Nanobeepers

Yesterday I drove over to Stanford to attend Paul Graham's talk on Handling Investors. As I pulled up to the Stanford Oval, I saw Paul and Trevor Blackwell walking to the event. Actually Trevor was riding a motorized unicycle that he had put together and which Paul described it as a one-wheeled segway. As we walked I asked how the application reviews went. They said that it was brutal and that there were about twice as many apps this time: 425.

When I did my app, I really liked the last question which was to tell about something 'surprising' or 'amusing' that you had discovered. So I asked if they had had many good answers to that one. Paul said that there were some really good ones, but that inexplicably many had left that question blank! Then he chuckled and said "Actually, there was a guy who submitted this wonderful algorithm for wiping his ass." I couldn't believe he remembered that - but I guess that would be hard not to remember and I told him "That was mine!" He answered "Really? That was really yours? Small world!" That was pretty funny. But, the saying is true, "A clean bum, is a happy bum" ;)

Anyway, we got to the auditorium and then I discovered that my battery was dying on my audio recorder. Total bummer - so I didn't get the podcast to share. Sorry guys :( But I did take pretty good notes and I'm sure Paul will be putting it out as a full essay at some point :) The following is taken from my notes and memory. These aren't direct quotes from his talk, but it's pretty close. Enjoy!

___ Update2 ___
PG has just released the official essay based on this talk at: http://paulgraham.com/guidetoinvestors.html
___ Update2 ___

___ Update ___
Comment from Paul on these notes: "This is fairly accurate as on-the-fly notes from a talk go, but certainly nothing like a transcription. There's a lot missing and a few things that have been misunderstood." Yep, that's why I prefer to podcast! Be sure to read Paul's real essay when it comes out :)
___ Update ___

There are two important principles at YCombinator:
The first most important thing is to build a great product that people want. The second is how to handle investors and is the focus of this talk. I've been dealing with investors for twelve years and have been on both sides of the investment table. It's a weird world and there are many surprising things about it. It's much different from the programming world.

Here's a list of 21 surprising things I've learned along the way.

1. Investors make silicon valley
I wrote an essay on what makes silicon valley what it is. Basically it comes down to two key elements: nerds and rich people. And it's more important to have the rich people, simply because the rich are less willing to move! If there is a group of great programmers in Cincinnati, the rich people are not going to move to be near them. It's always easier for the poor hackers to move to where the rich established people are.

To reproduce silicon valley, you need angels. Many are former founders. Shockley Semiconductors was the real seed of silicon valley. Urban legend tells the story of HP being the start of silicon valley, but this is only because nobody liked Shockley and they didn't want to give him credit. Shockley was a mad man, but he hired really smart people, who then left because of him and started their own successful companies that made the valley what it is today.

2. Of investors - angels are the most critical
Angels are very different from VC's. VC's are investing the aggregate funds of other people's money and they get paid whether the fund succeeds or not. Angels on the other hand are writing checks straight from their own bank account and thus much more 'invested' in the success of the startup. When there is investment, it is always the angels first since they will invest in riskier things. The way the media tells it, you would think that only VC's funded Google. However, Google was funded by angels and the VC's didn't come in till late in the game with Google at a 75 million valuation!

3. Angels don't like publicity
On the other hand VC's do. VC's have customers and have to market themselves to build their funds and attract talented founders.

Angels are private investors that do not have a staff and therefore do not want to be deluged with outside requests for their money.

Also, no deals are done by email. They're all done by introduction. Investors mostly see it as a sign of incompetence or laziness if you can't manage to get a real introduction. So email is basically a waste of time when it comes to seriously approaching investors.

4. Investors are not like founders
This is especially true for VC's. To hackers: What if someone said that you had to stay in an office all day making nothing and just looking at other people's proposals? You'd say their nuts! That's exactly what VC's do and why they are so different from good hackers. It's the same difference evident between novel writers and editors, artists and art critics. One set are the makers and the other set evaluates what the makers are making. It's true that you'll find some VC's with a technical background, and those VC's are the best, but they are still fundamentally different from hackers.

5. Most investors are momentum investors
VC's in general are pretty clueless, yet they make a lot of money. How do they do it? Well, it's possible to make lots of money even if you don't know anything. For example, in 'program trading' there is a simple algorithm which is basically:

when you see the stock start to go down quickly, then sell
when you see the stock start to go up quickly, then buy

That's how VC's operate - they basically just look at Alexa and buy what's hot.

