Monday, September 24, 2012

How to kiss your elbow

Even before Paul Graham's Growth post the recent VC meme was that entrepreneurs just aren't as ambitious as they used to be. They are too careful, husbanding their cash rather than boldly investing it in growth. I've heard this kvetch four or five times since Labor Day, each time from a smart and well-respected VC. They blamed the ease of raising seed money compared to the relative difficulty of Series A and B money, the incubators and angels corrupt the entrepreneurs.

I've had this conversation with many of my entrepreneurs over the years: the market's going to pass you by, carpe diem, etc. But I never really thought of it as a trend, it was just the usual learning opportunity for first-time entrepreneurs--there's no starting gun, no one is going to tell you when to start seizing the day, just do it. This has always been a conversation in the fifteen years I've been investing in early-stage. Is it really now a wider phenomenon?

I don't know, but I heard it enough that I ran the idea by a couple of my entrepreneurs. The entrepreneurs sighed and rolled their eyes. You advised me to raise just the money I needed to get to the A, they said, Paul Graham says I should "not need money." Which is it, they ask? Do you want me to not need money or do you want me to get aggressive and raise my next round sooner, maybe without the metrics I need to get a good valuation?

Fair enough, the world is full of conflicting advice. And I understand how annoying it can be when it's the same person giving you both sides of it.

And let's be clear: the dichotomy is not necessarily between the lean startup and "go big or go home." Suster makes the case that the latter is not always the best route. No argument. This is more along the lines of when to hit the gas, not if. I only invest in entrepreneurs who tell me they can and want to go big and then I try to get them to stick to what they told me. My favorite question when these conversations happen is "what's the bottleneck?" What can we do, who can we hire, who do we need to partner with or talk to or get in front of to make what we both think can happen actually happen, now?

When I ask this, the entrepreneur always knows the answer. In fact, they've usually been thinking about it night and day. But they also usually want to take it more slowly than I do. They want to take less risk. Once you've spent the money, that money is gone, and if you're too early it's gone for good.

So how do you know when it's not too early?

Christensen tells a story in The Innovator's Dilemma about Honda's entry into the US motorcycle market. Honda's entry strategy, after much examination, was to give Americans what they clearly wanted: large bikes to ride long distances at highway speeds. Honda's expertise was in designing small, efficient engines, as in their Supercub delivery bike. But Honda designed a big, fast bike for the American market and in 1959 sent three reps to live in LA to begin marketing it.

The bikes sucked. At highway speed they leaked oil and burned through clutches in record time. The cost of sending replacements for the warrantied bikes almost put Honda out of business.

To burn off steam the Honda reps used to go out and ride their Supercubs--small, zippy, 50cc bikes--in the hills east of LA on the weekends. Over time people started asking where they could buy one of these 'dirt' bikes. The reps special-ordered Supercubs from Japan for people. After a couple of years of this a buyer from Sears tried to place a bigger order. Honda ignored him. Finally the Honda reps convinced Honda to change direction, that the big bike strategy had failed but a small bike strategy would work. Innovative distribution channels were forged, sales took off, market entry was achieved.

But if Honda had been more aggresive with their strategy in 1959, if they had sent reps to Miami and Seattle and Dallas and Atlanta and Denver and Las Vegas at the same time as LA, there's an excellent chance Honda would have not only failed to enter the market but actually gone out of business. By taking it slow until they had found a product that fit the market, they bought the time they needed for success.

Bit of a buzzword that, product market fit. Mark Andreesen says "you can always feel product/market fit when it's happening." Unfortunately, this is simply not true. Honda took a couple of years to feel it and even longer to properly trust it. In B-to-B startups you can have a lot of buzz and a few amazing clients banging your door down and still have a product that doesn't really do much. Or you can have a product that is absolutely amazing that great clients are beta-testing but that no one is paying for. In B-to-C you can have a hundred thousand users and still be serving nothing but tech industry curiosity seekers. Or you can have millions of members and few users. These are not product-market fit.

There is a case for going slow, to a point. Your product has to provide real value to your users. You need to have a viable business model, know the metrics you need to make it work, and be on the path to meeting those metrics. And then you need a way to get to customers and convince them to sign up and/or pay. You need all these things before you can feel comfortable ramping up the spend. But if customers love your product, if those customers are profitable customers, and if those customers start presenting themselves, either directly or by making themselves extremely easy to get in front of, then you should let them become customers. And if there are more of them than you can get in front of personally, then you should hire a salesperson. If there are more of them than your salesperson can get in front of, then you should hire more salespeople. If you have a product and you have a market for that product, you should stop worrying and start scaling.

In the old hockey-stick curve there is a flat part and there is a steep part. That transition, the elbow in the curve, is hard to see, especially when you're spending all your time trying to run your company. Here's a question to ask yourself: if you think you can double revenue next year, what's holding you back from increasing revenue by 10x? If the answer is that there's no market yet, then keep grinding away at it. If the answer is not enough people or hardware for scaling, then start spending the money on hiring them, today.

The best possible Series A story: "we don't really need your money, but if we had it we could grow ten times faster starting tomorrow." Term sheet before you get home, guaranteed.

Thursday, June 28, 2012

Your personal data is not worth anywhere near what you think it's worth

I see a lot of Root Markets-like businesses. Companies creating a way for people to own their own data and profit from it rather than letting someone else profit from it. The idea is appealing: other people are selling your data, it's your data, why shouldn't you sell it yourself?

But most of the people I talk to don't have a good answer to the basic business question: can you sell your product or service for more than it costs you to buy or make it? In this case, can you sell personal data for more than it costs to garner it?

Well, can you?

The IAB says that in 2011 there was $31.74 billion in US interactive ad spend [pdf]. There were 245.2 million internet users in the US in 2011 according to, using data from Nielsen and the ITU. That works out to slightly less than $130 in ad spend per internet user per year in the US.

