Posts from April 2010

The Executive Session

Every board meeting should end with an executive session. The term executive session is an oxymoron because it is a meeting of all the board members other than the executives of the company.

The first time most CEOs hear of this idea, they hate it. The words "we want to meet without you" strike fear in the hearts of most CEOs. And understandably so.

But it is a critically important part of the Board's job to manage the CEO and to some extent the CEO's senior management team. The Board is required to regularly discuss the performance of the CEO and the senior team, to address their compensation, and to work to make sure the CEO and senior team are working together as well as they can.

You can't do that job with the executives of the company sitting in the meeting. And yet, you want the executives of the company in the Board meeting. The more the better in my opinion, at least for most of the meeting.

So "best practices" says that you should end every Board meeting with an executive session. Some executive sessions last 5 minutes or less. There is simply very little to discuss. Some executive sessions last hours. That's generally not a good thing. Most last 20 to 30 minutes.

I've been sitting on Boards since 1990 and have probably participated in over a thousand Board meetings. To be honest most of them did not end with an executive session. In addition to the CEO's discomfort, there is also the issue of timing. Most Board meetings end in a rush. It is seldom that the CEO's agenda fits into the set time slot. And Board members have schedules to keep. So it is the executive session that often gets skipped.

So I don't totally practice what I preach. But after having participated in a particularly excellent executive session recently, I am recommitting myself to executive sessions. I will need all of your help. Entrepreneurs and CEOs should embrace them and make sure they happen. And fellow VCs should do the same. They are incredibly important and we have a fiduciary duty to do them.

Reblog this post [with Zemanta]

#VC & Technology

Venture Hacks Daily Newsletter

I just subscribed (via email) to the Venture Hacks Daily Newsletter.

They read all the VC and startup blogs, they track Hacker News and Twitter, and they compile all of it in a daily newsletter. Here's a screen shot of what it looks like:


This should be a big time saver for me and for those of you who do the same thing Venture Hacks does every day. If you want to subscribe, here's how.

daily email, daily RSS, or real-time Twitter 

#VC & Technology

Mastering The VC Game

My friend and fellow VC Jeff Bussgang has written an excellent book about entrepreneurs and VCs called “Mastering The VC Game.” If you are an entrepreneur who plans to work with VCs at some point in your career, you should read the book. It’s a fast read. You could easily read it in a cross country flight.

It’s full of stories about entrepreneurs and VCs and how they worked together. There is one about me and Jack Dorsey and Twitter that TechCrunch has excerpted today

I’m not feeling very well today, I think it’s allergies but I’m not sure. So I don’t have a lot in me right now. So instead of reading AVC today, go to Techcrunch and read the excerpt and then hopefully you’ll want to go to Amazon and buy the book.

Reblog this post [with Zemanta]

#VC & Technology


Last week I told Mark Josephson, CEO of, that I have a greater appreciation for the hard hiring work all of you entrepreneurs do right now.

Yesterday all of the investment professionals in our firm sat in a conference room for seven hours and waded through hundreds of amazing applicants for the two open positions we have at Union Square Ventures. We cut the 616 applications (359 for Analyst and 257 for General Manager) down to 75 (47 for Analyst and 28 for General Manager).  

For the past week, my partners and I have done little else but work on this project. It is exhausting work. I believe we could hire any one of at least three quarters of the applicants and be very happy with our choice. But we only have two positions and we owe it to ourselves and the candidates to be thorough and make the very best choice we can.

People and Product are the two most important things a company has. In our case, our product is our ongoing work so we are even more people dependent than most of the companies we invest in. But regardless if you are a product oriented business or a service oriented business, you cannot invest too much time and money on your people. It is critical. 

Our hiring process has always been a bit unusual. We announce the open position on our blog and invite anyone to apply. We do not require a resume. In fact, we prefer that candidates not send one. We want to see the candidate's online presence and we want to know what they have accomplished.

Looking at over 600 online presences and mapping that to what they have accomplished is time consuming work but the insights you get from doing that are incredible. I said to my colleagues yesterday that it was like we interviewed 600 people yesterday.

That number is now down to 75. My partner Albert has a post up on the USV website explaining where we are at in the process and where we go from here (phone screens). This is our third hiring update post so far and I am sure we will do a few more. The candidates seem to love the transparency of this process and how we are communicating. That's a big takeaway for me. If you use the web to source your candidates, you also need to use it to communicate to everyone in the process. It works very well.

I'll end this post with a graphic from Albert's post. Check out his post because he has a bunch more graphical data including some great word clouds. This is a map of all the places in the US that our GM candidates come from. We also have a number of candidates from outside of the US but we couldn't get them on the map for some reason.

