By 2008, SpaceX had launched three rockets. They all failed to make it into orbit. Shortly after the third failure, Elon Musk was interviewed by Wired Magazine's Carl Hoffman:
Wired.com: At the end of the day you're still zero for three; you have so far failed to put a rocket into orbit.
Musk: We haven't gotten into orbit, true, but we've made considerable progress. If it's an all-or-nothing proposition then we've failed. But it's not all or nothing. We must get to orbit eventually, and we will. It might take us one, two or three more tries, but we will. We will make it work.
Wired.com: How do you maintain your optimism?
Musk: Do I sound optimistic?
Wired.com: Yeah, you always do.
Musk: Optimism, pessimism, fuck that; we're going to make it happen. As God is my bloody witness, I'm hell-bent on making it work.
In early 1997, an ambitious healthcare startup called Dr. Koop, which was founded by former US Surgeon General C. Everett Koop, went online. It was during the era when internet portals reigned, and the company spent its first year struggling to gain exposure. By the end of 1997, with no traction and zero revenue, it was running out of money.
On April 6th, 1998, the company closed a respectable $1M Series A financing. It wasn't enough, apparently, because just 22 days later, it closed another $6.8M Series B round. By the end of 1998, Dr. Koop had taken about $8 million in financing, made just $43,000 in revenue, and suffered a net loss of $8.997 million.
Nine months after the series B money was in the bank, and in desperate need of more cash, the company raised another $3.5M. It was January of 1999, at the height of one of the largest valuation bubbles in history, and Dr. Koop was running on fumes: with a burn rate of almost $1M per month, the company had less than four months of runway until it would be unable to meet its payroll obligations.
There was only one option. On March 5th, 1999, with $43,000 in revenue and 80,000 registered users, Dr. Koop filed its S-1 with the SEC to go public.
What Dr. Koop did, exactly, is not entirely clear. It appears to have been some kind of healthcare-specific web portal. The S-1 describes it like this:
Our website, www.drkoop.com, is a healthcare portal which integrates dynamic healthcare content on a wide variety of subjects, interactive communities and tools, as well as opportunities to purchase healthcare-related products and services on-line.
Fascinating how little it says with so many words. It continues:
Our company's founders, including former U.S. Surgeon General Dr. C. Everett Koop, created drkoop.com to empower consumers to better manage their personal health with comprehensive, relevant and timely information. Our objective is to establish the drkoop.com network as the most trusted source of consumer healthcare information and services on the Internet.
Again, nothing. But I suppose it doesn't really matter what it did.
On June 8th, 1999, Dr. Koop went public. The details[1]:
$9.00 Per share 27,514,591 Shares outstanding post-offering $247,631,319Initial market capitalization $5,906,250Fees paid to underwriters $2,332,260Fee paid to now-defunct Bear Sterns 5,759Revenue multiple (using revenue from 1998)
Its value popped up 83% on the first day–a “mildly positive review” from investors, as CNET eloquently reported–and it raised about $90 million. Now flush with an enormous stockpile of cash, Koop's top priority was to gain attention and drive traffic to the website.
Driving traffic back then wasn't very straightforward. In 1999, the portal sites controlled the internet. If you wanted traffic, you had to get featured or listed by one of the portals. Dr. Koop did what the other bubble companies did, and it entered into an agreement with AOL to get prominent positioning. How much did it pay?
$89 million per year.
Despite being prominently placed on AOL's portal, which drove an estimated 2.6 million visitors per month, Dr. Koop lost $82.5 million dollars in 1999. A few months later, in May of 2000, and with dwindling cash reserves, the company was forced to lay off 35% of its 185 employees. Its stock had fallen from a 52-week high of $45.75 to just $2.34 on May 18, 2000. In August, slightly over one year after its IPO, Dr. Koop was delisted from the Nasdaq.
For the next year or so, through 2001, the company experienced a series of setbacks. It ran out of money in August 2001, and was saved in the last hour–literally, three hours after it ran out of money–by a $20 million private equity infusion. It did nothing but prolong the inevitable.
