Posts from February 2010

I went down to Philly yesterday and spent the afternoon with students and faculty at Wharton Entrepreneurial Programs, the entity that administers the entrepreneurship major/courses at Wharton and runs a bunch of fantastic "outreach" programs like the Venture Initiation Program. 

WEP is run by Professor Raffi Amit and as we were making our way from one meeting to another, I said to Raffi that "you can't teach people to be entrepreneurs but you can teach entrepreneurs business." He replied to me that his research into the topic suggests that "there are no unique and defining characteristics of entrepreneurs" which leads him to believe that you can in fact teach people to be entrepreneurs.

That threw me and I've been ruminating on his conclusion ever since. I've been working with entrepreneurs for almost 25 years now and it is ingrained in my mind that someone is either born an entrepreneur or is not.

And I also believe that there are "unique and defining characteristics of entrepreneurs." Here are some of the ones I observe most frequently:

1) A stubborn belief in one's self

2) A confidence bordering on arrogance

3) A desire to accept risk and ambiguity, and the ability to live with them

4) An ability to construct a vision and sell it to many others

5) A magnet for talent

I accept that Raffi may be right. I will get my hands on his research and read it. And maybe it will change my mind on this topic. 

Venture Capital is a lot about pattern recognition. You learn to make quick judgements based on things you've observed in the past. And judgements about people are among the most important decisions we make. So this is a very important topic to me.

I access the web each day from at least seven browsers; 

Chrome on my macbook pro, 

Firefox on my macbook pro, 

Chrome on my windows desktop in the office, 

Firefox on our "kitchen laptop", 

Safari on our "kitchen laptop", 

The Android browser on my google phone, 

The blackberry browser on my Blackberry 9700

I know that I am not your typical web user. I keep both Firefox and Chrome open at the same time on my main machine (my macbook pro). I use three machines most every day, my macbook, my office desktop, and our "kitchen computer". And I use two mobile devices every day.

But I am illustrative of something important. Many people access the web from multiple devices and browsers on any given day. 

Each of those browsers I use every day drops a cookie identifying me as a "unique visitor" and the web analytics software the website is using counts me as up to seven unique visitors when I am only one.

I read some research done by a startup called Scout Analytics that reveals some interesting data on this trend.

Scout Analytics is a "behavioral analytics" provider. For this research, they tracked web users using some interesting techniques:

Scout Analytics used tracking techniques of device and biometric
signatures to follow the behaviors of hundreds of thousands of named
users accessing paid content products. The biometric signature
identified unique users through an individual’s typing pattern to
eliminate errors in user counting such as account-sharing. The device
signature identified unique devices through data elements collected
from the browser to eliminate errors in device counting such as cleared
cookies. By correlating the named user account, biometric signature,
and device signature, an accurate mapping of individual to devices
could be produced.

And what they found was that a "reliable cookie" overstates user counts 2 to 4 times. That's right, if your analytics software uses cookies, it is possible that your unique visitor counts are 2 to 4 times too high.

This is just one study by one company and I'd love to see more research on this topic, but this is not the first time this issue has been brought up on this blog. In April 2007, I wrote a post on this blog, citing some comScore research, that suggested that cookies overcount unique visitors by 2.5x.

Our portfolio companies are asking me all the time why panel based third party measurement services, like Nielsen and comScore, always generate unique visitor counts that are way lower than their internal numbers. I like to explain to them that there are many ways to count unique visitors and none of them is perfect. The panel based approach, used by Nielsen and comScore, is certainly "old school" and seems out of place in a world where you can just look at server logs. But server logs themselves are subject to inaccuracies.

I advise everyone to triangulate between the various approaches to get some idea of the real numbers. I don't think any service out there today can give you an entirely accurate read. And that is why I am so enthusiastic about comScore's hybrid approach of marrying panel data and tracking pixel data.

I was a founding investor in comScore over a dedade ago, was on their board for ten years, and a fund I help manage still owns a very small amount of comScore stock. So I am not unbiased in this discussion. And this discussion is not just about comScore. It's about web measurement and why, 15 years after the advent of the commercial web, we still aren't measuring it well enough.

For several years, we've been talking on this blog about programs like Y Combinator, TechStars, SeedCamp, and many others. I call them startup accelerators. One thing I've consistently heard is that NYC needs its own startup accelerator.

Well it is happening this summer, courtesy of the city's own NYC Seed fund.

