Start-Up: the book

When I published Start-Up, a friend and colleague told me: “Why do you want to write something about high-tech entrepreneurship and start-ups. Nobody reads anymore. Make movies, videos!” He may have been right. Now that I heard about a new documentary about Silicon Valley and I will talk about this later in this post, it gave me the opportunity to look backwards. Triumph of the Nerds is a 3-episode (50 mns each) produced in 1996.

It is great, sometimes boring, often funny. Its author Robert X. Cringely is also the author of the related and very good book, Accidental Empires. You can watch the videos on YouTube and read the transcripts on PBS. I found Part I, the best. Part II about Microsoft and IBM is more serious, Part III about Apple is in-between.

First I found the best definition of a Nerd: “I think a nerd is a person who uses the telephone to talk to other people about telephones. And a computer nerd therefore is somebody who uses a computer in order to use a computer.”

Then about the semiconductor industry: “Intel not only invented the chip, they are responsible for the laid-back Silicon Valley working style. Everyone was on a first-name basis. There were no reserved parking places, no offices, only cubicles. It’s still true today. Here’s the chairman’s cubicle… Gordon Moore is one of the Intel founders worth $3 billion. With money like that, I’d have a door.” [...] “Only Intel didn’t appreciate the brilliance of their own product, seeing it as useful mainly for powering calculators or traffic lights. Intel had all the elements necessary to invent the PC business, but they just didn’t get it.”

“What was needed was a version of some big computer language like BASIC, only modified for the PC. But it didn’t yet exist because the experts all thought that nothing would fit inside the tiny memory. Yet again the experts were wrong.” And here came … Microsoft … and Apple

Steve Jobs: “Remember that the Sixties happened in the early Seventies, right, so you have to remember that and that’s sort of when I came of age. So I saw a lot of this and to me the spark of that was that there was something beyond sort of what you see every day. It’s the same thing that causes people to want to be poets instead of bankers. And I think that’s a wonderful thing. And I think that that same spirit can be put into products, and those products can be manufactured and given to people and they can sense that spirit.”

Part III talks about how Xerox missed the high-tech revolution and Apple or Adobe used Xerox inventions. The output? “A software nerd is the richest man in the world.” We are in 1996. Gates: “You know, if you take the way the Internet is changing month by month, if somebody can predict what’s going to happen three months from now, nine months from now even today eh my hat’s off to them, I think we’ve got a phenomena here that is moving so rapidly that nobody knows exactly where it will go.”

Yes, it was an Accidental Empire.

There is another documentary Pirates of Silicon Valley but it looks very similar to Triumph of the Nerds, without the humour or Cringely. But the reason of this post, is the recent released of Something Ventured. This I will watch soon and hopefully show at EPFL to students and colleagues.

Here is the trailer:

The long-awaited filing of Facebook was finally published yesterday. Amazing numbers, amazing success. You’ll find below the capitalization table and revenue numbers I (approximately) built form the S-1 document and you can compare it to the exercise I had done in 2010.

According to my analysis (I tried to take into account existing shares as well as options and restricted shares differently), Zuckerberg owns 20% of the company, the investors (preferred stock) about the same. IPO shares could be 5%. You can also have a look at the different rounds. And the difference is common shares (which may include investors) and employee options. Finally, I cannot comment on founders’ shares and you may have a look at the old table again.

click on table to enlarge Facebook 2012 cap. table

Revenues of $3.7B, a profit of $1B and 3′200 employees in 2011. A possible market value of $100B and an additional $5B in the bank. Google did not have such numbers. (Google had $1.4B in revenues, 2′500 employees and raised $1.2B at the IPO. It was only 6 years old though whereas Facebook is one year older. In 2005, Google had $5B in sales, $1.5B profit and 6k employees!) I had already compared both in a post in 2010: Google vs. Facebook and I have update the curves below.

click on table to enlarge

In the last 4 years, the yearly growth of Facebook has been over 80% for revenues and over 50% for the number of employees. I might be over-optimistic by saying that the average employee stock value is $4M (because of investor ownership of these shares too). The cap. table which follows shows numbers as guessed in 2010 and published in a post entitled The Social Network, when the movie was released.

click on table to enlarge Facebook 2012 cap. table

click on table to enlarge

I noticed at least 4 IPO filings this month, not bad. These are Audience, Infoblox, Millennial Media and most important to me as a European citizen, AVG Technologies. European filings in the USA are sufficiently rare to be noticed, and this time the company has Czech origins. After discussing AVG, I will show you an update of my start-ups data coming from these filings.