On December 17th, 2001, embroiled in legal battles and sparking questions about healthcare privacy, Dr. Koop closed its doors.
The S-1
Thanks to Dalton Caldwell for bringing Dr. Koop's hilarious S-1 to my attention (he says he brings it up when people talk about the current crop of tech IPOs). Here are a couple of my favorite excerpts:
On the platform:
The drkoop.com web platform consists of readily available, off-the-shelf, computer systems, including dual Intel Pentium servers in a fully redundant configuration. The drkoop.com web platform was designed using a proprietary architecture deploying primarily Microsoft technology running the Windows/NT Operating System. Other Microsoft web enabling technologies used in the drkoop.com web platform include the Microsoft Membership and Personalization Server (software that captures user data and enables the drkoop.com experience to be customized for each user), Microsoft SQL Server (database software used to store user data and content) and Microsoft Internet Information Server (software which enables pages to be displayed to the user).
The affiliates section, which suggests that Dr. Koop paid other sites to embed their content:
drkoop.com has entered into an agreement with Infoseek under which The Go Network will distribute health-related content to their users. Through this agreement, users on The Go Network will be able to access various health information, services, interactive tools and related commerce opportunities through direct links to the drkoop.com website. We believe that this partnership will contribute substantially to our brand awareness and traffic generation efforts. This agreement with Infoseek has a one-year term. We will pay Infoseek a fee for running a minimum number of content segments on its Health Center homepage.
On Dr. Koop's “Unique Features and Tools”:
Our website is designed to provide easy access to innovative features and tools. Currently, our most popular tool educates consumers on the interaction among various drugs and other substances. In addition, we recently acquired the right to deploy a comprehensive personal medical record which will allow users to establish and maintain a lifelong record of their health and medical information in a secure portion of our database. We intend to continue to add useful tools to enable our users to personalize their on-line experience. We believe that our tools and features will continue to encourage users to visit our website frequently and increase the likelihood of users selecting drkoop.com as their preferred website for health-related issues.
On the benefits as an ad sales platform:
We believe our ability to target specific users, the interactive nature of our website and the demographic characteristics of our users will be attractive to pharmaceutical, healthcare and other companies that advertise on the Internet. By identifying users interested in a particular health-related topic or who desire to address a particular health condition, we believe we can sell advertising in a highly targeted manner, thereby commanding higher advertising rates.
Under “Risk Factors”:
We were incorporated in July 1997 and launched our Internet operations in July 1998. Accordingly, we have an extremely limited operating history. An investor in our common stock must consider the risks, uncertainties, expenses and difficulties frequently encountered by companies in their early stages of development, particularly companies in new and rapidly evolving markets, including the Internet market. These risks and difficulties include our ability to:
attract a larger audience of users to our Internet-based consumer healthcare network;
increase awareness of our brand;
strengthen user loyalty;
offer compelling on-line content, services and e-commerce opportunities;
maintain our current, and develop new, affiliate relationships;
attract a large number of advertisers;
respond effectively to competitive pressures;
continue to develop and upgrade our technology; and
attract, retain and motivate qualified personnel.
If only they had been able to “offer compelling on-line content, services, and e-commerce opportunities,” maybe they wouldn't have failed.
Yahoo has just announced Axis, a browser extension thing and mobile app that “redefines what it means to search and browse the Web [sic].”
• A group of people at Yahoo, including engineers, designers, and product managers had to conceive of, design, and build this product, which works basically identically to browser toolbars from the early 2000's. It does have a sync feature, but it requires that you use a new custom, dedicated browser on your mobile device.
• A group of people at Yahoo had to make the marketing website, which describes the product ambiguously and does not actually contain any real screenshots or information.
• A group of people had to write, design, direct, and edit the advertisement on that marketing website. It's an advertisement of a man punching 15-foot-tall transparent glass websites, followed by a giant pepper tree growing out of cement in an empty warehouse.
• A group of people had to design and approve the logo for Axis, which looks identical to Adobe's logo.
And none of those people said, “Hold on, this is a hideous piece of shit. No. We can't launch this, or Yahoo will look like a bunch of amateurs.”
