nPost Blog

Timing your startup

I never had the opportunity to invest in YouTube but I have to admit that if I did I probably would have passed (which of course would have been a huge mistake). I’d been around the web long enough to remember the dozens of companies before YouTube that tried to create crowdsourced video sites and failed. Based on “pattern recognition” (a dangerous thing to rely on), I was deeply skeptical of the space.

What I failed to appreciate was that the prior crowdsourced video sites were ahead of their time. YouTube built a great product, but, more importantly, got the market timing just right. By 2005, all the pieces were in place to enable crowdsourced video – the proliferation of home broadband, digital camcorders, a version of Flash where videos “just worked,” copyrighted web content that could be exported to YouTube, and blogs that wanted to embed videos.

Almost anything you build on the web has already been tried in one form or another. This should not deter you. Antecedents existed for Google, Facebook, Groupon, and almost every other tech startup that has succeeded since the dot-com bubble.

Entrepreneurs should always ask themselves “why will I succeed where others failed?” If the answer is simply “I’m doing it right” or “I’m smarter,” you are probably underestimating your antecedents, which were probably run by competent or even great entrepreneurs who did everything possible to succeed. Instead your answer should include an explanation about why the timing is right – about some fundamental changes in the world that enable the idea you are pursuing to finally succeed. If the necessary conditions were in place, say, a year ago, that might still be ok – YouTube happened to nail their product out of the gate, but if they hadn’t a company started later might have succeeded in their place.

Often the necessary conditions are only beginning to emerge and knowing when they will do so sufficiently is very hard to predict. We all know the internet will become fully social, personalized, mobile, location-based, interactive, etc. and lots of new, successful startups will be built as a result. What is very hard to know is when these things will happen at scale.

One way to mitigate timing risk is to manage your cash accordingly. If you are trying to ride existing trends you should ramp up aggressively. If you are betting on emerging trends it is better to keep your burn low and runway long.  This takes discipline and patience but is also the way you hit it really big.

Instrumenting the offline world

In the last decade there have been major advances in storing, analyzing, and acting upon extremely large data sets.  Data sets that were previously left dormant are now being put to (mostly) constructive use. But the vast majority of information in the world isn’t available for analysis because it isn’t being electronically collected.

This is changing rapidly as new data collection mechanisms are implemented – what engineers refer to as instrumentation. Common examples of instrumentation include thermometers, public safety cameras, and heart rate monitors.

Smart phones are one obvious new source of potential instrumentation.  A person’s location, activities, audio and visual environment – and probably many more things that haven’t been thought of yet – can now be monitored.  This of course raises privacy issues.  Hopefully these privacy issues will be solved by requiring explicit user opt-in.  If so, this will require creating incentives for people to do so.

Foursquare instruments location in an opt-in way through the check in. The incentives are social and game-like, but the data produced could be useful for many more “serious” purposes.  Fitbit instruments a person’s health-related activity. The immediate incentive is to measure and improve your own health, but the aggregate data could be analyzed by medical researchers to benefit others.

In manufacturing, there has been a lot of interesting innovation around monitoring machinery, for example by using loosely joined, inexpensive mesh networks.  In homes, protocols like ZigBee allow devices to communicate which allows, for example, automation of tedious tasks and improved energy efficiency.

In the next decade, there will be a massive amount of innovation and opportunity around the big data stack. Instrumentation will be the foundational layer of that stack.

You need to use social services to understand them

I don’t know if Malcolm Gladwell is right when he claims “the revolution will not be tweeted,” but I can say with certainty that the Twitter he describes is not the Twitter I know. Gladwell’s central argument is that Twitter creates weak ties but social movements require strong ties. I’ve made more strong ties through Twitter (and blogging) than I have through any communications medium I’ve ever used before. The relationships start off weak – a retweet, @ reply, or blog comment – but often strengthen through further discussions and eventually become new friendships and business relationships.

