale

YC 2014+ = Making the magic of PG, Jessica, Sam et al scalable

That's the whole idea, right? "Do things that don't scale" means not letting future logistical concerns get in the way of discovering a viable market. But once you do find a good one, you scale the fuck out of it.

Why don't you start a European subsidiary? That would be a really good way to scale without compromising on quality. The need is great for something like that.

Interesting. Personalized rejections (one worded statements given the quantity of applications) would be great.

Something that might scale better is just having a rating out of 5 for each of the different areas e.g. team, progress to date, risk, potential reward, quality of idea etc.

At least it would allow them to focus on specific areas.

We're working on a way to do this.

Suggestion: when you folks send out that "You were in the top 10% of startups....please re-apply" email, would be great to receive it along with the rejection. Would be a much needed ray of hope at that point :-)

That's a good idea. Thanks!

As you have scaled other things, can't you scale feedback for companies? Atleast, for what you consider top 10% or so?

We applied last batch and got a question around a competitor that had just gotten funding. Not sure if that was the reason of the reject.

Unfortunately with over 8000 applicants each batch that is very very hard but I will talk to our admissions team to see if there is anything we can do to improve in this area. I apologize for how frustrating it is to get a YC rejection.

P.S I'll advise everyone to visit siebel's Molly page (https://molly.com/michael/q/eyJpZCI6IDEwMTR9/what-are-some-o... are answers to most of your questions on there. It's a great guide for your application process and you would be able to deduce why you are accepted/rejected from the analytics of those answers. With +-8000 applications you'll need all the help you can get to make your application stand out.

I think you should be commended for actually giving rejections. It's a very kind thing in a world of radio silence.

How about just adding a "feedback" box that reviewers could add feedback to if they wanted which could then be included with the rejection email?

This documents a lot of the mistakes startups make when applying to YC:


When people compare how many companies get into YC vs how many of them become successful companies should understand that beyond what YC brings to startups which is a lot, the success also depends on the founding team execution and the ability to find a market fit, sell and grow the product. I am a first-time applicant and what makes me take the decision was all the great resources YC can bring to my company (advisors, network, knowledge) more than looking at how many big companies or brands went through the accelerator program which is also a way of effectiveness measurement, but for me not the most important.

Michael how much of that early magic is lost now with PG not actively involved, with the bigger batches and the more industrialized machine?

Don't get me wrong, what YC has become is pretty amazing but boy what I wouldn't give to have been part of one of those first few batches or especially that first one with Aaron Swartz.

I talk about this specifically in my post... Having done YC in 2007, 2012, and helping to run it now. I think today YC founders are given many more tools and much more money to succeed.

Sure, no doubt today's founders are much better equipped for success no arguing that.

I guess I meant just from a pure experience perspective.

I think today's founders care about being equipped for success. The Y combinator in the lambda calculus is about what it enables you express, not itself.

I'm a "today's founder" (although I will admit I'm pretty old) but the history of PG, YC & the people that were in those first few batches I find incredibly interesting. In fact I find the whole evolution of this industry as a whole in Silicon Valley interesting & this is just the next evolution of that.

Sure, our focus is all about how to grow faster & I didn't mean to imply that today's YC is worse than it was when it started as it relates to achieving a founders actual goals, I didn't mean that at all.

My question to Michael was just based on this aspect of his post: "For me, there were two amazing things in the early days of YC that are different today. First, in 2007, batches were much smaller so it was easier to meet all of your batchmates. Today, you have to work harder to meet and become friends with all the talented people in your batch. We put companies into groups of about 30-40 in order to replicate the feeling of the past. Second, getting to work directly with PG was special."

From strictly an experience aspect since Michael is uniquely qualified to give a perspective on being a part of both then and now there had to be a certain magic in that 2007 batch that would be pretty tough to recreate today, or maybe not I guess, I don't know, it's why I asked.

Im looking to get in but I have been stalking the hacker page trying to skim every single information I can and believe me...I'm a walking hacker news...lol. Seibel is doing a great Job in encouraging soon to be/ start-ups to apply to Yc...he's shown an uncommon belief in startups regardless of what they do or where they from. That to me is a great change to the psyche of yc. If a lot more people are open as you are we'll have more start-ups. I'm not trying to schmooze you incase I say less about yc community...I haven't had that experience yet.

YC just announced YC China[1]. What will you say when they ask you to call it the "June Fourth Incident" instead of "Tienanmen Square Massacre"? You're not naive, you must see that moment in the future, or have some idea to dodge it somehow? Will you kowtow?

What's your attitude towards the Republic of China (Taiwan)?

[1] https://blog.ycombinator.com/yc-china-qi-lu/

A fair question and one that might be influenced by the Google MD's response to a request for a chat: https://www.nytimes.com/2018/09/28/technology/google-pichai-...

What's the average/median number of applications for startups who get in?

It's pretty awesome that you can actually apply 3 times within one year and did right that with https://fulfilli.com. It's good to take a look back and see how we've progressed.

What percentage of YC founders end up working for another YC company after their venture fails?

I don't know the percentage but it's definitely very common. Lots of YC companies look specifically to hire other former YC founders.


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