Gm candidates

Reblog this post [with Zemanta]

#VC & Technology

Projections, Budgeting and Forecasting

MBA Mondays is starting a new topic this week. It's a big one and I think we'll end up doing at least four and maybe even five posts on this topic in the coming weeks.

I said the following in one of my first MBA Mondays posts:

companies are worth the "present value" of "future cash flows"

The point being that the past doesn't matter too much when it comes to valuing companies. It's all about what is going to happen in the future. And that requires projecting the future.

There is another big reason why projections matter. They are used for goal and expectation setting. Generally speaking goal setting is used to manage the team and expectation setting is used to manage the board, investors, and other important stakeholders.

And finally, projections matter because they tell you what your financing needs are. It is critical to know when you will need additional financing so you can start planning and executing the process well in advance of running out of cash (I like 6 months).

There are three important kinds of projections. I'll outline each of them.

1) Projections – These are a set of numbers, both financial and operational, that you make about your business for various purposes, including raising capital. They are aspirational and are often done with a "what could be" perspective.

2) Budgets – These are a set of numbers, both financial and operational, that the management team prepares each year, usually in the fall, that outline what the company plans to achieve in the coming year. They are presented and approved by the board and the management team's compensation is often driven by them.

3) Forecasts – These are iterations of the budget that are done intra-year by the management team to indicate what is likely to occur. They reflect the fact that the actual performance is going to vary from budget (in both positive and negative ways) and it is important to know where the numbers will actually end up.

Over the coming weeks, I will go through the processes companies use to project, budget, and forecast. Because I do not do this work myself, I've enlisted one of our portfolio companies to help me with these posts. 

I've been working with Return Path for ten years now. Matt Blumberg, CEO, and Jack Sinclair, CFO and sometimes COO, have done over ten sets of projections, budgets, and forecasts for me and other investors, board members, and team members. In the process they have evolved from a raw startup to a well oiled machine. With their help, I will talk about the how three "model companies" go about projecting, budgeting, and forecasting. These companies will be 10 person, 75 person, and 150 person. These are the typical sizes of companies that I work with and are probably also the sizes of companies that most of the readers of this blog are dealing with.

I'll end this post with a picture that Matt sent me last week. This is ten years worth of board books that include Return Path's projections, budgets, and forecasts. The goal of MBA Mondays in the coming weeks will be to get all of you to a place where you can create something similar.

Rp budgets

Reblog this post [with Zemanta]

#MBA Mondays

How A Little Hack Can Be Lifechanging

One of the things I love about mobile browsers is they recognize phone numbers and activate them. You search for a store on your mobile phone, you find the phone number, you click on it, and you place the call.

I've always wanted to work like that at home, but I really don't like calling on my laptop. I love the ability to roam around my home or office with my mobile phone or cordless phone and as much as I love my laptop, it makes a lousy phone.

On Thursday I posted about voice in the cloud and SIP phones. And Ken Berger left this comment letting us know about a Firefox extension for OnSip.

I installed the Firefox extension on my laptop yesterday morning and now every phone number I come across in Firefox is activated just like in my mobile browser. When I click on the number, my wireless SIP phone rings, I pick up the call, and I am connected. I can walk around my home and do the call just like I was on my cell phone.

This is a little hack. It wasn't a huge amount of effort for OnSip to build this extension and the extension has only been downloaded 902 times as of this morning at the Firefox add on service. So in the grand scheme of things, this isn't a big game changing web service.

But it is a lifechanger for me. It simplifies making calls and makes the wired experience mimic the mobile one.

All of this is possible when your phone becomes a device on your network and the Internet just like your computer. As I said yesterday, this convergence is a big deal.

As Ken said in another comment, "Fones are finally fun."

Enhanced by Zemanta

One Graph To Rule Them All?

My partner Albert posted this on his blog yesterday:

But I see at least one flaw with this plan for domination. I simply don’t believe that there is a single social graph that makes sense. I may very well follow someone’s booksmarks on that I don’t want to have any other relationship with. Or take the group of people that I feel comfortable sharing my foursquare checkins with — these are all people I trust and would enjoy if they showed up right there and then. That group in turn is different from the people I work with on Google docs for various projects which is why I would be nervous about using the Microsoft docs connected to Facebook. Trying to shoe-horn all of these into a single graph is unlikely to work well.

He's talking about Facebook and the new services they announced this week at f8. Clearly Facebook is executing fast and well at a scale that few Internet companies have ever reached. It is very impressive to watch.

But I am with Albert on the issue of one social graph for the entire web. I don't think it will happen either. And we are putting our money where our mouths are by backing companies like Twitter, Tumblr, Foursquare, Disqus, GetGlue, and others (remember that are building different social graphs.