Here is a breakdown of this per capita number, by channel, and a guess as to how much is potentially available for third party data sellers:

$ per Addressable
Channel User Market
Search 47% $60.84 $0.00
Display / Banner 22% $28.48 $7.12
Classifieds 8% $10.36 $0.00
Digital Video 6% $7.77 $1.55
Lead Generation 5% $6.47 $3.24
Mobile 5% $6.47 $1.29
Rich Media 4% $5.18 $1.04
Sponsorship 4% $5.18 $0.00
Email 1%   $1.29   $0.97
Total $129.45 $15.21

The $130 needs to pay for several different functions. The $28 for display, for instance, pays for account management, creative, media planning, targeting, media buying, ad serving, analytics, verification, and--not least--the actual inventory the ad is placed in. I'm guessing that the maximum amount available to a company selling data to target display ads is 25% of the ad revenue*. The opportunity to use data to optimize lead gen is potentially larger, while the opportunity in sponsorship, classifieds and search is pretty much nil**.

If this is right, and given the fuzziness of the IAB numbers, it means that there is maybe $1.00 to $1.50 per person's data per month available to data sellers.

But keep in mind that Google does not need your data. Nor does Facebook. They are a large part of the market. Your data is competing with everyone else's data--first, second, and third-party data--for this $1 per month. And some of the data you are competing with is so closely tied to the awareness generating process that it can't be pried away and placed in a 'wallet' somewhere.

Take context. The context of an ad can account for somewhere between 50% and 90% of its effectiveness. Context correlates to demographics, purchase intent, state of mind, and behavior. If you are looking at a review of the new Mac Book Pro I don't need any personal information to make an educated guess that you are in the market for a new computer. I can confidently put a computer ad next to that article without any other data, and the only way someone else can intermediate my guess is by blocking the content or ad entirely. Same argument different data for Facebook, and for much mobile usage.

This means that of the $1 per month much less is actually available to you as a collector of the data.

The original Root business model was to allow users to own their data and rent it out to people who wanted to market to them. The problem: users think their data is worth far more than $1 per month. But $1 per month is all that is available, on average. To a single company, it's maybe $0.10 at best. And then there has to be a commission paid to the new intermediary--the Root-like company. The user ends up with maybe a dollar a year. Nobody cares about a dollar a year. There's no business model. I could even imagine a world where each user was worth $0.20 a month, but that price is still nowhere near where it has to be to have users take it seriously.

There is a business model for businesses that gather data very efficiently. There are several pretty large companies that do this. But they have figured out a way to gather the data for much less than $0.10 per person and to collect data on hundreds of millions of people. The Root model simply costs more per person than the data is worth.

I spent several years of my life trying to build a business that lets people take control of their own data while still leaving a way for marketers to find them. I believe in privacy. And I believe that marketers finding customers is key to economic efficiency. I would love to see someone square this circle, but the Root model is not the way to do it.

* This takes into account the fact that I think the IAB/PwC revenue number is the amount paid to publishers, not the amount spent by marketers. The amount spent by marketers may be 50% to 100% more than that paid to publishers on average. Hard to know. This is an important point though: marketing is much, much more than advertising. The amount that companies spend on marketing in total is far higher than the amount that publishers make from selling ads. There are companies selling data that sell into this marketing market that are worth billions, they are not the focus of this post.
** The best businesses are the ones where everyone else thinks you're wrong. My saying there's no opportunity means that if you have a way to use data to optimize these channels, you may have an opportunity that no one else has seen. I like those.

Monday, June 18, 2012

Great Riches and Low Theft

In early 1998 I walked the open plan floor of what was then one of the largest web development shops. The founder was giving me a tour so I could see the scores of web developers working diligently. They looked the part, the founder looked the part, the place had good energy. I liked it.

The founder was looking for venture capital to expand internationally. He was a much more experienced businessperson than I was and he knew it. At one point he gave me a sly look and said "How do you know that I didn't just hire a bunch of extras to fill an empty office building floor for the day so I could impress you?"

Now I had done my due diligence and some of the people who would have had to be in on that sort of scam were people who had more to lose lying to me than they could possibly gain lying for him. I trusted my due diligence and I knew the company was real. But I no longer trusted him. At the end of the tour I told him we were passing on investing.

He was pissed. He went over my head to the CEO of my company, directly and through common clients. When I was called on the carpet to explain myself I said only that I did not want to work with that founder. I did not say why. The CEO did not give me a soul-searching stare, he did not grill me, or even ask why that would be so. He knew me well enough to let me have my reasons. He just waved me out of his office.

It's an odd fact that our capitalist system--our brutal, unsentimental, Darwinian, sink-or-swim system--relies almost entirely on its protaganists' ethical behavior to function. Our entire economy relies on trust. You probably don't think about this much. Most people don't think about it at all. I think about it a lot. What I do--what all VCs do--would not be possible without the honest behavior of an overwhelming majority of founders. If even 10% of founders decided to start cannily lying the entire startup ecosystem would come tumbling down shockingly quickly.

I hear objections. Let me distinguish between transactions and relationships. Many transactions are entirely caveat emptor: you need to know what you are doing and what questions to ask. Transactions have a simple API and learning how it works is your responsibility. But a business relationship is different: it is too complex, there are too many ways to be dishonest. It is not possible for both parties in a business relationship to verify everything the other side has told them; if they had to the cost of doing so would make it infeasible to have business relationships at all.

There are many gradations and steps between transactions and relationships; navigating through them requires experience. But if you do not trust a person you should not have a business relationship with them.

Some of the oldest business advice in the world: "A good name is better than great riches."* What happens to those of ill-repute? "The sons of men of no name, they were driven out of the land."** In our community a bad reputation results in being driven out of the land. If you're known for not being trustworthy your career amongst the highly interconnected venture capital community is probably at an end***.

The flip-side has always been that our community hesitates to accuse other people of certain types of ethical lapses. I can only think of one time in my fifteen years of venture investing that I have gotten a third-party reference from a venture capitalist that called someone's ethics into question. The closest a VC will come to saying something bad about someone is to refuse to say anything of substance at all. If you don't like someone, you don't have to do business with them. But impugning someone's character can put their life's ambitions at risk. You need to be extremely sure of what you're doing and cognizant of the effect your words might have before you do this. If you don't, you can do a great amount more damage than your dislike of that person deserves.