This is what VC's care about:
A) traffic
B) what other VC's think of you
C) the quality of your team

It can be a perpetuating cycle when all the VC's are constantly looking at what all the other VC's are doing. This can be to your advantage when looking for funding. Once you have a small amount of funding from an investor, you can go to the next investor with the fact that so-and-so is investing in you, and you're much more likely to get funding, then you can go to the next investor with the fact that you have these two investors and so on.

6. VC's look for big hits
You need to seem like your going to go public, and in order to go public, you need to have huge revenues. This is why companies that are essentially just a 'feature' have trouble getting funding. You can't build a huge hit off a feature typically.

You need to demonstrate that you're ambitious to the point of insanity. When you walk in, it needs to be clear that you are going to build a huge company.

This is one reason investors love Max Levchin. After PayPal, he certainly didn't need the money, but started Slide because of his naturally high ambition.

7. Investors look for stars
In the past VC's made the mistake of looking for the next Bill Gates. But Microsoft was a bad example to use. They essentially got a monopoly dropped on them by IBM. It was like IBM was walking by real slow with their wallet full of money half-way hanging out of their back pocket!

During the bubble one of the big mistakes investors made was investing in MBA's as founders who would then hire hackers to develop their idea. That's like investing in Steve Ballmer and hoping he hires Bill Gates!

Fortunately, most have learned their lesson and now look to invest in the next Larry and Sergey.

8. The contributions of investors is underestimated
One of the main reasons for this is that neither the investors nor the founders have an incentive to share this information. The investors don't want credit for the success of a company, because then they would be wrapped up in the value of the company and that could make it harder to sell since they don't come with the startup when it's acquired. So it's in their best interest to promote the company as successful on its own merits. The founders likewise are also motivated to market the company as a self-contained success story for the same acquisition-related reason.

Since all the credit usually goes to the founders, other potential founders often think they have to be god-like in everything they do. This is a dangerous notion, because it's not true - there are many huge contributions from others. For example, does anyone know who did Gmail, Adsense, and the 'Don't Be Evil' slogan? Most people have never heard of him, but it was Paul Buchheit, not Larry or Sergey.

One of the great things about Startup School is that potential founders get to come and hear successful founders speak and then realize that some of those guys really aren't all that smart.

9. VC's want to invest large amounts
This is because they have a large amount of money in their fund, but only a few partners to distribute it. So to avoid spreading themselves too thin, it's in their best interest to invest big in a smaller number of startups.

Some of the lame VC's are in it just for their percentage on the fund and will raise a large fund, invest big in a small portfolio of companies, and watch it fizzle.

Angels are different in that they want to invest the least amount possible since it's coming right from their pockets.

10. Valuations are made up
They are strictly fiction at the beginning. The VC wants to invest $X and they want 1/3 of the company, so they'll just use those figures to get the valuation!

VC's want to invest more, but this decreases the liquidity possibility. If you take a high valuation, then you have to sell for a SUPER high valuation.

It's tempting though for founders to seek a high valuation for ego/competitive reasons.

11. VC's are afraid of limited partners
This is surprising. You would think that VC's would just tell their fund investors that if they don't like what they're doing, they can take their money and leave. But VC's are very concerned about how they look to their limited partners and this affects the risks they are willing to take. Even if a deal looks super promising, if it has the potential to make them look dumb, then they'll pass on it. They have to wait until that startup is doing well, or other VC's are investing in similar startups before they will invest themselves.

12. There are 2 kinds of judgments: Admissions and Grades
If you get a bad grade in a class, but think you did a great job, then you feel entitled to go and talk to the teacher and challenge the decision. On the other hand, that entitlement feeling isn't present if you apply to a college and get denied. This is because in admissions there are a limited number of seats available and a cutoff point at which no more people can be accepted. Like in the olympic tryouts - the top X% of swimmers for a country may get to go to the games, but those below that threshold don't get to go. It's not much different in getting funding - you can just slide off the edge and just not make the 'admission'. When I was rejected by Harvard I thought I sucked. Now I see it as laughable. Have you ever been to the admissions office? A rejection in an admissions scenario is not necessarily a reflection on you. At YCombinator we go through and rank all of the applications, then we take the top 30 for interviews. Recently we looked at some of our early teams and compared how they are doing now with how we rated them in the application process. It was scary! One of our most successful teams was at number 30! Just one more down and they would have missed the cutoff.