I can see why Gladwell gets this wrong – he doesn’t seem to really use Twitter (he does blog occasionally). I barely tweeted or blogged for a long time too. I read blogs basically since their advent, but social services are fundamentally participatory: reading blogs/tweets is to social services as watching TV is to a real life conversations. I finally relented at the insistence of Caterina, who had the foresight to insist that everyone at Hunch blog, tweet, contribute to open source projects, etc. I now get some of my best ideas from responses to tweets and blog posts, and have developed dozens of strong relationships through the experience.

I made some jokes on Twitter the past few days about Kleiner Perkins’ new social fund.  These were meant to be lighthearted: I only know one person at KP and from everything I’ve seen they seem to be smart, friendly people. But underneath the jokes lies a real issue: the partners there don’t seem to really participate in social services (something they only underscored by announcing their new fund at a press conference that targeted traditional media outlets).

I’d love to engage in a debate with smart people like Gladwell about the impact of the social web on culture, politics, activism and so on. I also think it’s great to see savvy investors like KP allocate significant resources to the next wave of social web innovation. But it’s hard for me to take them seriously when they don’t seem to take their subject matter seriously.

Online privacy: what’s at stake

It is widely believed that a flourishing democracy requires an independent, diverse, and financially solvent press.  With print newspapers set to disappear in the next few years, the future of quality journalism is highly uncertain. This year, the online version of the New York Times will generate about $200M in revenue, a number that will need to approximately triple to support the current Times newsroom.

Most people who understand Internet economics believe that the best hope for online journalism is online advertising. Luckily, online advertising has significant room for improvement. Most of the revenue of the Times’ online business is generated through display ads. The main metric used to price display ads is derived from the rate at which users click on the ads, a rate which today is dismally low.  Thus the Times could continue to support its current newsroom staff if display ads became even moderately effective.

Lots of smart people are working on improving the efficacy of display advertising. Large companies like Google and Microsoft are investing billions in the problem. As usual, though, the best ideas are coming from startups. Companies like Blue Kai and Magnetic are bringing search intent (particularly purchasing intent – the core of Google’s profits) to display ads.  Companies like Media6Degrees are using social relations to target ads based on the principle that “birds of a feather flock together” (Facebook will likely start doing this soon as well).  Solve Media turns the hassle of registration into an engaging marketing event.  Convertro is working on properly attributing online purchases “up the funnel” from sites that harvest intent (search, coupon sites) to sites that generate intent (media, commerce guides). All told, there are a few hundred well-funded ad tech startups developing clever methods to improve display advertising.

Many of these targeting technologies rely on gathering information about users, something that inevitably raises concerns about privacy. Until recently, online privacy depended mostly on anonymity. There is a big difference between advertisers knowing, say, users’ sexual preferences and knowing users’ sexual preferences plus personally identifiable information like their names.  Like most people, I don’t mind if it’s easy to find my real name along with my job history, but I do mind if it’s easy to discover other personal details about me. When I’m not anonymous (e.g. on Facebook) I want to control what is disclosed – to have some privacy – but when I’m anonymous I’m far less concerned about information gathered for marketing purposes.

Before the rise of social networks, online ad targeting services (mostly) tracked people anonymously, through cookies that weren’t linked to personally identifiable information.  Social networks have provided the means to de-anonymize information that was previously anonymous. Apparently, the wall has been breached between 1) my real identity plus my self-moderated public information, and 2) my anonymous, non-self-moderated private information.

The good news is that the things users want to keep secret are almost always the least important things to online advertisers. It turns out that knowing people are trying to buy new washing machines or plane tickets to Hawaii is vastly more monetizeable than their names, who they were dating, or the dumb things they did in college. Thus, there are probably a set of policies that allow ad targeting to succeed while also letting users control what is associated with their real identities.  Hopefully, we can have an informed and nuanced debate about what these policies might be. The stakes are high.

Note:  As with almost everything I write on this blog, I have a ton of conflicts of interest.  Among them: I’m an investor, directly or indirectly, in a bunch of technology startups.  Some of these – including some companies mentioned above – are trying to create new advertising technologies. I am currently the CEO & Cofounder of Hunch, which among other things is trying to personalize the internet through an explicit user opt-in mechanism.

