I’m always a bit depressed after the last dinner of the funding cycle. In years past, I’d plan a vacation starting the next day—just to distract myself from being sad. I wind up liking the founders so much that I feel like an empty nester once the weekly dinners end.

We have a bench signing tradition, where each founder signs the bottom of one of our benches at the last dinner.

But this summer was slightly different. The day before the last din, I was already preparing to be sad and was talking to Paul about it. I realized that this time the cause was not just because I’d miss everyone but something more: based on past experience, I knew some of the startups would die over the next several months.

I don’t mean I was bummed from an investment perspective. We’re in the seed funding business; we don’t expect all of our investments to pan out. It was the founders that I couldn’t help thinking about.

Most take funding from us and show up with grand ambitions. They beaver away for the next 3 months and get swept up in this amazing wave of excitement, achievement, and sense of invincibility.

But startups are hard. If they weren’t more people would do it. And it’s usually the next year that separates the wheat from the chaff.

That’s why I was so pleased that Paul decided to give a talk at the last dinner about how not to die. This topic may seem too grim to be introduced during such a celebratory night, but I think it was exactly the kind of thing the founders needed to hear. Because it could prevent them from making mistakes that inevitably lead to failure. For example, founders may not have control over market conditions, but they can decide whether or not they go to grad school. After 5 funding cycles, I’ve seen some predictors of early startup death and Paul hit them in his talk, which he posted online.

I suppose ultimately running out of money is the biggest reason for failure. But that’s a symptom of some other cause and doesn’t usually happen if you are doing things right. I think one of the biggest things that kills startups in the earliest phases is not focusing 100% on the company. Paul wrote, “If you find yourself saying a sentence that ends with ‘but we're going to keep working on the startup,’ you are in big trouble.”

You probably won’t make something people want right away. It will take so much perseverance on your part that if you put yourself in a situation where the startup is allowed to fail silently while you are (fill in the blank) [back at school] [consulting] or [working on another startup idea with a friend], then it probably will.

My favorite sentence was, “So if you want to get millions of dollars, put yourself in a position where failure will be public and humiliating.”

All this said, I don’t blame the founders if they try hard and fail. They of course have to pay their bills. Never having done it before, some people can find a startup too overwhelming.

Paul already warned everyone that “bad shit is coming;” and it probably is. But so is some really great stuff too. Like closing a funding round, launching, getting users, moving into your first office, creating something really useful, starting to be well known…

Though I’ll miss them, I can’t wait to see what the future holds for the summer ’07 founders. Good luck!

The gang gave us this digital picture frame, with photos of everyone to remind me of them. I'll treasure it!

Mitch Kapor was our guest speaker last week. It was a magical moment when he spoke to the room full of founders. Not just because he’s rich and famous, but because he is the kind of person they are. There seemed a special sort of fellowship of founders.  

One of the guys wrote to me the next day, “Wow, Mitch was truly great.  Definitely a nominee for Best Speaker Ever.”

I first met Mitch when I interviewed him for Founders at Work. I remember being on the edge of my seat when he told me about the early days of Lotus.

Chapter 6, Founders at Work: "I was so convinced that VisiCalc had a lock on the market that I had to convince myself that we were going to do something that wasn’t fundamentally a spreadsheet.  Of course, what we did was fundamentally a spreadsheet, but the self-deception I engaged in wasn’t sufficiently damaging to be fatal. But there was a big push to call it integrated software, to add other capabilities, to wrap other things in it.

The galvanizing event was when IBM announced the IBM PC in August 1981, and it was very important in the history of PCs because it legitimized the whole field--  because of IBM’s imprimatur. Until then, the personal computer hardware companies were Apple, Tandy and Commodore. IBM was the first “real” computer company to come out with a PC, legitimizing it for the business marketplace.  And that was not lost on me. And they made some very smart decisions." Read (a little) more.

Mitch’s interview is filled with stories that we cite to founders. But it’s not just his experience with Lotus that makes him so knowledgeable. He’s a cofounder of the EFF and was an early investor in companies like RealNetworks, UUNET, and Second Life’s Linden Labs. He talked about his experiences with these and other startups at the most recent Startup School. You can tell immediately that he is genuinely interested in new technology.  

What I like best about listening to him talk is that he’s so candid. I found myself writing down much of his advice. Dinner talks are off-the-record, but he’s given me permission to share one of my favorite quotes. Someone asked what it was like running Lotus and taking the company public. He replied:

“I actually woke up scared every single day that I would do something so stupid that the company would crash as fast as it went up.”