13. VC's are like High School girls
A) fickle
B) your stock is based on what the other girls think
One good strategy would be to bribe 10% of the popular girls to like you, then the other girls would follow!
Remember that if you are not sufficiently deferential and responsive to them they will drop you. One of the top angels told me that if someone doesn't return his email within 24 hours then he writes them off.

14. Investment process is never 'done' till the money is in the bank
You can't stop selling yourself till the deal is completely closed and the funds have been transfered. You need to make sure that you seem so hot that at any moment another investor could swoop in and take their place.

15. Investors like to co-invest
A) makes them feel better
B) makes it less likely for a competitor company to get funding if multiple VC's are committed to this one

16. Investors collude
Investors are not covered by anti-trust law. Sometimes they will get together to hose a startup. Like if the army and air force of a repressive regime play a game of baseball together, they are competing against each other, but in the end they are still united in ruling the country's population.

This can also be a positive for founders since VC's will often pass a deal on to another VC if the company is too similar to another company already in their portfolio.

17. VC's will sometimes hose one company to benefit another
VC's win if their *whole* portfolio wins. So there have been times when a VC will sell one of their portfolio companies at a slashed down price to another one of their companies. The VC and the acquiring company win. The sold company loses.

18. Investors have a different risk attitude than founders
Investors are rich enough to be rational. Founders are typically poor and thus more risk averse. So when a deal for a $30 million acquisition comes along, the founders may be much more motivated to take the deal rather than keep going to build the company. The investor may be more motivated to have the founders keep going since his risk is much less in percentage terms.

19. VC's vary greatly
They all look alike if you go and look at their website profiles: blue button shirts, jacket, slacks, etc. But the best VC's are the ones that have *good* product ideas and have *useful* advice. There is a very sharp dropoff in VC quality.

20. You need investors
There is much empirical evidence that this is true. Very very few companies make it without the help of investors.
The wrong question to ask is "Do I need investment?"
The right question to ask is "Will taking investment help me?"
Remember that your competitors may have investment and thus a competitive advantage over you if you don't.

21. Investors like it when you don't need them
Some founders approach investors as if they are asking for permission to start a company. That is very much the wrong approach. You need to make it clear that the train is leaving and if they want to be on it, they better get on. That being said, you have to mean it 100%, because if you don't, they'll be able to tell. A good way to mean it 100% is to have a solid backup plan. What also helps is to start cheap and avoid doing the expensive things till later so you don't *need* the investment, but can get real momentum going for your company.

You're startup needs to be like a cockroach - hard to kill, even after a nuclear war. Don't be a delicate beautiful flower, be a cockroach.

Imagine the VC on his hands and knees with a cockroach on his back riding him with a harness. That's the approach you want to take.

_____________________

Update: View the comments for this post at: http://news.ycombinator.com/comments?id=12282

I've been developing a framework using EC2, S3, S3DFS (DFS = Distributed File System) along with SQLite to try and solve the database problem on Amazon Web Services.

S3DFS allows you to mount an S3 bucket to one or multiple EC2 instances as if it were a local filesystem. It works at the block level and has read/write caching - so is very fast. (Note: It's free to use for development, but otherwise requires a paid commercial license.)

Since SQLite uses flat files and has no database server or client, it is ideal for use on S3. SQLite only scales well to a point, so to get around its limitations I am separating each user into their own database (each database is a separate flat file). This can sometimes be a bad idea, but SQLite allows you to attach multiple databases together to essentially create a temporary master database where you can run queries across all the user databases at once. This will be handy for search indexing, site-wide stats, etc. In order to change the user databases I will have utilities to allow global scheme alterations.

So, by adding EC2 instances that are all attached to the same S3 bucket, and using round-robin dns to distribute traffic among them, you can theoretically scale a web service infinitely. Just launch a new EC2 instance and add it to the round-robin dns pool and you're done! Even that could be automated to respond dynamically to traffic.

Imagine the long-term cost-savings from only having to build your app and architecture once and not having to go through a more complex scaling process.