The “ladies’ night” strategy

Many singles bars have “ladies’ night” where women are offered price discounts. Singles bars do this for women but not for men because (heterosexually-focused) bars are what economists call two-sided markets – platforms that have two distinct user groups and that get more valuable to each group the more the other group joins the platform – and women are apparently harder to attract to singles bars than men.

Businesses that target two-sided markets are extremely hard to build but also extremely hard to compete against once they reach scale. Tech businesses that have created successful two-sided markets include Ebay (sellers and buyers), Google (advertisers and publishers), Paypal (buyers and merchants), and Microsoft (Windows users and developers). In some cases individuals/institutions are consistently on one side (buyers and merchants) while in other cases they fluctuate between sides (Ebay sellers are also often buyers).

In almost every two-sided market, one side is harder to acquire than the other. The most common way to attract the hard side is the ladies’ night strategy: reduce prices for the hard side, even to zero (e.g. Adobe Flash & PDF for end-users), or below zero (e.g. party promotors paying celebrities to attend). Rarer ways to attract the hard side is 1) getting them to invest the platform itself (e.g. Visa & Mastercard), and 2) interoperating with existing hard sides (e.g. Playstation 3 running Playstation 2 games).

If you are starting a company that targets a two-sided market you need to figure out which side is the hard side and then focus your efforts on marketing to that side. Generally, the more asymmetric your market the better, as it allows you to market to each side more in serial than in parallel.

The segmentation of the venture industry

Ford Motors dominated the auto market in the early 20th century with a single car model, the Model T.  At the time, customers were seeking low-cost, functional cars, and were satisfied by an extremely standardized product (Ford famously quipped that “customers can choose it in any color, as long as it’s black”). But as technology improved and serious competitors emerged, customers began wanting cars that were tailored to their specific needs and desires. The basis of competition shifted from price and basic functionality to ”style, power, and prestige“. General Motors surpassed Ford by capitalizing on this desire for segmentation. They created Cadillacs for wealthy older folks, Pontiacs for hipsters, and so on.

Today, the venture financing industry is going through a similar segmentation process. Venture capital has only existed in its modern form for about 35 years.  In the early days there were relatively few VCs. Entrepreneurs were happy simply getting money and general business guidance.  Today, there is a surplus of venture capital and entrepreneurs have become increasingly savvy “shoppers.”  As a result, competition amongst venture financiers has increased and their “customers” (entrepreneurs) have flocked to more specialized “products.”

Some of this segmentation has been by industry (IT, cleantech, health care) and subindustry (iPhone apps, financial tech, etc). But more pronounced, especially lately, has been the segmentation by company stage.  Today at least four distinct types of venture financing “products” have become popular.

1) Mentorship programs like Y Combinator help startups ideate, form founding teams, and build initial products. I suspect many of the companies they hatch wouldn’t exist at all (and certainly wouldn’t be as savvy) if it weren’t for these programs.

2) So-called super angels provide capital and guidance to a) hire non-founder employees, b) further product development c) market the initial product (usually to early adopters), and d) raise follow on VC funding. Often current or former entrepreneurs themselves, super angels have gone through this stage many times as founders and angel investors.

3) Traditional VCs (Sequoia, Kleiner, etc) help companies scale and get to profitability. They often have broad networks to help with hiring, sales, bizdev and other scaling functions. They are also experts at selling companies and raising follow-on financing.

4) Accelerator funds (most prominent recently is DST) focus on providing partial liquidity and preparing the company for an IPO or big M&A exit.

In the past, traditional VC’s played all of of these roles (hence they called themselves “lifecycle” investors). They incubated companies, provided smalls seed financings, and in some cases provided later stage liquidity. But mostly the mentorship and angel investing roles were played by entrepreneurs who had expertise but shallow pockets and limited time and infrastructure.