I won't do business with someone I don't trust. When someone I worked with has turned out to be a liar I have ended my business relationship with them. Luckily I have not had to do that often and not in almost ten years. But likewise I won't have anything to do with someone who puts someone else's life's work in jeopardy by carelessly judging their ethics in public. These offenses--breaching trust and baseless accusations--are two sides of the same coin. The ignominy of the offenders should be likewise the same. If a good name is better than great riches then heedlessly sullying someone's reputation is low theft, and leaves the perpetrator, the victim, and the rest of us equally impoverished.

Proverbs 22:1
** Job 30:8
*** Every field has different standards for trust. With startups we expect baseless optimism for instance, where in academia this would be frowned on. With startups we expect confident predictions of the future as if it has already come to pass, while in banking this would be looked at askance. People in the community know the norms. And, in our community, are willing to give allowance for the fact that many entrepreneurs were not part of the community before starting their company so may be unfamiliar with our ways. Mistakes made with good intentions are not ethical lapses, they are mistakes.

Saturday, May 12, 2012

Response to comments: Training VCs

Many of the comments on yesterday's post were about training future VCs, or not. Both Brad Feld and Fred Wilson said they did not have junior VCs because they did not want to burden entrepreneurs with inexperienced VCs. This makes a ton of sense. But, then, where should experienced VCs come from? Andy Weissman comments that perhaps VCs are best trained by being entrepreneurs.

You either believe that Venture capital is not a profession--i.e. there are no special skills or knowledge needed that can't be picked up as you're doing the job--or it is. If it is then it looks like the best firms are akin to 'boutiques' in other professions. In fact, almost all the firms are akin to boutiques in other professions. Of course, in other professions boutiques are formed by people who were trained at the mainstream firms. If there were no mainstream firms there would be no people to form boutiques.

Law firms train lawyers. Accounting firms train accountants. Banks train bankers. Even advertising agencies train advertising people. VCs by and large do not train VCs. Maybe VC is not something you can learn by just doing VC, although Fred is a prominent counter-example. Professions train professionals partly because they think their professions are important, so they feel the obligation to pass on embedded knowledge to the next generation. And partly because they can skim some of their underlings' earnings (thus the pyramid structure of professional services firms.)

VC does not have the pyramid structure of some professions, like law or accounting, where most tasks can be delegated with oversight to junior people. So it's true that junior people in VC probably would cost more than they generate if they were truly being trained (as opposed to just cold-calling and spreadsheet-jockeying.) But if we care about entrepreneurs, as we all profess to do, we should want not just the best for today's entrepreneurs, but also for tomorrow's.

If you do, and still don't think training VCs is worthwhile, then it must be that you simply do not believe that VCs can be trained, that VC is not in fact a profession at all.

I do not think this is true. The best--in fact almost all--VCs have historically come from one of five places: VC, banking, law, technology firm management, or journalism. Check the VC genealogy to confirm this. (I don't think any journalists are represented there, but Mike Moritz is a prime example.)

Each of these paths teaches people some of the necessary skills to be a venture capitalist, but not all of them. Witness Kleiner's and Perkins' struggles as they started out, making ridiculously wrong bets on markets they did not understand. Or the revealing comment Fred Adler made* about two of his partners that left to start their own fund: "These fellows came out of Citicorp where they were quite senior and they didn't go through [my] intense interrogation" justifying the investments they were making. The implication being that the two partners did not know enough to make good investments and Adler did not feel they would accept his instruction since they were so senior. In other words, being senior at Citicorp had not taught them all they needed to know to make good venture investments. Those two partners must have learned something on Adler's dime though, because the fund they started was Accel**.

Many venture capitalists made similar mistakes early on. The ones that didn't seemed to either have extensive angel investing experience (and so their early mistakes are not part of the record) or they "played the follower strategy" (as Wilson has it) and managed to get into more experienced VC's deals in order to learn the business.

So if specialized knowledge is needed, how to generate it? Kauffman has their Fellows program to train VCs. Andy thinks being an entrepreneur is the best training. I disagree with both. I think only doing the job teaches the job. And since no one wants anyone doing the job who doesn't know the job, this means a long apprenticeship. But the best VCs seem to not be interested in having apprentices. So, then what?

In my opinion, if we want better trained VCs, then either the culture has to change so VCs feel an obligation to train the next generation***, even though it costs them money, or the LPs need to start looking out for their future returns in addition to their present ones and compel VCs to have a bench. It would be interesting to hear from experienced venturers how they learned the business.

* Quoted in John Wilson's The New Venturers.
** Wilson's book was published in 1985, Accel was founded in 1983, so Wilson had no way of knowing that Accel would go on to be one of the premier venture funds. This just makes the quote that much better.
*** I would be happy if VCs would even just blog more with other investors as the audience, instead of writing the same frickin how-to posts for entrepreneurs over and over. The world does not need another "How to Read a Term Sheet" post, it really doesn't. It could, however, use a few more "How You Will Get Screwed if You Write a Bad Term Sheet" posts.

Friday, May 11, 2012

On fixing VC ourselves

What good is it for me to sing helplessness blues
Why should I wait for anyone else?
- Fleet Foxes, Helplessness Blues

That's for Fred Wilson.

In his post last month "Can the Crowd Be More Patient", Fred says
We need new medical approaches to preventing and/or curing disease. We need new scientific approaches to generating, storing, and being more efficient with energy. Maybe we need more space exploration. Maybe we need more undersea exploration.
He says this in the context of not being able to fund these things because that is not what venture capital is. But you know what? At this exact point in time Venture Capital can be whatever Fred says it is. If he wants a 20 year fund instead of 10, he could raise it. He wants an evergreen fund? He can do that. I think he could probably raise a fund to do whatever he wants.

So I'm not sure what he's saying. But if what he's saying is that these things are not fundable, no matter the time frame, that they are simply bad early-stage investments, then I strenuously disagree. Investing in the things that make our collective lives better--the very things we think of as progress--should be the only good investments. Venture capital is a means to an end, that end being the commercialization of innovation that makes our lives better.