AVG did not experience the typical start-up process. Indeed the founders sold their shares to a private equity group in 2001, ten years after the incorporation. The investors then grew the company and attracted new investors including Intel Capital, TA Associates as well as a Polish fund. You may know about AVG, I am using it as a free anti-virus but I did not know it was a European start-up…

Click on picture to enlarge

The revenue growth is quite impressive as you can see in the cap. table (about $150M in 2010 from $100M in 2008). I found a 2000 presentation where the founder gave the facts and figures for 1996-2000. Then the revenues were respectively 17M and 55M Czech Korunas. One Krona was about €0.03, which means in the €0.5-1.5M. Not a bad growth at all. Why did the founders sell, I do not know and I am not even sure what they do today. They do not seem to be role models in Brno. Tomas Hofer seems to be active in another start-up however. If someone has more information on the founders, please comment or contact me.

Tomas Hofer

You can visualize the other cap. tables in my full data document. I do not have much to say about them, but I have updated my stats in the tables which follow, including new data on the amount of VC money raised. I also did a new classification in addition to geography and fields: years of incorporation.

Click on picture to access full pdf data

I end 2012 with two posts related to my beloved Silicon Valley. This one is about the two great Venture Capital firms Sequoia Capital and Kleiner, Perkins, Caufield and Byers. The next one will be about Palo Alto-based author of thrillers, Keith Raffel.

I have already said a lot about these two firms. You can for example read again the following on KP:
- About KP first fund (3 posts)
- Tom Perkins, a Silicon Valley venture capitalist
- Robert Swanson, 1947-1999
and about Sequoia:
- When Valentine was talking (2 posts)

The recent IPO of Jive is the motivation for this new post because Jive has both funds as co-investors. I am obviously providing my now-usual cap. table and what you can discover here is the huge amounts of money both funds have poured in the start-up ($57M for Sequoia and $40M for KP) … is this still venture capital? I am not sure.

Click on picture to enlarge

I am not writing an article on Jive here but let me add that we have again here two founders who had each 50% of the start-up at creation and end up with 8%, the investors have 30%. What is really unusual is that the company raised money in 2007, six years after inception. A sign of a new trend in high-tech?

Now back to Sequoia and KP. When they co-invested in Google in 1999, I thought it had been a very unusual event. David Vise in his Google Story (pages 66-68; I also have mine!) explains how the start-up founders desired to have both funds to “divide and conquer”, hoping no single fund would control them. When I met Pierre Lamond, then at Sequoia, in 2006, I was surprised to learn from him that in fact the two funds has regularly co-invested together. As often in Silicon Valley, it is about co-opetition, not just competition.

So I did my short analysis. A first Internet search got me the following:
- The question on Quora “How unusual is it for both Kleiner Perkins and Sequoia to co-invest in a company?” (August 2010) gives 11 recent investments, including Jive and Google.
- Russ Garland in the Wall Street Journal adresses the topic in “Kleiner Perkins, Sequoia Combo Has Solid Track Record” (July 2010). He says: “But the two Menlo Park, Calif.-based firms have done plenty of other deals together – at least 53, according to VentureWire records. It’s been a fruitful relationship: 29 of them have gone public. They include Cypress Semiconductor Corp., Electronic Arts Inc., Flextronics International and Symantec Corp. That track record lends credibility to the excitement generated by the Jive investment. But most of those 53 deals were done prior to 2000; the two firms have been less collaborative since then. Of the handful of companies that both Kleiner Perkins and Sequoia have backed since 2000, at least one is out of business. That would be Abeona Networks, a developer of technology for Internet-based services.”
- Now in my own Equity List, I have 4 (Tandem, Cypress, EA, google) plus Jive.

So I did a more systematic analysis and found 55 companies. More than the WSJ! I will not put the full list here, but let me give more data: Kleiner has invested a total of $267M whereas Sequoia put $268M [this is strangely similar!], i.e. about $5M per start-up. On average, KP invested in round #2.07 and Sequoia in round #2.63, so a little later. Time to exit from foundation is 6.5 years. I found 27 IPOs (I miss two compared to the WSJ)

Is Garland right when he claims “But most of those 53 deals were done prior to 2000; the two firms have been less collaborative since then”? Here is my analysis:

Number of co-investments related to the start-up foundation year

If I look at the decades, it gives,
70s: 6,
80s: 30,
90s: 11,
00s: 7.
Clearly KP and Sequoia co-invested a lot in the 80s, much less in the 90s and 00s. Whereas the fields are

So what? I am not sure . KP and Sequoia are clearly two impressive funds and as a conclusion, I’d like to thank Fredrik who pointed me to Business Week’s The Venture Capital Winners of 2011.