What we are witnessing now is a the VC industry segmenting as it matures. Mentorship and angel funding are performed more effectively by specialized firms.  Entrepreneurs seem to realize this and prefer these specialized “products.”  There is a lot of angst and controversy on tech blogs that tends to focus on individual players and events. But this is just a (sometimes salacious) byproduct of the larger trends. The segmentation of the venture industry is healthy for startups and innovation at large, even if at the moment it might be uncomfortable and confusing for some of the people involved.

If you aren’t getting rejected on a daily basis, your goals aren’t ambitious enough

My most useful career experience was about eight years ago when I was trying to break into the world of VC-backed startups. I applied to hundreds of jobs:  low-level VC roles, startups jobs, even to big tech companies.  I got rejected from every single one.  Big companies rejected me outright or gave me a courtesy interview before rejecting me. VCs told me they wanted someone with VC experience.  Startups at the time were laying people off.  The economy was bad (particularly where I was looking – consumer internet) and I had a strange resume (computer programmer, small bootstrapped startups, undergrad and masters studying Philosophy/mathematical logic).

The reason this period was so useful was that it helped me develop a really thick skin.  I came to realize that employers weren’t really rejecting me as a person or on my potential – they were rejecting a resume.  As it became depersonalized, I became bolder in my tactics. I eventually landed a job at Bessemer (thanks to their willingness to take chances and look beyond resumes), which led to getting my first VC-backed startup funded, and things got better from there.

One of the great things about looking for a job is that your “payoff” is almost always a max function (the best of all attempts), not an average. This is also generally true for raising VC financing, doing bizdev partnerships, hiring programmers, finding good advisors/mentors, even blogging and marketing.  I probably got rejected by someone once a day last week alone. In one case a friend who tried to help called me to console me. He seemed surprised when I told him: “no worries – this is a daily occurrence – we’ll just keep trying.”  If you aren’t getting rejected on a daily basis, your goals aren’t ambitious enough.

Howard Lindzon’s “Web is Dead” series

Howard’s Stocktwits interviews are always really fun.  Some people don’t get his subtly self-deprecating sense of humor but I love it. Besides discussing the usual suspects (Facebook, Twitter, Apple), we spend some time trashing Wall Street and chatting about some early-stage startups including Founder Collective investments Bnter, Giiv, Ze Frank Games, and Canvas (founder of 4chan Moot’s new startup).  Of course I also shamelessly promote Hunch.

Also, Fred Wilson’s interview with Howard is a must watch.

Web services should be both federated and extensible

One of the most important developments of the web 2.0 era is the proliferation of full featured, bidirectional APIs.  APIs provide a way to “federate” web services from a single website to a distributed network of 3rd party sites. Another important web 2.0 development is the proliferation of web Apps (e.g. Facebook Apps). Apps provide a way to make websites “extensible.”

The next step in this evolution is to create web services that are both federated (APIs) and extensible (Apps).

In my ideal world, the social graph would not be controlled by a private company. That said, Facebook, to its credit, has aggressively promoted a fairly open API through Facebook Connect. Facebook has also been a leader in promoting Apps. For Facebook, creating extensible, federated services would mean providing a framework for Facebook Connect Apps – apps that extend Facebook functionality but reside on non-Facebook.com websites.

Consider the following scenario.  Imagine that in the future a geolocation data/algorithm provider like SimpleGeo takes Facebook Places check-in data and, using algorithms and non-Facebook data, produces new data sets, for example: map directions, venue recommendations, and location-based coupons. The combination of Facebook’s data (social graph and check-ins) and SimpleGeo data/algorithms would create much more advanced feature possibilities than either service acting alone.

With today’s APIs, if, say, Gowalla wanted to integrate Facebook plus SimpleGeo into their app*, they would basically have 3 choices:

1) Embed Facebook widgets in Gowalla.  These are simple iframes (effectively separate little websites) that don’t interact with SimpleGeo.  Gowalla would just have to sit and wait and hope that Facebook decided to bake in SimpleGeo-like functionality.

2) Pre-import SimpleGeo data. This significantly limits the size and dynamism of the SimpleGeo data sets and doesn’t incorporate SimpleGeo algorithms, thus severely limiting functionality.