Here's a thought:
Entrepreneurship is a self-actualizing and a self-transcending activity that—through responsiveness to the market—integrates the self, the entrepreneur, with society. Unavoidably, therefore, entrepreneurship is an exercise in social responsibility. To suppress or constrain innovation and improvement—and their implementation—ignores a society’s needs and wants, holds it back, and diminishes its future. Entrepreneurship is the unique process that, by fusing innovation and implementation, allows individuals to bring new ideas into being for the benefit of themselves and others. It is sui generis, an irreducible form of freedom.
That's from the Kauffman Foundation's 2008 report on Entrepreneurship in American Higher Education [pdf]. Meanwhile, this week Kauffman had a new report [pdf] that--as Ed Zimmerman had it--blames investors in VC funds for being co-dependent enablers of bad VC behavior (for the tl;dr, see Fred Destin's excellent post on the report.) I've heard a lot of opinions on this report: some agreeing, some denying Kauffman's conclusions (for instance, Brad Svrluga's rebuttal.) But while the degree to which venture investors are doing a bad job is arguable, the fact that, as a whole, we are is not.

Nice as it would be to agree with Kauffman and say "I'm doing a bad job because I'm being managed poorly," that's no excuse. Every VC I know complains about bad practices in the industry, because bad venture capital practices affect us all. And while the bad behavior might make short-term economic sense, as outlined in Kauffman's report, I am not in this for the money and neither is anyone else I know.

Yes, I want to make money; in fact, I need to make money if I'm going to keep investing, I'm also competitive by nature and making money is how we keep score. And then, if it's true that the market ends up choosing companies as winners because they are the ones that contribute most to societal growth, then making money is not a bad metric in the long-term. But the real reason I'm in the business is that I want to contribute, I want to be useful. "Entrepreneurship is an exercise in social responsibility." That's what I want to enable. Every good VC I know feels the same way, but almost all of them feel helpless to change the current broken incentive model.

How would we do that, as venture investors? I'm sure there are smarter people than me thinking about this, but here are a few ideas.
  1. Change LP Behavior.

  2. I agree with Kauffman about misaligned incentives, not that my agreeing is going to change anyone's behavior except my own. But if Fred Wilson and his ilk agree with Kauffman, then it does make a difference, if they want it to. Fred talks his talk in public and he walks his walk in private, so maybe he's already in the process of convincing LPs to accept a different model so he can make the investments he thinks make a difference. I hope he is. And I hope he's not just volunteering USV, but the whole industry. When you're really good at something, explicitly raising the bar for the entire industry is a killer strategy, so convincing LPs to hold VCs to a higher standard would just be good business for him.

  3. Professionalize VC.

  4. One of the odd things about venture is the lack of seriousness about what we do. Venture is the only professional services business which does not think training its employees is a good idea. Witness Brad Feld's comment--ironically, in the textbook that Kauffman asks its Fellows to read--"We don't intend to hire associates and train them; [when we retire] we are just going to shut shop and go home. Done!" This après moi le déluge attitude means that our industry continues to be half-staffed by people who half know the job. I am constantly amazed at the crazy things other angels do, usually sins of omission, and VCs I know express the same sentiment about other VCs. In no other profession do they expect people to just show up and do the job well. In our profession many show up and do the job poorly. We all suffer. If we care about innovation--not just making money--we should be training people how to invest in and manage investments in startups.

  5. Think bigger.

  6. Wired publishes "When Will this Low Innovation Internet Era End?" at the same time as the Guardian has an article called "Has the Internet Run Out of Ideas Already?" Rick Webb calls a bubble in the very part of the startup world that has the least to do with societally useful progress (progress defined as improving GDP per capita and thus living standards.) Fred's complaint: it's true.
    As an industry, we are funding too many ideas which do not make a difference. We can take pride in helping build companies that create jobs. But creating jobs is not as good a goal as we make it out to be if those companies and those jobs disappear three years later. Jobs come and go, but technological progress is forever. Funding progress makes a difference. This is not a "they promised us jetpacks" rant. Jetpacks are stupid. I don't want a jetpack. I don't want you to have a jetpack. I think all of us having jetpacks would not make the world a better place in the least**. That's not progress. Google was progress. Twitter is progress. These are tools that enable us to think better, to communicate better, to find the things we need to know more efficiently.
    Paul Graham had a post on "Frighteningly Ambitious Startup Ideas." I think that his ideas as a whole were not ambitious enough. A new search engine, replacing email? OK, those are big ideas, and they're ideas a small team can make progress on over the course of a YC session. But the big ideas are more akin to his latter ones: a wholesale reconfiguration of existing industries that suck, efficiency-wise or societally: Hollywood, medical care. I like companies that are trying to destroy and replace our most hated industries***. But there's big and there's bigger. How do you create a company that doesn't solve a specific problem but rather makes us better at solving problems in general?
    Google and Twitter both make us better at solving problems. They don't just make us more efficient, they make us more efficient at finding efficiencies. They are tools to make our brains better. But they are primitive tools. We should be building companies that make us--as a species--more creative, better problem solvers. Our bottleneck in making more progress is ourselves as people: we can not on our own think any harder or better. Where are the startups that change that? I don't want a company that cures a disease, I want a company that helps researchers figure out how to cure diseases. The best, and best returning, industries that venture capital has funded have done just this: the computer industry, the biotech industry. These were meta-tools.
    What's the next meta-tool? If I knew I'd be building it. I don't know. So instead I spend my days looking for the type of people who think they do know. That is the job of the venture investor. We need to do more of this, and less of what we are doing now.
Kauffman's report implicitly suggested that there should be only 20 funds, each of $400 million or less. If this advice were taken, the venture industry would be a fifth the size it is today. Almost all VCs would be out of jobs. Entrepreneurs would be back in the bad old days when ARD funded DEC with $85,000 and received 70% of the company in return. No one wants that, except maybe the LPs, but that's what is in the cards if that's what it takes to make the investment class work. To avoid that, we--the venture investors--need to do better and we need to do it preemptively.