I like this because it shows that even a successful startup founders didn’t always think he had all the answers.

Since then, he’s seen a lot in the world of startups-- and continues to (he gave us a peek at some stuff he was working on with Foxmarks). But what struck me last week was that he does not seem to have lost any enthusiasm for new ideas.

What a week we all had out in California!

Tuesday Dinner

We hosted our weekly dinner at our office in Mountain View. We share space with Trevor’s company, Anybots, so the founders were introduced to his robots, Dexter and Monty.

Steve Anderson of Baseline Ventures and Sam Altman of Loopt were the guest speakers. Steve gave an investor’s advice on the fundraising process, and Sam offered the founder’s perspective.

The Bay area alumni descended, making dinner an impromptu YC reunion. I had forgotten what a force of nature all the past founders can be when they are in one room.

YC founders listen to the guest speakers

Ashwin (Buxfer) and Dan (Weebly)

Pelle and Sam

Paul and Steve (Reddit)

Demo Day

The big day was last Thursday and it came off without a hitch. Between the east and west coast Demo Days, the founders demoed to pretty much all the big Internet VCs, along with a lot of angels and even some acquirers.

In fact, the biggest problem we had was the crowd. Our California space is pretty big, but it was still uncomfortably tight. We’re going to have to try something different this winter. I think our plan will be to be to host a special session a few days in advance of Demo Day: a sneak preview for any investor who has previously invested in any YC company.

The investors seemed pleased: it was a very efficient way for them to see a lot of new startups. Many mentioned how impressed they were with the founders’ presentation skills (Paul warns the audience that we choose people because they are good hackers, not good presenters). One VC told me he was on a high from “having quaffed 19 shots at the entrepreneurial bar.”

We had some nice write-ups by the press.

Afterward, Trevor’s toys (the Segwell and Eunicycle) came out, and luckily no one was seriously injured.

Peter Nixey (and his iron-on shirt) on the "Eunicycle"

Pete Couldridge (blurred) almost takes down Paul Graham (with camera)

The proof of the pudding is in the eating

Though it feels like Demo Day was a success, the real test is how many of the startups can get more funding. I hate to scare the founders, but all the hard work they’ve done so far may seem like a walk in the park compared to actually getting investors to close.

Paul’s written some great essays about this. I think the hardest part of being a YC-funded company may be the second three months. We’ve seen a lot of our startups do well during this period, but I’m afraid we’ve also seen some wither and die. So we’re working with the founders now as they try to get more funding.

Xobni party

We capped off the week with the Xobni office-warming party on Friday night. The Xobnis’ new office is just right: funky, good light, not too fancy. A good place to take over the world from.

It was especially fun to see everyone at an event that I was not hosting. Though it was a party, it still kind of amazed me how cheery everyone seemed.

There are so many YC alumni now that they’ve become a real society. Paul told me later that there was nothing like this around when he, Robert and Trevor were starting Viaweb. No one really understood what they were doing. Their friends would ask, “So, are you still doing that company thing?” At the Xobni party, practically everyone there had either started or worked for a startup. They understood.

Ten weeks into each Y Combinator funding cycle, the founders present to investors at Demo Day. It’s probably the most exciting part of the 3 months we spend together.

10 weeks is not really a whole lot of time. But it’s enough to build something good enough to convince investors of your potential.

We tried something new this summer: in addition to presenting to investors in Boston, we’re flying all the founders out to Silicon Valley this week to present again at our office in Mountain View. We found that of all the previous startups that received follow-on funding, about 90% got it from west coast investors. Even ones that presented to Boston investors in summer ’05 and ’06 wound up moving to the Bay area for funding.

It’s gratifying how the east coast Demo Day has grown in popularity. The first summer we were praying people would show up. This year we had to put up a tent in our parking lot to accommodate everyone.

I think the founders were filled with a mix of excitement and anxiety as Demo Day approached. They had heard over and over what a big deal DDay was from alumni, but it wasn’t till a week before-- when they first rehearsed their presentations with the group and saw the list of attendees-- that it finally hit them. Y Combinator founders typically fall towards the Woz end of the continuum rather than the Jobs’ end; most aren’t that comfortable presenting to large audiences.