Does no one at Yahoo have the power to say that? Really?
To be fair, there are some great ideas in Axis, and some of those ideas will almost certainly make up parts of browsers in the near future. But Axis is not a product that Yahoo should be building, and it's not finished. It was executed very poorly. Look at the Terms & Conditions. It's embarrassing. Someone should have said, “No.”
This morning, I received a LinkedIn message from a man named James Holm, who identified himself as a corporate recruiter with Weyland Industries. Weyland is the corporation that funded the the Prometheus project in the upcoming Ridley Scott movie of the same name.
This message was clearly tailored specifically for me, and I think it's pretty awesome. But it does raise some interesting questions about the divide between so-called viral marketing and reality. Mr. Holm has a complete profile on LinkedIn which documents his career. But he does not exist.
Because these kinds of campaigns have to exploit existing reality, marketers have to find places where the benefit of advertising revenue for a distribution, content, or platform company (like LinkedIn) outweighs the cost to that company of sacrificing user trust in their experience. In this case, LinkedIn was probably paid a bundle of money to allow Prometheus' marketers to create fake profiles and send fake messages to users.
The result is that LinkedIn sold my reality to Fox so they could exploit it for the few seconds before I realized the message was fake. But it's that moment of realization that makes this sort of thing enjoyable. It ends up being net positive for everyone. LinkedIn gets money, Prometheus gets more exposure, and I get a satisfying moment of realization.
Hilariously, in this instance, the marketers became suckers in their own game, and chose a poor tertiary partner that basically ruined the experience for me. Microsoft's IE 9 advertising is horrendous, and I can't imagine that it's worth the enormous expense I'm sure they had to pay Fox for such a blatant and absurd inclusion in the LinkedIn message.
See also: The Game, which is based on similar themes
“It’s almost a mathematical impossibility to imagine that, out of the thousands of things for sale on a given day, the most attractively priced is the one being sold by a knowledgeable seller (company insiders) to a less-knowledgeable buyer (investors).”
In a blog post today announcing Twitter's new tailored suggestions system is something that has left me shocked: an overt admission by Twitter that it is transparently tracking your movements around the web. Othman Laraki, on the Twitter blog:
These tailored suggestions are based on accounts followed by other Twitter users and visits to websites in the Twitter ecosystem. We receive visit information when sites have integrated Twitter buttons or widgets, similar to what many other web companies — including LinkedIn, Facebook and YouTube — do when they’re integrated into websites. By recognizing which accounts are frequently followed by people who visit popular sites, we can recommend those accounts to others who have visited those sites within the last ten days.
Basically, every time you visit a site that has a follow button, a “tweet this” button, or a hovercard, Twitter is recording your behavior. It is transparently watching your movements and storing them somewhere for later use. Right now, that data will make better suggestions for accounts you might want to follow. But what other things can it be used for? The privacy implications of such behavior by a company so large are sweeping and absolute.
If tracking your behavior transparently is acceptable in the pursuit of a better user experience, why isn't it also acceptable in the pursuit of monetization? Is it okay for Twitter to sell your web browsing history to advertisers? The company is playing with a very slippery slope.
I'm not particularly surprised that Twitter is doing this kind of data analysis or collection. Facebook is almost certainly doing the same thing. But it is wrong. People do not expect Twitter or Facebook to know about their movements on the web. But they do. And that information is being stored somewhere. It is a violation of privacy and trust.
I'm amazed that Twitter is overtly admitting to this behavior without considering the privacy implications. How many people have access to the data Twitter is collecting? Can any Twitter employee who has production database access look at Mitt Romney's browsing history? Can they look at your browsing history?
These kinds of questions are extremely important, but there are no answers.
Update: It is true that Twitter has committed to following the “Do Not Track” flag, which “asks” websites to not track your behavior. I think that is a distraction from the real issue; tracking should not be opt-out. It should be opt-in. At least until such behavior by companies is commonly understood. I have no problem with Twitter or Facebook tracking me, as long as I know about it and as long as normal people who use those services know about it.