We are not helpless, we should not wait for anyone else.

* Many small companies do make a difference in people's lives, and certainly do so in the aggregate. But in some sense it's just as much work to build a small company as a big one. My philosophy is to aim for the moon and land on the roof: a big idea can produce a moderate size outcome or a big outcome; a small idea can produce a small outcome and that's it. Many entrepreneurs I've worked with have showed up with a modest idea. I've pushed them all to be wildly immodest--there's always a big idea surrounding a small idea, go for the big idea.
** Just go buy yourself a Ducati.
*** My current bets are in banking, mobile telecomm, and, of course, advertising.

Tuesday, April 24, 2012

Venture Capital Family Tree

About six months ago my friend Chris Fralic (@ChrisFRC) invited me to a screening of Something Ventured, a film about the origins of the US venture capital industry. Definitely worth checking out. One of the things it got me to thinking about was how intertwined the early VC firms were. So, in the spirit of one of those genealogies of rock music posters, I gathered some data and made a visualization. It's not pretty like the rock and roll one. And it's woefully incomplete, which I need your help on. We'll get to that.

The whole thing

Zoomed in on some interesting '90s reshuffles

I used jquery and d3, although I had to customize d3 a bit. Because it's d3 and there are a lot of svg objects, it could be slow on slower computers. Also, screen size makes a big difference here, not recommended for mobile viewing.

I was trying to put in firms from the early days and firms that were critical links between the early days and today, so there are a very few firms founded in the last ten years in there. It's a historical study, not a contemporary view. The founders of the firms are noted, but not seminal later partners (i.e. Kleiner and Perkins but not Doerr or even Caulfield.) Although new partners can have a huge impact on a firm's direction*, I just didn't have time. Maybe in the next iteration.

Firms founded in New York are blue, in Boston are red, and in Silicon Valley are green. Others--including those that do not yet have an entry for place--are orange. Type a firm name or founder into the box (there's autocomplete) and click to zoom in on that firm. Also, pan and zoom using the mouse.

Venture firms never really die, they just fade away. And some firms stop being VCs (in the '80s many firms abandoned venture for PE) or were financial orgs and became VCs. This is denoted by a 'tear' at the beginning or end of the firm's bar--the firm had a life before or after, but was not an active venture investor.

I've also started adding noteworthy investments, but there are only a few. As far as I can tell, there is no extensive list of who backed who when. Which brings me to my ask.

I've put in info as I came across it for the last six months, but I have other commitments, so it's been slow. And the easy info sources are running dry. So I slapped on a form, hoping you all would help. All contributions are welcome, but in the spirit of the thing, I'd like to prioritize adding firms that were either critical links between the past and present, that trained a bunch of people who went on to found their own firms, have been influential for a long time, or were influential in the past and have disappeared (i.e. TVI, MPA&E.)

When contributing investments made by firms, I would rather not add every investment. I've tried to add investments that were important or household names. This is a public historical document, I think it's more interesting (and puts the better foot forward) to show that Starbucks and McDonnell Aircraft were venture backed (or Pizza Time, for that matter, even though it failed) than, say,**.

To contribute either click on the '+' button on the upper right to add a new firm, or click on the name of a firm to edit/augment their info. When you hit 'submit', it should reflect locally, but it doesn't add it to the database (I'm not a back-end guy) it emails me. I will edit and add data, I don't expect to get a ton of submissions. The data is open source (cc by-sa), and any contributions will be considered open as well. I will add your name to the contributor list on the help page if you put it in the 'Comments' box on the form (there's no other way for me to know who you are.) Also, put your email address there if you want so I can contact you re your submissions.

But please, do submit! It struck me as odd when doing the research here how little the venture community values its roots. Law firms have web pages and sometimes whole self-published books celebrating their founders and history. Your typical VC firm comes across as if it's in the witness protection program. It's crazy that I can't figure out who all four founders of Menlo Ventures are and where they came from, or who backed Federal Express and when. I've got decent google-fu, and I looked, trust me. Someone out there knows, and you should enter this stuff. The mainstream industry is now some 50 years old. We are in danger of losing our past.

If someone knows of a source of data for this (that is either free or you can get me access to), I will port stuff.

Primary sources were firm web sites, Wikipedia, and The New Venturers by John Wilson, a great book now out of print. Some data was taken from Venture Capital at the Crossroads, Creative Capital, Valley BoyThe Startup Game, and Elfer's Greylock. I'm planning on skimming Done Deals and Venture Capitalists at Work for more. Other sources are in with the rest of the data as 'cites'.

*  In a couple of case, Paul Bancroft joining Bessemer from DG&A, for example, I think it was significant enough to consider that the start of Bessemer's venture activity, even though that is not strictly true.
** Well, hard to say. I can only think of interesting companies because I only remember the interesting ones. Kozmo might actually be an interesting investment from a dot-com bubble point of view. Maybe just send me whatever you want and I'll figure out some way to highlight some and not others. I don't know.

Thursday, April 19, 2012

That joke's not funny anymore

"[Y]ou have to assume that humans are capable of looking at facts, finding root causes and formulating solutions. On my planet there's not much evidence to support this assumption... If humans had the ability to look at facts and make good decisions, think about how different the world would be. There would be only six kinds of cars on the market and nobody would buy a car that was second best in its price range. There would be no such thing as jury selection since all jurors would reach the same conclusion after viewing the facts, and all elections would be decided unanimously. That's not the world we live in. Our brains are wired backwards. We make decisions first--based on irrational forces and personal motives--then we do the analysis. The facts get whittled until they fit into the right holes."
-- Scott Adams*

[The first section of this post was previously published on AdExchanger, and there's an excellent comment thread there, so you should go read it. This is the rambling version.]

Jokes all start with one of a few stock set-ups. "A man walked into a bar." But the punchlines are all different. VC pitches are the opposite: the set-ups are all different, but the punchlines are all the same.