Certainly you should only worry about scaling after your app is up and running, but what if you didn't have to ever worry about it? That's the goal of this framework.

This is pre-alpha of course and I'm still in the development and experimentation stage, but I wanted to share with the community and get some feedback!

Update: To follow the feedback on this topic, see: http://news.ycombinator.com/comments?id=10001

Update2: I have decided to change my strategy a bit to bypass some of the limitations inherent in SQLite and S3DFS. Rather than have the user database files sit directly on S3 via S3DFS, I will have them reside on the EC2 instance and just back them up frequently to S3 via S3DFS. This will be faster, less complex, and more stable. When I get to the scaling part of the app, I'll probably use a reverse-proxy solution and divide the user database files across multiple EC2 instances. Thanks to the guys at Lunchgeeks for their help on this!

VC Rick Segal's approach seems the most charitable.

His tact works similarly in dating.
(Note that I date with the goal of marriage in mind - I'm definitely not a 'player', which makes this a good analogy to the good VC's - they're hoping to 'marry' founders when it's a good match)

Many guys burn bridges when they have to break up with a girl. However, that's totally unnecessary! I've certainly done that when I was a teen, but have found the charitable approach much more effective. Let me be clear that 'charitable' does not mean unclear or deceptive in the breakup. On the contrary - it's being very clear and honest about the breakup with the girl. That being said, you *do* withhold any information that would unnecessarily hurt her (ex: you're not attracted to something about her that she can't help).

The breakup needs to establish that:
1) the relationship is not a match for you, but that that is no reflection on her personally in the slightest
2) you genuinely admire the N number of amazing things about her
3) you hope that you both can still be good friends, but that you understand and honor her wish if she thinks it's better not to
4) the romantic possibilities of your relationship are just not there sufficient for marriage and never will be (let her move on with closure!) Don't worry, if you change your mind later you can crawl back on your knees to her and tell her you were wrong - but you have to let her go and pursue other possibilities.
5) you are open to talk with her about it at any time - this will usually relieve a lot of her anxiety about having to wonder why you broke up later.

Note: Be careful about the 'still being friends' thing. If you are hanging out too much with her, most other guys will avoid asking her out - and you don't want to hinder her ability to move on. Be sure to honor her space and let her know that you are doing that so she doesn't feel rejected by it.

Success stories:

2004 girlfriend: Good breakup, she decided it was better for her to not have contact. I honored that and imagine she has gone on to great things.

2005 girlfriend: Good breakup, though we did get back together several times. Ultimately it ended, but our communication was so good that we have remained best friends since then and talk regularly about our new relationships. We both are huge advocates of each other - she sells me to other girls and I sell her to other guys. We do business together too since there is such a high level of trust.

2006 girlfriend: Had a great time together, but we both knew it wouldn't work. We had great communication and though it was hard for me to let her go, I did and that really impressed her to the point where we have a similar high-trust, high-advocate relationship as with 2005 girlfriend.

2007 girlfriend: Ended pretty recently. So far so good. Very hard for both of us. But we've had good communication since and it is evolving well into a good friend relationship.

Note that these tips assume that you are only dating reasonably sane girls - with psycho chics this probably won't apply ;)

It's important to remember that going the charitable route is often harder in the short-term than the alternatives, but it's important to suck it up and have the kahoonas to do it right. Some guys may be tempted to not give her a clean break or feel they have to diminish her as a person to satisfy their own insecurities. DON'T DO IT! Seriously, it will hammer your karma and make life much more difficult for you in the long-term. To illustrate: 2006 girlfriend only dated me because of the good things 2005 girlfriend told her about how good our relationship and breakup was. The same with 2007 girlfriend. It's really the way to go - be respectful and honor them, because more often than not, they will return the sentiment.

Granted that my experience is pretty limited, please share your own success stories in the comments!

Update: When thinking more about this, I find that it applies also very well to leaving a job with the best possible feelings.

http://feeds.feedburner.com/Ycombinator-StartupSchool

This podcast contains all the published podcasts from the last 3 years of Startup School. Talks from 2005 and 2006 are linked to the official recordings and those from 2007 are linked to my lower quality recordings until the official ones are released.

If you were not there this year and are short on time - I'd recommend listening to Paul Buchheit, Paul Graham, and Greg McAdoo first.