Sequoia and KP may not be #1 and #2, but their track record remains more than impressive. Here is a bad picture taken on an iPad!

I am surprised not to have published this before. It was one of my first work before I even wrote my book. It became its chapter 4. Venture capital is about 50 years old and it has changed a lot in parallel to innovation and high-tech. I hope you will enjoy these very visual slides!

I recently read Mind the Gap: The University Gap Funding Report published by innovosource.

Disclaimer: I usually do not mention my activities at EPFL on this blog and this report deals exactly with the type of funds I manage there: the Innogrants. I was indeed interviewed for this report as one of the active members of the Gap Funding community and the Innogrants are one of the examples mentioned.

Mind The Gap is a great report because it describes a concept which was born a little before 2005, the seed funding, I should even say the pre-seed funding, by universities of their innovations, including start-ups. The next figure illustrates not only gap funding but all the existing tools enabling academic innovation.

Let me just briefly quote it (but you should know the report is not free, so I cannot summarize it in too much detail. The author allowed me however to give you a 25% discount code: USHAPE). In any case, it is extremely rich in data and information.

“This “gap” extends from where the government funding of basic research ends to where existing companies or investors are willing to accept the risk to commercialize the technology.” [page 9]. The author reminds us that “Failure is commonplace in these sorts of pursuits, but ask where you would find yourself (or where you are going) without GPS and the internet, and most recently a little iPhone “assistant” named Siri that originated from the DARPA funded CALO (Cognitive Agent that Learns and Organizes) program through a university consortium.” [page 20]

As a side element, there are also the emerging accelerators, “Popularized in recent years with the likes of Y Combinator and TechStars, accelerators combine access to talent and support services with “stage-appropriate” capital in return for a stake in the company or other repayment structures.” [page 22] but this is another subject!

There are already some “famous” gap funding tools: “another study by the Kauffman Foundation [1] investigated two well-known proof of concept centers at MIT (Desphande Center) and UC-San Diego (von Lebig Center) and reported general process and impacts.” [page 26]

[1] C. and Audretsch, D. Gulbranson, “Proof of Concept Centers: Accelerating the Commercialization of Univeristy Innovation,” Kauffman Foundation, 2008.

I do not want to quote much more this 100-page deep and very interesting analysis. My final comment is that a critical element is the leverage gap funding enables. You will find a full analysis on pages 88-90. In his Report Summary, the author depicts the value of gap funding through:

High commercialization rates
- 76-81% of funded projects commercialized on average
Attraction of early stage capital
- $2.8B leveraged from public and private investment sources
Business formation and job creation
- 395 new start-up companies
- 188 technology licenses to existing companies
- 7,761 new jobs, at cost of $13,600 gap fund dollars per job
Building a community of innovation
- Thousands of faculty and students engaged in the process
- Incorporate networks of technical and business professionals in the evaluation, mentorships, and leadership of these technologies
Organizational returns
- $75M returned to the organizations through repayments, royalties, and equity sales
- Maximize resource allocation and downstream savings, by permitting early failures through exploratory and evaluation tactics
- Empower universities to continue to take risks that support the type of breakthroughs that define our present, and the type of innovation that will carry us into the future

Let me finish with what I contributed to the report, i.e. a short description the EPFL innogrants:

When I met Jochen Mundinger in October 2006 it did not take me much time to make up my mind. I had previously seen many startup ideas and Jochenʼs Internet project looked to me original and powerful. Prior to any due diligence, I told him that if my analysis was positive, he would get a 12-month grant to work on his start-up. Because of this program, I am lucky enough to be able to make fast decisions and by January 2007, Jochen was working on his project. He did not wait until the end of his grant to found routeRank and by October 2007, with the support of business angels. Today, the service has grown and been recognized by the famous MIT TR35 award in 2010.

And then there was Andre Mercanzini, a Canadian citizen, who certainly has the drive and enthusiasm of many North-Americans. Andre obtained his PhD at EPFL following a few start-up experiences in the US. Andre has developed electrodes for Deep Brain Stimulation. The path was not as fast and easy as for Jochen. Though Aleva Neurotherapeutics was founded in mid-2008, Andreʼs prototypes needed further validation to attract venture capital (a major use of the grants). The Swiss ecosystem is rich with mentors and support so that Andre developed further his project to the point of raising $10M in his series A round in August 2011.