Stop me if you've heard this one before. "We enable the half of advertising that is not yet online: brand advertising!" Been hearing that one for more than ten years now. It permeates the hopes and dreams of every adtech entrepreneur and investor. And still no one has cracked the code.

My friend Tim Hanlon got together with Tim Chang a few weeks ago over on AdExchanger to offer some reasons why this might be. Worth your time, but here's the tl;dr**:
  • No standards or consistent measures of “success” other than outdated or inadequate metrics like CPM and CTR;
  • Limited real-time intelligence;
  • Unsuitable display ad formats; and
  • Lack of creativity in formats.
The prize is huge. As Tim and Tim point out, two-thirds of advertising spending is brand advertising, but online only one quarter is. In fact, if brand advertising dollars moved online in the same proportion that sales advertising has, it would almost exactly close the famous gap between time spent online and ad dollars spent online. The $50 billion gap that Mary Meeker mentions is exactly equal to the missing brand spend.

So I understand the urgent desire to figure out online brand advertising. If we did, we'd more than double the online advertising market. Online pubs would rejoice, online marketing pros would have more excuses to go out drinking with prospective clients, my portfolio value would quintuple overnight. Good things all. And I appreciate the optimism that Tim and Tim have, their willingness to keep suggesting solutions. But I think it's the triumph of hope over experience. Each of these things has been tried, and tried and tried. And still we believe that this time it's different, that this year an online branding play will work. Online video maybe, or Facebook, or Pinterest. Every new company is touted as the one that will make branding work online.

But what if we try all these things, like we've tried everything before, and they don't work? What if we eliminate all the possibilities and what remains is... nothing? I'm going to be branded a heretic for saying this, but what if online just doesn't work for branding?

I mean, not to be defeatist, but we understand branding pretty well. Marketers have been creating brands nigh on one hundred years now, it's not a black art. And the solutions I hear, even Tim and Tim's, are not untried. More, they are not what makes brand advertising effective in other media. I don't buy that these are the solutions. I think it's distinctly possible that there are no solutions.

Maybe the medium itself is antithetical to the way brands are built. Like direct mail, maybe the very fact of delivering your message in a low-budget, specifically targeted way can not in any way build a brand. Brands attempt to exist autonomously, they are objects of desire, they want to distinguish what otherwise is indistinguishable. The psychological processes of branding are inimical to the idea that the brand has been chosen for you. Brands do not choose you; you choose brands. Brands are aloof, they aspire to be the Platonic ideal, their competitors just shadows.

Perhaps. I mean, I could be wrong. It could be that even though we sell ourselves to clients as brand-building geniuses, we don't know what we're doing; that we're groping in the dark, throwing random darts and just haven't hit the bulls-eye yet. We could be a bunch of monkeys at typewriters and Shakespeare will just roll on out one day. Could be. But it seems unlikely.

What if I'm right? What if online branding is a mug's game? If it is, it won't be too much longer before marketers get wise and just stop listening to online branding pitches. Maybe they already have. Maybe they never did listen to them. What's the fallback plan? How do we go about getting the brand advertising dollars if online brand advertising doesn't work? What can we do that will cause brand advertisers to move their branding dollars out of advertising altogether into some other online channel? How do we disrupt branding?


What is branding, really? Why does it work? For the marketer, branding is a way of wrapping all of a product's attributes in a neat package and giving it a handle so they can refer to it easily. For the consumer, brands are a shortcut: for products where the potential benefit of making a choice is smaller than the cost of choosing, a brand is a fallback. Brands are Scott Adams' whittled down facts, except that marketers have done the whittling for us so they can control the outcome to their advantage. The hole they fit into is your brain. And research shows people only have a few of these holes in their brain.

This can work in a couple of ways. A consumer confronted with a dozen pasta brands in the spaghetti aisle*** would have to expend some time and effort deciding which was best for them. The small potential benefit from finding a better quality pasta is less than the cost in time and effort to determine this, for most consumers. So even if De Cecco is a better pasta, it is a rational for someone familiar with Ronzoni to buy Ronzoni. The pastas are similar enough that someone who just wants a bowl of spaghetti should not expend any effort distinguishing them: just choose your brand and move on.

On the other end of the spectrum, some things are extremely costly to evaluate. Choosing a motorcycle, for instance. The variables that come into play include not just the motorcycle itself, but complementary goods like service quality and availability of third-party components (if the stock pipes are too tame, say.) There are also intangibles, like aesthetics, community, and how you will be perceived among your peers if you are riding a Honda instead of a Harley. A brand can ensure that, even when the objective technical qualities of the bike itself are the same or inferior, it has an advantage among certain consumers because the cost of objectively evaluating differences between bikes is, for most people, impossibly high.

This cost/benefit analysis masks another obvious aspect of branding: risk-mitigation. I have, driving down the highway with my kids, chosen McDonalds though this would be at other times not on the list of possible dinner spots. But I know exactly what I'm getting, how long it will take, and how much it will cost. Because there's a fixed cost of investigating new options, even if for one-time use, the risk/reward curve is not linear.

The answer seems obvious. If branding is a needed compensation for something our brains are just not good at, a low quality way to reduce search costs, an easy alternative to remembering tons of facts, then the answer is to provide a better, more efficient way to sort alternatives. The internet, in its ability to instantly connect you to huge data sources and extremely fast algorithms no matter where you are and what random question you're asking, seems to be the perfect answer.

This is what computers are good at. Lots and lots of data, changing prices, personal utility curves. Right now I might walk into the supermarket looking for a relatively healthy breakfast cereal that my kids will eat. Given the huge number of choices, I might settle on Frosted Cheerios (even though they're probably about as healthy as a glazed donut) because Cheerios has pounded the idea that they are healthy into my brain. I can imagine, instead, walking into the supermarket, scanning the Lucky Charms with my smartphone and asking it to rank healthier alternatives for me. I can imagine a world where I tell an application what I like and dislike about my current pair of sneakers and it recommends a pair that would be better for me. I can imagine a world where I enter the specs and spec tradeoffs for any good imaginable--skis, cars, laundry detergent--and my computer finds me the best match.