Also, Twitter's communications team emailed me with some specific information in response to my questions. From the email:
Is it okay for Twitter to sell your web browsing history to advertisers?
As we state in our Privacy Policy, this is not something we do, nor would we do. That’s counter to how we treat our users and their data. Here's a link to the Privacy Policy: https://twitter.com/privacy.
Further, regarding browsing history:
You also mention “browsing history” several times in your post. On that point, to protect your privacy, we do not maintain browsing history. We start the process of deleting your visits to pages in the Twitter ecosystem after a maximum of 10 days. We only keep tailored suggestions for you, as explained in our privacy policy.
Just days before Facebook’s historic stock offering, General Motors said it plans to stop advertising on the social media site, concluding that its paid ads don’t have a big impact on consumers. […]
The news comes at an awkward time for Facebook, whose $105 billion IPO is scheduled for Friday.
“Awkward?” No. For someone, strategic. (Or, maybe, vengeant).
You can call Steve Ballmer many things, but you cannot call him the “the worst CEO of a large publicly traded American company today,” as Forbes' Adam Hartung did in a recent article. It's easy to see Microsoft as a bumbling fool of the tech world, but when you look closely at its business, the company's core competencies, and Ballmer's decisions, a coherent picture begins to form. It's a picture of a company being run from a very rational and respectable set of philosophies.
While Microsoft is no disruptive force in tech today, the truth is that it has never been. The company's core competency is a process it uses for entering and consuming existing industries. After it enters a market, it rides off the innovations of its competitors, uses its existing brand power and sheer size to tackle a large surface area at the bottom of the market, and then, finally, it develops a valuable platform on top of the new market. As the platform grows, it slowly squeezes out the existing players.
The results speak for themselves: Windows, Office, Windows Mobile, servers, business solutions, and Xbox. The process works.
Microsoft's talent is in understanding the power of platforms. Before anyone was talking about the Facebook Platform or the App Store, the biggest ecosystem of all was quietly running the world's computers: the operating system. Microsoft knows how important it is to own the platform. Their attempts to secure control of it in each industry is a recurring theme in everything they do, including their three most important business units: Entertainment & Devices (mobile and gaming), Business/Server & Tools (enterprise solutions), and Windows.
Mobile
Microsoft's execution in mobile has been excellent. I know that probably sounds ridiculous, but here's what I'm thinking: in 2007, the company found itself in the middle of a crowded street with its pants pulled down; Windows Mobile was a piece of junk and there was nothing good in the pipeline. After the iPhone, Microsoft apparently had an epiphany and made the difficult decision to start over. Kudos to them for doing that, because starting fresh is the only true way to win when you're competing against a new, totally disruptive force.
Shockingly, Windows Phone 7 is really good. It's genuinely good software, which is more than I can say about other things Microsoft has shipped. That's a good sign, isn't it? More than that, though, the company has managed to realize that the most important pieces of Apple's iOS ecosystem–including the exclusivity of apps and the strict platform design constraints on users, manufacturers, and developers–are far more important than the raw software approach Microsoft is used to taking. Microsoft deserves more credit for identifying these potential advantages, adopting them, and making them work.
Long term, assuming Microsoft can use its 600lbs to force adoption, I think Windows Phone 7 will be a better platform experience than Android. With Nokia as a close partner, WP7's chances are even better.
How could Microsoft have executed better in mobile? I can't think of many ways, even in hindsight. They were slow. But everyone was.
The Enterprise
Microsoft is not a PC company. It's not a mobile company. It's not a gaming company. It's not even really a software company. When it comes to making money, Microsoft is an enterprise services company. It provides a nebulous “solution” to the enterprise which is difficult to comprehend as a single cohesive product. Microsoft offers large businesses things not many others can: reliability, consistency, and end-to-end solutions.
Microsoft makes an endless quantity of software that is alluring to big businesses, like Office, SQL Server, Exchange, SharePoint, etc… The list goes on and on, and each of the pieces fit together like a beautiful profit-building puzzle.