These are just two examples of EPFL innogrants. Initially backed by Swiss bank Lombard Odier, it has since received support from KPMG and Helbling, an engineering firm. The fact that similar initiatives were launched in Switzerland is another illustration that gap funding attracts and seduces. The Innogrants are a bet on young people. Since 2005, 48 projects have been funded out of more than 300 ideas and 24 companies created. We admit at EPFL that failure is part of the process and even if no start-up is ever launched, the grant is a learning experience. We also have the vision that Innogrants become role models and hope that more and more students will be less shy about expressing their dreams.

Venture Capitalists are always proud to mention which companies they successfully backed. It is because of their success stories that Sequoia and Kleiner Perkins are so famous in this industry. But the deals the VCs decline are much less famous. In my book, I had mentioned some examples by some pioneers of venture capital:

A few VCs use the humour to tell their biggest mistakes. A colleague of mine (thanks Amin ) recently mentioned to me that Bessemer has a full list on their anti-portfolio: A123, Apollo, Apple, Check Point, eBay, Federal Express, Google, Ikanos, Intel, Intuit, Lotus and Compaq,
PayPal, Stratacom.

The most striking miss is probably Google: “[One of Bessemer's partner] Cowan’s college friend rented her garage to Sergey and Larry for their first year. In 1999 and 2000 she tried to introduce Cowan to “these two really smart Stanford students writing a search engine”. Students? A new search engine? In the most important moment ever for Bessemer’s anti-portfolio, Cowan asked her, “How can I get out of this house without going anywhere near your garage?”

But then what about OVP’s ironic style in their Missed Deals including

Starbucks.

“A guy walks into your office in the late 1980’s and says he wants to open a chain of retail shops selling a commodity product you can get anywhere for 25 cents, but he will charge 2 dollars. Of course, you listen politely, and then fall off your chair laughing when he leaves. Howard Shultz didn’t see this as humorous. And we didn’t make 500 times our money.

To get even (wasn’t our not making money enough?) years later, Howard opened his own venture capital firm right down the street. “

Amazon.

“The Internet boom was just beginning. Amazon had sales of $4M a year. We had a handshake on a term sheet with the CEO to put $2M into Amazon for 20% of the company (a $10M post money value). At the eleventh hour, some guy named John Doerr flew up and offered $8M going in for 20% of the company (a $40M post money value). Handshake? What handshake?

To get even, we buy all our books at Barnes & Noble. We don’t think Amazon has noticed.”

Just a few lessons about the difficulty in reading the future. If you have other links, please comment.

Do not ask me why this book is entitled The Monk and the Riddle as I will let you discover it if you decide to read this “old” book (a more than 10 year-old great piece of Silicon Valley description). Its subtitle is clear though: The Education of a Silicon Valley Entrepreneur.

Not all agree on the fact it is a great book as you may find at the end of this post, from the comment by the Red Herring in 2000. Still, I loved reading this book and let me explain why. Randy Komisar, today a partner at Kleiner Perkins and former enrtepreneur, has written a book about passion and inspiration. He does not tell you how to do your start-up (but he tells you how not to do it). He also explains also very well what Silicon Valley is, the locus of risk taking, where failure is tolerated, where a start-up is more a romantic act than a financial endeavour. “Business isn’t primarily a financial institution. It’s a creative institution. Like painting and sculpting.” [page 55] Here are a more few extracts I scanned from Google Books.

First Mr. Komisar explains that an entrepreneur is a flexible visionary and why the business plan does not have to be strictly followed (or should not always be) [page 37]:

Of course, venture capitalists look for such people [page 38]:

But there is a danger with VCs: the down round which is the consequence of failed momentum [page 52]:

Mr. Komisar gives much more than basic advice. Even if he admits he may not have followed these when he was younger, he understands now how important they are. His book his about the meaning of life where he defines the Deferred Life Plan (that should not be followed) [page 65]:

He therefore considers that personal risks are more important than business risks [page 154]:
Personal risks include:
- the risk of working with people you don’t respect,
- the risk of working for a company whose values are inconsistent with your own;
- the risk of compromising what’s important;
- the risk of doing something you don’t care about; and
- the risk of doing something that fails to express – or even contradicts –who you are.
And there is the most dangerous risk of all – the risk of spending your life not doing what you want on the bet you can buy yourself the freedom to do it later.
[… page 156...]
If your life were to end suddenly and unexpectedly tomorrow, would you be able to say you’ve been doing what you truly care about today?

He also explains why Hard Work is a critical and necessary value of Silicon Valley [page 125]:

But People and Culture remain the most important elements [page 128]:

Another interesting concept is the fact that start-ups need 3 CEOS [page 128]:

But nothing replaces Vision [page 144]:

When I wrote above that Silicon Valley is about tolerance to failure [page 150]:

Obviously it means even success should be mitigated [page 151]:

I really advise you to read this great book, not only for the Riddle but also for the nice, funny and sad story of Lenny and Allison. Enjoy!