The program is called SeedStart. The program mirrors much of what the pioneer in this sector, Y Combinator, does. It is an eight week summer program. They will select up to ten teams who will get $20k in funding, space to work, and mentorship in return for 5% equity in the business and the opportunity to launch at the end of the program to a large audience.

If you are interested, you need to hurry. The application deadline is the end of the month, February 28th. The acceptances will be made on March 15th. If this is for you, apply here.

This is a great thing for NYC and the first time entrepreneurs who really benefit from this kind of accelerator program. I am really thankful for Owen Davis' leadership in making this happen. I am also grateful to our colleagues at Contour Ventures, IA Ventures, Polaris, and RRE for their support of this effort.

I spent a few days on the beach this Presidents Weekend and while lying in the sun, I read The Blue Sweater, a terrific book by the founder of the Acumen Fund, Jacqueline Novogratz.

Jacqueline has spent her entire career working at the intersection of philanthropy and the financial markets. Her story is all about finding a middle ground between the two.

The Acumen Fund was started almost a decade ago and it is a non-profit entity that receives grants from its donors. But instead of giving away those funds, it invests in entrepreneurs who are making a difference with for profit initiatives in the developing world; particularly India, Pakistan, and Africa.

I have always believed in the power of entrepreneurs and for profit initiatives to change the world. If you share that view, then you should read this book. I'm not suggesting that there isn't a role for pure philanthropy in this world. There is and will always be. But I am also suggesting that in the middle ground between philanthropy and raw capitalism lies a very powerful model that is worth understanding, promoting, and developing. Read this book if you want to understand that model better.

It's time for MBA Mondays again. For the third week in a row, the topic of the post has been suggested by a reader. Last week, Elia Freedman wrote:

"A suggestion for your next post. The logical follow-on is to explain the second half of the TVM (time value of money), which is compounding interest."

Before I address the issue of compounding interest, I'd like to recognize two things about the MBA Monday series. The first is that each post has a very rich comment thread attached to it. If you are seriously interested in learning this stuff, you would be well served to take the time to read the comments and the replies to them, including mine. The second is that the readers are building the curriculum for me. Each post has resulted in at least one suggestion for the next week's post. I dove into MBA Mondays without thinking through the logical progression of topics. At this point, I'm just going to run with whatever people suggest and try to assemble it on the fly. It's working well so far. So if you have a suggestion for next week's topic, or any topic, please leave a comment.

Last week, I described interest as the rate of change in the time value of money. And we broke interest rates down into the real rate, the inflation factor, and the risk factor. And we calculated that if you invested $900 today at an 11.1% rate of interest, you'd end up with $1000 a year from now.

But what happens if you wait a few years to get your money back and receive annual interest payments along the way? Let's say you invest the same $900, receive $100 each year for four years, and then in the last year, you receive $1000 (your $900 back plus the final year's $100 interest payment).

There are two scenarios here and they depend on what you do with the annual interest payments.

In the first scenario, you pocket the cash and do something else with it. In that scenario, you will realize the 11.1% rate of interest that you would have realized had you taken the $1000 one year later. It's basically the same deal, just with a longer time horizon. And your total proceeds on your $900 investment are $1400 (your $900 return of "principal" plus five $100 interest payments).

In the second scenario, you reinvest the interest payments at 11.1% each year and take a final payment in year five. If you reinvest each interest payment at 11.1% interest, at the end of year five, you will receive $1524 as your final payment. Notice that the total proceeds in this scenario are $124 higher than in the other scenario. That is because you reinvested the interest payments instead of pocketing them.

Both scenarios produce a "rate of return" of 11.1%. If you look at this google spreadsheet, you can see how these two scenarios map out. And you can see the calculation of total profit and "internal rate of return".

The fact that you make a larger profit on one versus the other at the same "rate of interest" shows the power of compounding interest. It really helps if you reinvest your interest payments instead of pocketing them. While $124 over five years doesn't seem like much, let's look at the power of compounding interest over a longer horizon.

Let's say you inherit $100,000 around the time you graduate from college. Instead of spending it on something, you decide to invest it for your retirement 45 years later. If you invest it at the 11.1% rate of interest that we've been using, the differences between pocketing the $11,100 you'd get each year and reinvesting it are HUGE.

If you pocket the $11,000 of interest each year, you will receive $599,500 on your $100,000 investment over 45 years.