Well, the week they spent overhauling and practicing their presentations totally paid off. The founders really pulled it together under pressure!

But this often happens. What was surprising this summer was how many investors showed up, and that they all stayed to the end.

I think investors realize that it’s an efficient deal for them: they get to see 19 new startups in one afternoon. Based on past batches, it’s statistically almost certain there will be very successful companies among them.

And of course it’s also a good deal for the founders, who in two days get exposure to the top investors in the country.

10 weeks doesn’t seem like much time to make something to show to the nation’s top investors, but Y Combinator’s funding cycles are so concentrated that you get more done than you’d think possible. My three partners are all hackers, and one of their big principles is that hackers can be much more productive when they’re freed from distractions. Dinner each Tuesday is a big event, but the rest of the week we leave the founders alone to work, and during these stretches they get amazing amounts done.

So now 50 founders are all making their way across the country to the Bay Area, getting ready to fine-tune their presentations and try to convince another room full of investors that they are going to take over the world.

Paul Buchheit (r) with founder Josh Wilson

Paul Buchheit was YC’s guest speaker last week. He also spent the day meeting with the new batch of founders and giving them product advice. With 19 startups, this was a Herculean effort!

I first met Paul when I interviewed him for Founders at Work. He’s the author of Gmail, developed the first prototype of AdSense and suggested Google’s now-famous motto “Don’t be evil.” As much as anyone, he set off the “Web 2.0” phenomenon.

Needless to say, we were delighted to have him join us in Cambridge. Paul never fails to be thought-provoking when he gives talks. I’ve heard him speak a few times and  I usually come away with this overall message: listen to users, not convention.

This talk to the founders was no exception. I was really sick and had to leave before Paul’s talk, but I asked Paul Graham for a few highlights. My favorites were:

“Anything that’s different or innovative, people are going to predict isn’t going to work.”

It’s true. Anything that hasn’t been done before people tend to reject. In fact, practically everything really good was rejected at first-- from the Apple computer to Google. 

“You just have to find some group of people you can make really happy.”

It’s much easier to broaden the reach of something some people love than to get people to love something they only like.

Thanks again Paul for coming to YC and for all your support!

Loopt's Sam Altman was named to Inc. Magazine's list of "30 Under 30: America's Coolest Young Entrepreneurs."

Not vesting their stock is a common legal mistake new startup founders make.

Early on in our funding cycles, I help the founders set up all the initial legal and incorporation paperwork. (I’m not a lawyer, but I’ve gone through it about 60 times by now.) Y Combinator’s philosophy on legal paperwork is that it should be as simple and straightforward as possible. We hate it as much as founders do. When we first started out, we didn’t include vesting in the paperwork because we didn’t want to be overbearing investors.

Boy, am I singing a different tune more than 2 years and 5 funding cycles later. It’s like I’ve taken a “Scared Straight” course in corporate law. I’m exaggerating a bit, but the point is that there are so many disasters looming out there for startup founders, why not arrange to avoid one?

What is vesting? It’s when the founders voluntarily agree that instead of getting all their stock up front, they’ll earn it over time. There's no set rule for vesting schedules, but the most common are 25% per year (vesting monthly or quarterly) over 4 years, or 20% per year over 5. Usually there's a 1-year "cliff," meaning people who leave after less than a year get nothing.

Once you’ve issued your stock, you then need to file a form called an 83(b) election with the IRS. You have 30 days to do this or you can face hideous tax repercussions. Many paperwork oversights can be fixed later, but this one can’t.

Yawn. You must now realize why founders don’t want to deal with all this: it’s boring! Why attend to legal paperwork when you can be working on your software and planning to take over the world? Because if you don’t have vesting and one of the founders leaves with a large chunk of your startup’s stock, you will waste a lot of time and money to fix this situation. One YC-funded startup that didn’t have vesting had one of its founders leave within the first year. Here’s what one of the remaining founders told me about the aftermath:

“The biggest thing by far was the legal and accounting fees we had to incur (+$20k) and the time and stress associated with dealing with something like this.  Hours and hours on the phone, reading over documents that weren't explained well by lawyers, trying to understand from an accounting and tax perspective what we were getting ourselves into, etc.  If we had the vesting agreement, it would have ALL been unnecessary.

Everything is roses when the company is first starting-- you're excited, you're even thinking about the people you're working with differently, i.e. any misgivings you may have you're lessening in a ‘let's see how it goes’ attitude, or ‘things will be fine once we get started’ attitude, so obviously you don't worry about it as much.”