Here is what the Red Herring published. The analysis is not wrong, but even 10 years later, I am not sure Silicon Valley is so well understood as the RH thought…

A new report on venture capital brings interesting conclusions and updates. Here is the summary that you can also fidn on the Nesta web site:

1. The returns performance of UK and US VC funds in recent years has been very similar. UK funds have historically underperformed US funds, but this gap has significantly narrowed. The gap in fund returns (net IRR) between the average US and UK fund has fallen from over 20 percentage points before the dotcom bubble (funds raised in 1990-1997) to one percentage point afterwards (funds raised in 1998-2005). However, this convergence has been driven by declining returns in the US after the burst of the dotcom bubble, rather than by increasing returns in the UK. Average returns for funds raised after the bubble in both the UK and the US have been relatively poor, but VC performance is likely to move upwards as VC funds start to cash out their investments in social networks (particularly in the US).

2. The wider environment in which UK funds and the companies they finance operate was a major contributor to the historical gap in VC returns. While there are some large differences in the observable characteristics of VC funds between both countries, they cannot account for the historical returns gap.

3. Average returns obscure the large variability in returns within countries. The dispersion in returns across funds was highest during the pre-bubble years, and has fallen significantly since then. But in both periods the gap in returns between good and bad performing funds within a country was much larger than the gap in the average returns across countries. Thirteen per cent of UK funds established since 1990, would have got into the top quartile of US funds by returns (this has increased to 22 per cent for funds established in the post bubble period), while 45 per cent of UK funds outperformed the median US fund. Selecting the right fund manager is thus more important than choosing a particular country.

4. The strongest quantifiable predictors of VC returns performance are

(a) whether the fund managers’ prior funds outperform the market benchmark;

(b) whether the fund invests in early rounds;

(c) whether the fund managers have prior experience; and

(d) whether the fund is optimally sized (neither too big nor too small).

Moreover, historical performance has been higher for funds located in one of the four largest investor hubs (Silicon Valley, New York, Massachusetts and London) and for investments in information and communication technology.

5. UK government-backed funds have historically underperformed their private counterparts, but the gap between public and private returns has narrowed in recent periods. This suggests that in later years governments have become savvier when designing new VC schemes.

Most US funds have traditionally only invested locally, with less than a third of US funds raised between 1990 and 2005 having invested in one or more companies outside the US. In contrast, the majority of European funds have invested outside of their home market.

The situation has changed somewhat in recent times. A higher proportion of European funds raised in 2006-2009 have chosen to invest locally while US-based funds are becoming more global. As a result, the proportion of European VC capital being invested in the US has halved, falling to 10 per cent, and a slightly larger share of US VC capital is coming to Europe.

Overall, this analysis suggests that Europe does not offer an attractive proposition to US VC funds. Europe has a less developed VC market than the US, so attracting US funds (their money but also, crucially, their expertise) ought to benefit European economies. Instead, the opposite is happening. A much larger share of European VC funds invest in the US than the other way around. While Europe is likely to benefit from its funds investing in the US (for the returns it provides, the network it builds and the experience it generates), the small flow in the opposite direction is a cause for concern.

In conclusion

- The global venture capital industry is concentrated in very few hubs (and does not exist in a vacuum)

- The convergence in returns is not the result of changes in the characteristics of UK funds

- Small funds underperform medium sized funds, but larger is not always better

- More experienced fund managers achieved higher returns

- Past performance predicts future performance

- Funds in investor hubs had better returns

Investing in earlier rounds leads to better performance

- But much of the variability in returns is not explained by these factors

Finally some advice on Policy:

Remember venture capital activity does not exist in a vacuum.

Resist the temptation to overengineer public support schemes

Avoid initiatives that are too small.

I also found interesting two figures:

In my current reading of old Red Herring and Upside magazines (see for example Google in 2000 and Funny Data in the Internet Bubble), I just discovered an interesting ar4ticle about how angels may replace venture capitalists (Upside 1999).

They take a one example sendmail which did not go with Kleiner Perkins, Accel or IVP but closed a $6M round with business angels at a $20M valuation. The article also mentions the Band of Angels, agroup of then 120 investors having invested a total of $44M with an average investment of $600k and the Angel’s Forum with 20 investors putting up to $500k per start-up.

Sendmail raised $35M in 2000 (series D), $14M in 2002, as well as debt financing (at least $7M) as recently as 2009. Sendmail is still private so difficult to say if it is/will be a success or not.

I had doubts in a recent post on Super Angels being new stuff, this shows it is clearly not that new…