Today Facebook is the mainstream social network and where most people keep their entire social graph. The other services I mentioned, with the exception of Twitter, have not built large mainstream social graphs. But I think they will for the exact reasons Albert articulated.

I want to share some things with the widest group of people that is possible. Those things end up on this blog and/or Twitter. I want to share some things with the smallest group possible (like checkins on Foursquare and financial transactions on Blippy). That behavior requires a very tight, very private social graph. 

Facebook sits in the middle of all of this and has created the largest social graph out there. These other social graphs can and will grow in the wake of Facebook. I am not sure if Facebook's ambition is to create the one social graph to rule them all but if it is, I don't think they will succeed with that. If it is to empower the creation of many social graphs for various activities and to be in the center of that activity and driving it, I think they are already there and will continue to be there for many years to come.

I want to thank my friend Mo Koyfman and my partners Brad and Albert for helping me crystalize my thinking about this over the past few weeks. It is a very important topic for those of us who invest in the social web.

Reblog this post [with Zemanta]

#VC & Technology#Web/Tech

SIP Phones and Voice In The Cloud

Snom m3  Yesterday evening, I got a SNOM M3 phone that I had ordered from Amazon. I unpacked it and charged the phone. While it was charging, I pointed my web browser to and set up a 30 day free trial account. It took me about five minutes to set up an account, one user, and get a phone number. I then took down the sip configuration data from the onsip account I had set up.

I went back to the M3 which was still charging, and plugged the base station into my home network. The phone gave me its IP address on my home network. I then pointed my web browser to that IP address and entered the sip configuration data into my phone's configuration. And then I rebooted the phone's base station.

That whole process took about ten minutes, maybe less. I then picked up the M3, dialed a number, and made a phone call.

I realize that voice over IP (VOIP) has been around for over a decade. I invested in a VOIP company in 1997 so this is not new technology. 

But there is something really powerful when voice moves into the cloud. In about five minutes, I was able to provision myself a phone number in the cloud that had dialtone. And then make a phone call.

I chose to use the M3 phone, but I could have chosen any number of SIP phones. Here are four pages of SIP phones you can buy on Amazon.

I then walked down the street to our new apartment. I plugged the base station into our internet network in our new apartment, picked up the M3, and made a call.

Contrast this to the experience of getting a phone line provisioned from Verizon or any other traditional phone company. That requires talking to someone, getting a tech to come out and provision the line, and then running wires around the house or office.

It has taken VOIP at least a decade to get here. We needed it to get into the cloud, and we needed get a wide selection of high quality sip phones. But we are here now. You can provision a phone line from the cloud in five minutes, you can connect any number of high quality sip phones to that cloud-based phone number. And you can use that phone and phone number from any internet connection you want.

I am so happy to see voice and data converging and moving into the cloud. It's the way I've always wanted voice to work. I think we are going to see a lot of big changes as a result of this convergence.

#VC & Technology#Web/Tech

1 Billion Pageviews

Yesterday our portfolio company Tumblr announced that they had quietly done a round of financing with the existing investors. And they also announced that the Tumblr service is serving over 1bn pageviews per month.


I thought it might be useful to put 1bn pageviews in perspective. I went into comScore and looked at how many web properties serve over a billion pageviews worldwide each month. The number is 57. Here are the comScore stats on that. Monthly pageviews are the farthest column on the right.

Comscore 1bn pvs

Now, it is important to state that we are comparing apples to oranges here. Tumblr is not measured directly by comScore. The data in that first chart comes from Quantcast. And comScore does not roll up all the Tumblr pages that run on other domains. Quantcast does that. If comScore did roll up all of Tumblr and if it measured Tumblr directly, it might not report 1bn pageviews per month. Or it might. 

But that's sort of beside the point. 1bn pageviews is a seriously big number. Only 57 properties in the world report that kind of number on comScore. 

And Tumblr got there on $5mm of invested capital (we just put in another $5mm). 

There are several reasons why they were able to do that. First David and Marco are an amazing team. They did all of the work themselves for the first year. And now they have a small team who maintain the service and develop new features. They get a lot done with very little. And finally the service is amazing. The Gotham Gal started a tumblog this past weekend. She has blogged for almost seven years so she is no novice to this kind of software. She got her tumblog up in about five minutes and posted a bunch of stuff. And then she turned to me and said "wow, this is so easy."

There are some lessons here. First, make your software super easy to use. Second, you don't need hundreds of employees to build a big time web service. You can keep it lean and scale if you have the right team. That's how Tumblr got to a billion page views and we just made a bet that they will be able to take that number a lot higher.

#VC & Technology