But I'm not leading you down a garden path here. I don't do Socratic dialogue. I do not know the answer to the question I'm asking. I do not think these ideas will work because nobody will pay for them.


Generally, and certainly with advertised goods, the seller is the one paying to find buyers and not vice-versa. This has resulted in all sorts of market distortions. Sellers are motivated to sell, and not necessarily only if the product is right for the buyer. That the seller is paying to find buyers--any buyers--instead of the buyer paying to find the perfect seller is a bit of an historic accident. The media was once solely a broadcast mechanism, a mass-reach vehicle. Before the internet, seller-financed advertising was cost efficient while buyer-financed search was cost prohibitive. Even though that is no longer necessarily true, advertising is stuck in a local maximum.

It would certainly be more systemically efficient today if consumers decided what they wanted and then went out and searched for their best match themselves. Then advertising would be less effective so there would be a lot less of it. If sellers did not need to advertise, they could lower the cost of the product and this lower cost would--I'm guessing--more than compensate buyers for the time needed to find the right product. The buyer would end up even on cost (lower product cost ~= higher search costs) but with a more appropriate product. That's the ideal world, but it requires massive behavior change from both sides of the market at once. There's no way to achieve that kind of coordination.

As long as advertising continues to be cost-effective, sellers will advertise. As long as they advertise, they will not lower prices. And as long as they don't lower prices, buyers would have to pay twice if they decide to do the search themselves: once for the advertising and once for the search. This is a long way of saying that no one except the sellers themselves will pay for anything to do with informing consumers about products and services. We are stuck with what we have, efficient or no.

Want to quibble? We now have some seventeen years of evidence that, even if it's a better way of doing things, consumers will not pay for search in any way other than by looking at ads. This is, if you think about it, probably the most bizarre thing about internet marketing. People pay for media that compares and analyzes products, but in a way that undermines the value of these comparisons. Yelp, Google (like Car & Driver magazine offline) critique the very industries that finance them. Their interests are in question. Marketers are attempting to influence people right as they are trying to make uninfluenced choices. Search for a product on Google and you are inundated with ads. Odder, many are clicked. People are searching for someone to convince them, they are going through the motions of choosing and then avoiding making choices. The media soothes the cognitive dissonance with a pleasing veneer of objectivity, but the objectivity is--has to be--a sham. Follow the money.

The non-profit Consumer Reports has overcome this criticism by refusing to accept advertising. But they may be the exception that proves the rule: despite almost certainly being worth the subscription price for anyone who buys even one thing in any category they cover, they only have some seven million monthly subscribers. Consumers will not pay to inform themselves. That's why we're stuck with marketing.


Some other, possibly spurious, correlations to note:
TVPrintOnline Display
Measurability Low Medium High
Involvement Absorbed Absorbing Engaged
Audience Mass Select Targeted
Branding Yes Sure Not so much
CPMs High OK Low
Many intelligent observers, when contemplating low CPMs or recalcitrant brand advertisers say we just need more measurement, more engagement, or more specific targeting. But these things seem to go the wrong way. On the other hand, there are some exceptions: the trade press is more targeted and has higher CPMs; search is more engaging and has higher (effective) CPMs; etc. So the point here is not that we're doomed, but that the easy answers will not do--we're probably analyzing success along the wrong dimensions.


Regardless, I believe in the power of the internet. I think that we can achieve a better product-consumer match by using personalization, community, data and machine learning. In fact, this seems almost too obvious to say. The internet has the power to create a much better branding mechanism: one that works better for brands and for consumers.

I also believe that brands can be valuable. They are proprietary marks, so can guarantee implicit promises and ensure repeat business. They allow trust, and trust is a necessary lubricant for commerce. If there were no newspaper brands, no one would read newspapers, because they would not be able to judge the quality of the news they were reading. If there were no retail brands, every purchase would be like walking into a generic electronics storefront in Times Square: buyer beware.

My objections are not to the internet or to branding. My objections are to the way we are approaching disrupting**** branding. I do not believe the success of online brand advertising is about waiting a bit longer, or measuring better, or creating more engagement. We've waited long enough, we measure better than any other medium, and we are as interactive a medium as they come. If you are espousing those ideas, then you have to also explain why you are right now when you would have been wrong all these long internet years. Things do change, but sudden change is either because of a compounding effect or a catalyst. It does not look like to me that online brand advertising is increasing in an exponential way. Nor do I recognize a catalyst*****.

Or, and I think this is a more promising path, we need to accept that branding may not change to accomodate us, we may have to change to accomodate branding. No more complaining that brand marketers just don't get it. No more waiting on incremental change in measurement or attribution technology. Find a way to allow brands to hone and prove their promises, while giving them a much larger payoff for doing so. Don't think about how to service Procter & Gamble or Coca-Cola--disruption starts off by creating new markets, not servicing old ones--think about how you could help a quality product or service build a brand from the ground up for far, far less than a TV branding campaign would cost. If you do that you will have the big brands' attention, and an amazing business.

* Quoted in Bruce Tremper's Staying Alive in Avalanche Terrain, a must read if you like the backcountry in Winter.
** Seriously? What are you doing here?
*** I was in a supermarket in a hispanic part of Pennsylvania recently where pasta was in an aisle labeled 'Ethnic' while the Goya black beans were in an aisle labelled 'Beans.' Where I live in Hoboken, the exact opposite is true. I wonder if there's a place where the ethnic aisle stocks Oscar Meyer and Easy Cheese.
**** I wrote a typically long blog post on disruption and what it means in this context here.
***** A catalyst has to be a new technology or an entirely new way of utilizing it. The internet itself, say, or social media, or collective intelligence, or online video, or data-driven matching. These could all have been catalysts but, as it turned out, they were not.

Tuesday, February 28, 2012

Selecting, not filtering: Give me a reason to say yes

Raising money for my last startup was humbling, frustrating, time-consuming. But that part was okay: any highly selective process will be humbling, frustrating, and time-consuming. The part that really bothered me wasn't that it took so much time, but that so much of that time was a complete and total waste. In almost all of the VC meetings, we did not leave with a check. But in some 80% of the meetings we also did not leave with any insight as to why not*.