3) Host an instance of SimpleGeo’s servers internally.  This requires heavy technical integration, undermining the main benefit of APIs.

In a world of extensible APIs (or “API Apps”), Gowalla could instead send Facebook data back to SimpleGeo.  The data flow would look something like this:

(Note how there are three parties involved – @peretti calls this a “data threesome”). This configuration is much simpler to integrate – and potentially much more powerful and dynamic – than the other configurations listed above.  You could implement this today, but it would create user experience challenges.  For example, Gowalla would be sending Facebook data to a 3rd party (step 3), which might (depending on the data sent) require explicit user opt-in. Things become more onerous if SimpleGeo wanted to share its own user data with Gowalla. That would require an additional oAuth to SimpleGeo (authorizing step 4).

Allowing websites to be federated and extensible will open up a whole new wave of innovation.  Ideally some spec like oAuth could include the multiple authorizations in a single authorization screen.  Facebook could also do this by allowing 3rd parties to be part of the Facebook Connect authorization process.  Inasmuch as Facebook’s seems to be trying to embed their social graph as deeply as possible into the core experiences of other websites, allowing extensible APIs would seem to be a smart move.

* I have no connection to any of these companies (Facebook, Gowalla, SimpleGeo) and have no knowledge of their product plans beyond their public websites.  I am imagining functionality that Gowalla and SimpleGeo might include someday but for all I know they have no interest in these features – I just picked them somewhat arbitrarily as examples.

Converts versus equity deals

There has been a debate going on the past few days over whether seed deals should be funded using equity or convertible notes (converts). Paul Graham kicked it off by noting that all the financings in the recent YC batch were converts. Prominent investors including Mark Suster and Seth Levine weighed in (I highly recommend reading their posts). While this debate might sound technical, at its core it is really about a difference in seed investing philosophy.

I am a proponent of convertibles, but only with a cap (I’ve written about the problems of convertibles without caps before and never invest in them).  I believe that pretty much every other seed investor who advocates converts also assumes they have a cap.  So any discussion of convertibles without caps seems to me a red herring.

There are two kind of rights that investors get when they put money in company.  The first are economic rights: basically that they make money when the investment is successful.  The second are control rights: board seats, the ability to block financings and acquisitions, the ability to change management, etc.  Converts give investors economics rights with basically zero control rights (legally it is just a loan with some special conversion provisions). Equity financings normally give investors explicit rights (most equity terms sheets specify board seats, specific blocking conditions, etc) in addition to standard shareholder rights under whatever state the company incorporated in (usually CA or Delaware).

To the extent that I know anything about seed investing, I learned it from Ron Conway.  I remember one deal he showed me where the entire deal was done on a one page fax (not the term sheet – the entire deal).  Having learned about venture investing as a junior employee at a VC firm I was shocked. I asked him “what if X or Y happens and the entrepreneur screws you.”  Ron said something like “then I lose my money and never do business with that person again.”  It turned out he did very well on that company and has funded that entrepreneur repeatedly with great success.

You can hire lawyers to try to cover every situation where founders or follow on investors try to screw you. But the reality is that if the founders want to screw you, you made a bet on bad people and will probably lose your money. You think legal documents will protect you? Imagine investors getting into a lawsuit with a two person early-stage team, or trying to fire and swap out the founders – the very thing they bet on.  And follow on investors (normally VCs) have a variety of ways to screw seed investors if they want to, whether the seed deal was a convert of equity.  So as a seed investor all you can really do is get economic rights and then make sure you pick good founders and VCs.

Seed investing is a people business.  Good entrepreneurs understand this.  Ron was an investor in my last two companies and never had any control rights but had massive sway because he worked so hard to help us and gave such sage advice.  And most importantly, he carried great moral authority. We always knew he was speaking from deep experience and looking out for the company’s best interests – sometimes against his own economic interests.

Like it or not, the seed investment world runs on trust and reputation – not legal documents.

hosting