Several of our startups have had founders leave early on. For those groups without vesting, some founders gave back the stock voluntarily and some didn’t. But you don’t want to have to rely on your cofounder’s opinion about how much stock he/she deserves to keep. You want to be clear from the start about what someone will walk away with if they leave.

I couldn’t blame someone for leaving. It usually means things aren’t going well for the startup-- you can’t get users, you can’t get funding, your bank account is dwindling, you are demoralized-- and you need to pay your bills after all. But what about the founder who sticks around, making huge sacrifices to keep the startup alive a la Evan Williams? How would that founder like to be broke and working around-the-clock on the startup while the founder who left is sharing equally in the upside? Kind of embittered I think.

Also, think of your future investors. It will be a major red flag if you have to inform them that someone who owns 25% of the company now works elsewhere and doesn’t have anything to do with the startup. They don’t like it when someone owns a large portion of stock and isn’t “adding value.” They’d want (as you should too) that stock to be motivating a new employee who would be busting his hump on the startup.

Professional investors expect vesting and will likely impose it upon the founders anyway as part of their investment.

If you are lucky enough to afford lawyers to set up your corporation, they will obviously help you with all this. But many founders can’t afford their fees so early on. If you are using online services to incorporate (which I wouldn’t recommend), make sure you set up vesting. If you have already issued stock to the founders without vesting, look into setting it up retroactively. I think it’s pretty easy to do.

Most founders don’t think they are going to need vesting, but roughly 20% of the startups we’ve funded had a founder leave within the first year. So if you are thinking, “We don’t need vesting,” you are in the company of a lot of people who were wrong.

Facebook announced its first acquisition today.

Congratulations Blake and Joe!

Robert Morris with Loopt founder and CEO Sam Altman (summer '05) that summer.

Sprint will soon offer Loopt’s social mapping service on its phones. Loopt's mobile application uses the GPS capabilities on phones to let users keep in touch with their friends. A user's private network of friends is displayed as dots (yes, they move) on a map on the phone.

Loopt had till now been available only through Boost, so even though Loopt has been very successful, the average Web 2.0 fan may not have known about them. I think that’s about to change.

Congratulations!

Working with the founders is probably my favorite part of Y Combinator. So when “alumni” come to visit, I get especially excited. I’m spoiled when we’re in Silicon Valley, since so many YCers live in the Bay area. Many will crash dinners or else I’ll stop by the Y Scraper when I’m in the city to say hello.

Besides the 19 new startups participating in this summer’s funding cycle, I think there’s only one active alumni startup in Boston (more on this in a future post). So I was delighted when a few of our alumni visited Cambridge last week.   

Zenter: Victory lap

Wayne and Robby of Zenter flew out to meet the new founders and talk about what they wished they’d known in the first month of YC. This was the first time we’d seen them since the Google acquisition and it was nice to see for myself how happy they seemed. Their main advice for the founders: channel all your hacking ability into making something users want. Especially in the months leading up to Demo Day.

Wufoo: Getting it right

Kevin Hale of Wufoo also visited and spent hours giving the founders design feedback. (He’s extremely talented in this area.) Kevin founded Wufoo with brothers Chris and Ryan Campbell. They moved from Tampa to Mountain View in January ’06 and, like the Zenters a year later, took advantage of the next 3 months to work nonstop on their product. They rewarded themselves occasionally by watching LOST, but other than that, they were programming animals.

They raised some angel funding before returning to Tampa in April ’06. We worried the Wufoos would be hosing themselves with this move (do you know any software startups from Tampa?) but it seems to have worked out OK. In the 15 months since, the founders have accomplished a lot. They launched a year ago and have consistently added new features. They also care a lot about customer service. Apparently their average response time to emails from users is six and a half minutes-- and it’s only three guys!

They have all sorts of clever tricks to help them stay productive. I can’t say too much about the most amusing one, since I think Kevin plans to write about it on Particle Tree, but the payoff is a trip to the winner’s destination of choice with the loser assuming the role of “trip bitch.”

Kevin’s most exciting news was that Wufoo is now profitable. He showed me a graph with a nice smooth upward curve. Assuming they don’t make some huge mistake, the Wufoos are in a place most “Web 2.0” companies would envy: they’ll never have to ask investors for permission to keep working on their product.