But if you reinvest the $11,000 of interest each year at 11.1% interest, you will receive $11.4 million dollars when you retire. That's right. $11.4 million dollars versus $599,500. That is the power of compounding interest over a long period of time. 

You can see how this models out in this google spreadsheet (sheets two and three).

Now let's tie this issue to startups and venture capital. Venture capital investments are often held for a fairly long time. I am currently serving on several boards of companies that my prior firm, Flatiron Partners, invested in during 1999 and 2000. Our hold periods for these investments are into their second decade. Of course not every venture capital investment lasts a decade or more. But the average hold period for a venture capital investment tends to be about seven or eight years.

And during those seven to eight years, there are no annual interest payments. So when you calculate the rate of return on the investment, the spreadsheet looks like this. It's a compound interest situation. 

If you go back to the $100,000 over 45 years example, you'll see that a return of 114x your money over 45 years produces the same "return" as 6x your money with annual interest payments.

The differences are not as great over seven or eight years but they are made greater by virtue of the fact that VCs seek to make 40-50% annual rates of return on their capital. If you read last week's post, you'll know that comes from the risk factor involved. The more risk an investment has, the higher rate of return an investor will require on their money in a successful outcome.

If you want to generate a 50% rate of return compounded over eight years on $100,000, you will need to return $2.562 million, or 25.6x your investment. See this google spreadsheet (sheet 4) for the details.

The good news is that most venture capital investments are made over time, not all at once in the first year. So the "hold periods" on the later rounds are not as long and make this math a bit easier on everyone involved (maybe a topic for next week or some other time?). 

But as you can see, compounding interest over any length of time increases significantly the amount of money you need to return in order to pay the same rate of return as a security with annual interest payments. There are two big takeaways here. The first is if you are an investor, you should reinvest your interest payments instead of spending them. It makes a huge difference on the outcome of your investment. The second is if you are an entrepreneur, you should take as little money as you can at the start and always understand that your investors are seeking a return and that the time value of money compounds and makes your job as the producer of that return particularly hard.

We had a very interesting discussion about Chatroulette here last week. Since then, reporters at the NY Times got the founder to talk and they posted the email exchange here.

Here are some interesting facts from the NY Times piece:

• The founder, Andrey Ternovskiy, is a 17 year old high school student who lives in Moscow
• He created the site for "fun" and had no "business goals" for it
• It was inspired by his extensive use of Skype web chat with his friends
• It spread entirely by word of mouth
• He's had to rewrite the code several times in order to allow it to scale
• His relatives invested some "funds" so he could buy more servers
• Right now, he's doing it all himself
• Chatroulette runs on seven servers in Frankfurt, Germany
• He is planning to add more servers in new geographic locations
• Andrey has never been to the USA but would love to visit
• He has ideas for more "weird in a good sense" features
• He's not sure what Chatroulette is any more
• He thinks it would be best to "found Chatroulette" as a US-based company

I think we'll reach out to Andrey and offer him a visit to NYC. I'm still not sure if this is something we should invest it, but I'd sure like to meet this guy. He reminds me of many great young entrepreneurs we've worked with and his story sounds so familiar.

For years many industry observers including me have stated that email is the ultimate social network.

If you take all the people I email with regularly, you’ll quickly figure out my top relationships. Then you can build a social network out of that.

And that’s exactly what Google did with Buzz. But what I learned from that experience is that email is not one social network, it is many. And it may be a mistake to combine them all together into one network.

I’m writing this on a plane on my blackberry. I can’t easily link to to Miguel Heft’s piece in today’s NY Times, but I can quote from it. He says:

“Email, as it turns out, can hold many secrets, from the names of personal physicians and illicit lovers to the identities of whistle blowers and antigovernment activists.”

That was my experience with Buzz too. Many of the people that Buzz suggested I follow are people that I would never think of following on Twitter or friending on Facebook.

One of the more recent lessons for me in the world of social media is the value of the explicit connection. At one time, I thought that automatically building my networks would be useful.

But if you compare Google Latitude, where you broadcast your location, to Foursquare, where you explicitly checkin to a location, I think you’ll agree that the explicit gesture is better.

And so it turns out that implicity deriving social relationships is tricky and potentially dangerous. That doesn’t mean the idea isn’t powerful. So was Facebook’s Beacon. It just means you have to roll it out very carefully.

Reece asked me in the comments the other day what I thought about Buzz. I told him I hadn't had the time to turn it on and play with it. I did that later in the day and have been using Buzz for a couple days. So here goes Reece.