The best VCs listened to us and then gave us some insight into their thinking. Fred Wilson and Brad Burnham actually said no to us and then worked through their thinking about what we were doing in great and helpful detail. Josh Kopelman and Howard Morgan told us it wouldn't work, told us exactly why, then invested, and--after that--introduced us to people who helped us fix the flaws in the plan. But many others gave us either no response at all ("we'll get back to you") or generic non-responses ("we'd like to see more traction.")

We had a pretty firm idea of the problem we wanted to solve, but we were somewhat flexible about how we would solve it. We used the feedback from the money-raising process to hone our ideas. When we got no feedback, we felt we had given the VCs critical market intelligence and gotten nothing in return.

One of my ideals when I started investing was to always provide feedback when I said no. But here I am four years in, going through the pitches that piled up last week while I was on vacation. I'm finding it hard to live up to that ideal. I'm saying no to companies that I don't have a concrete reason to say no to. After a bit of introspection, I think that finding a reason to say no is not really how I make the hard decisions.

I have a tangible reason to say no to some 85% of the pitches I see, and I say yes to less than 2% (some of these I don't end up doing because we can't agree on a deal.) Here's a swag at how my dealflow works out:

  • 40%: No; I do not know your market well enough to help you succeed (also known as I do not know your market well enough to make a good decision about investing);
  • 20%: No; I do not think your idea will work and I can't see where else you will be able to put the technology you're building to work/you are completely inflexible about entertaining other potential markets for your technology/you are too flexible about where you will put your technology to work (the "we're a platform!" syndrome);
  • 10%: No; You are creating something merely better, not different;
  • 5%: No; You have the wrong team/your team does not seem to gel/you do not seem to think you need a team at all/you are coding in .NET;
  • 5%: No; Other explainable reasons;
  • 5%: No; Bad**;
  • 13%: Meh;
  • 2%: Like.
I always explain, in as much detail as the entrepreneur wants, my thinking behind the 85% where I can say no. And I am always happy to explain why I like the 2% I like.

The rub is in the penultimate 13%. These are companies that I don't have a real reason to say no to, companies where I analytically think they have a venture-capital-winner expected value but where I just can't get excited about them. The reality is that with these companies--and, in fact, with all companies--I am not looking for a reason to say no; I am looking for a reason to say yes. With the 85%, there is a glaring reason why I can't say yes. With the 13%, I just can't get the word to come out of my mouth.

For the companies I can easily say no to, some dimension of their plan (team, market, vision, product, customer, etc.) does not rise above my threshold of yes. For the 15%, all aspects do. Analytically the fitness function then necessarily also rises above my threshold.

The difference between the 'meh' and the 'like' is that the 'meh' companies are good enough in all aspects but not great in any of them. The 'like' companies are the ones where they really excel in at least a couple of ways: a great team, a big market, a compelling vision. I try to select not just for how good a company is, but how good it will be. It's easy to improve along one dimension, it's possible to improve along a couple of dimensions, but it's almost impossible to improve along all dimensions. The companies that are just good enough in all dimensions need to improve in all dimensions. The companies that are great in a few just need to improve in a few others, not all, to be great overall.

In fact, some of my favorite companies are the ones that may not even rise above the threshold in one or two dimensions but make up for it by having a superstar team or a gigantic market or a world-beating vision. These are the companies that have a shot at being legendary.

I don't know what to say to 'meh' companies after they pitch me. It's hard for you to recover from a "we're not so bad" pitch. But if you're dreaming up your startup right now my recommendation is to be good at everything, but to be insanely great at something. That's what gets me excited.

* And, I should note, the founding team knew the venture market inside and out. We had done our research on which firms to approach based on what they were interested in, which partners to approach, had pre-sold the idea before the physical meeting, had customized the deck to highlight the aspects that particular firm/partner could grab onto most quickly, etc. Highly suggested in any case.

** My dealflow right now is pretty highly curated so I don't get a lot of pitches that are just, well, bad. Not to be judgemental. Bad, to me, is a founder who simply does not know what they're doing: a non-coder trying to enter a market either (i) that they just don't know anything about--generally where they've had a bad customer experience but have not done the research to understand the institutional framework behind the root cause, (ii) where there are great companies already doing exactly what they want to do and they've never heard of them, or (iii) that is so small that even revolutionizing it will create almost no societal value. Or, they could give a damn about creating societal value, they just want to make some money quick.

Tuesday, January 17, 2012

VC/Company Investment Visualizer

A friend asked me last week if I knew a tool to help him visualize which VCs were investing in a sector. I did not. But I realized I could pretty quickly repurpose the VC Bar Chart code and some unpublished code that pulls in data from a Google spreadsheet to show a force-directed graph. So, weekend project.

Data from Crunchbase, visualizaton using the d3.js library.

Here's my portfolio.

The site is here. Just start typing company names in the upper-left hand corner box and hit plus to add. Real name to Crunchbase permalink translation uses the list of companies as of Friday* or so, so if the company was added to CB later, autocomplete finds nothing;  just type in the permalink and the company will still be added. In the screenshot above a few of my companies had no CB investor entries, so they're just floating out there. Many of my other companies are not linked to me because CB does not mention me as an investor.

One way to explore is to enter a bunch of companies in your area of interest and see how the graph falls out.  Here's one of the AdTech industry.

The save functionality is experimental (to me, that is.) It uses HTML5 localStorage. The caveat is that you can't email visualizations around that way, and there may be times when your browser clears localStorage (sometimes when clearing cookies, for example.) If it does, you lose all saved visualizations.

The code is all client-side, so it's right there in your browser if you want to look at it. I found myself late last night using a non-analytical debugging process** when I was trying to get the 'load visualization' piece to work. I'll put it up on GitHub some time after I clean it up.

* And I redacted the list to only include companies that CB showed having investors. The full list was too big to load efficiently.
** Mainly making random code deletions.