I'll state right up front that there are elements of Buzz that are derivative of Twitter, Tumblr, Disqus, and Foursquare (and many other services) and being that our firm is invested in some of those services, I can't be and won't be unbiased in my views of Buzz.

I think Buzz is Google's finest implementation of "social" to date. It is a given in the web/tech circles that "google doesn't get social". So maybe the most important thing about Buzz is that it puts that idea to rest. 

I also think that the decision to place Buzz into gmail was a smart move. Every time I move to my gmail tabs (I keep gmail open in two to three tabs at all times), I see something like this:

That's quite a temptation to click on Buzz and see what those 42 messages are about. And I've done that at least a dozen times the past couple days.

But that is also one of my big issues with Buzz. That (42) number includes replies to buzzes (can I call them tweets?) that the people I am following leave. Buzz copies the FriendFeed user experience for the most part. And as much as I admire FriendFeed and the people who built it, I don't believe that is a compelling experience for the mainstream user.

When I follow Pete Cashmore in Buzz, I'm basically following all of his fans. And my Buzz timeline is filled with all of their replies to his posts. I think that user experience works well in something like this blog and the comments. I don't think it works well in a mass medium where I want to follow hundreds of people.

Like FriendFeed, Buzz allows me to "pump my data into it". It is an aggregator as well as a updating service. But that poses a problem in some ways. What does this service want to be? Once I pumped this blog, my tumblog, and my twitter into Buzz, I was done. You can all follow me in Buzz here.

I've updated Buzz once or twice since then, but since I can update Buzz via this blog, twitter, and tumblr, I think that will be my favored approach. That's what I do with Facebook and that's what I did with FriendFeeed. 

I'm not saying that everyone will do that. Clearly that is not the case. But of the 9 million posts and comments that people have submitted to Buzz since it launched, how many of those were brand new status updates and how many were items automatically pumped into Buzz and how many were comments?

I believe the most compelling experiences on the Internet do one thing and do it very well and then open themselves up to other services via an API. Buzz is trying to do way too much in my opinion and while that may work for many people, it does't work that well for me.

I think mobile presents some amazing opportunities for Google and Buzz. You can already see Buzzes near you on Google Maps on Android (via a layer in the map). I think that's a very interesting place and way to receive Buzzes. And when you search for and select a specific place on Google Maps on Android, you can "buzz about that place." I did that the other day using my Google phone in my office. That's like a Foursquare checkin and you know that I think mobile checkins are a fantastic way to create social utility.

My partner Albert who still builds web apps, posted about Buzz and he had some favorable things to say about the Buzz API. If third party app developers start building Buzz apps, or building Buzz into their existing mobile apps, that will be a big plus. I'm sure it will happen.

So in summary, Buzz doesn't really do it for me but I think it's an important new entrant in the world of social media and given that it's from google, it's in gmail, it's nicely integrated into their mobile offerings, it has an excellent API, and it will be constantly improved, it's likely that Buzz will turn out to be an important player in the social media world.

Startl is a very interesting new project based here in NYC. It's a startup accelerator (like Y Combinator, TechStars, Seedcamp, etc) but focused entirely on entrepreneurs working in education. It's a non-profit backed by the Hewlett, MacArthur, and Gates foundations which is quite a collection if you ask me. 

Another thing that makes Startl interesting is the partnerships it has with IDEO, DreamIT Ventures, and Berkeley & Noyes (leading investment bankers in education).

Startl has two programs that anyone working on "hacking education" should be interested in:

Design Boost is a five day bootcamp here in NYC from March 15th to 19th. This first Design Boost will focus on the design of mobile apps for learning and will feature designers from IDEO who will help the selected teams with their design and development process. Up to 15 teams will be selected for Design Boost. Details on how to apply are here. The application deadline for Design Boost is February 22nd.

Accelerator is part of the DreamIT Ventures summer program in Philadelphia from May to August. Startl will select five "learning companies" to participate in the regular DreamIT Ventures program. Details on how to apply are here. The application deadline for Accelerator is March 15th.

Startl was started by two friends of mine and our firm, Diana Rhoten and Laurie Racine, and their colleague Phoenix Wang. We've known Diana and Laurie for a long time, they are committed to the concept that the innovation/startup world can have a big social impact, particularly on education. I am excited by their work on Startl and hope that some of you will apply and take advantage of these two programs.