Unfortunately, people tend to discount enterprise software as unimportant fluff, especially when compared to the sexiness of products from other companies and industries. But the numbers do not lie.
3,770 M Business Division 2,952 M Windows 1,738 M Server & Tools Division (229) M Entertainment and Devices (479) M Online Services
6,374 M Total income
If you compare Microsoft's earnings history to these recent numbers, the trend is clear: solving problems for companies that have a lot of money is very lucrative. Microsoft's strategy, smartly, has been to focus on them. Most new consumer-focused initiatives lose money, and those failings are disproportionately public, hence the negative sentiment toward Microsoft as a company.
Good enough and better
Ballmer has led Microsoft through an insanely tumultuous twelve years. The entire technology industry has dramatically changed at least three times. Some companies are lucky to last through one major industry change. Ballmer and Microsoft have identified each of these shifts and then doubled down on owning the platforms of the future. The strategy often fails. But sometimes it succeeds, and the successes more than make up for the failures.
I don't think it's fair nor constructive to hold Microsoft to a standard very much higher than the one it currently adheres to. Microsoft, as an enterprise services company that also builds occasionally successful consumer products, is undeniably successful. While the future for Microsoft is not so clearly defined, the past has grown out of a series of rational decisions.
Unfortunately, while fiscally rational decisions have been good enough to get Microsoft to where it is today, such decisions have never and will never catapult a company into the top of the future. It's the difference between a CEO who is good enough and one who is better. Ballmer, I think, is firmly in the good enough camp.
He might even be slightly better, because we should not forget this very consistently true fact: Microsoft makes around $5.5 billion every three months. In pure profit.
The question and answer site Quora just announced that it recently closed a $50 million series B financing which valued the company at $400 million. Usually, the purpose of a large funding round like this is clear; Square, for example, raised a bundle of cash because they're attempting to accelerate growth with expensive marketing and by giving away hardware dongles. With Quora, the purpose is less clear. There are only so many things you can do with raw cash to accelerate the growth of a question and answer community. So why would they raise so much? How could they raise so much?
Mr. Heiliger said one of the key reasons he invested in Quora is because of his friendship with Mr. D'Angelo and Mr. Cheever when they worked together at Facebook several years ago.
“I know Adam and Charlie and think highly of both of those guys,” Mr. Heiliger said. “We worked hand-in-hand at Facebook.” […]
Josh Hanna, a partner with venture firm Matrix Partners and the only non-Facebook alum to invest in the current round, said what they learned from Facebook is that “you want to be the platform,” he said. “That's where the value is.”
Interesting. Friendships and platform vision. With around 20-30k daily active users (according to AppData), it probably wasn't money raised based on user adoption.
Adam D'Angelo has posted an answer on Quora explaining, in very broad terms, what they plan to do with the money they've raised:
We are going to continue making Quora the best place for people to share knowledge, and do our best to grow it into one of the most important products on the internet for everyone. To do that, we need to:
Build out the very best team to achieve this. […] Scale up technically. […] Focus on the long term. […]
The goals he lists do not seem particularly difficult; any decent startup will do those things. Why does Quora need $50 million to achieve them? Read D'Angelo's full response here.
In an article published a couple of days ago, Forbes writer Erik Kain suggested that HBO's president, Eric Kessler, called internet streaming distribution a “temporary phenomenon,” and then he blamed HBO's supposed incompetence for causing the extremely high rates of piracy for its original content. It was a very misleading and shallow article.
Who would believe that the president of HBO is so dense that he would make such an absurd comment? It's so unbelievable that I went to the primary source, a 40 minute video interview with Mr. Kessler, which draws a fascinating picture of HBO's business strategy. After listening to the entire interview twice, I could not pinpoint where Kessler actually said “temporary phenomenon.” He hinted at the opposite, in fact, and then he articulately rationalized HBO's position from a business perspective. (He did mention that the depressed economy was leading some people to cancel cable television, but he expects and hopes they will resubscribe as conditions improve.)
I personally believe that HBO's long term viability is extremely grim, but their current strategy seems very rational to me.
In the interview, Kessler talks about why moving to internet distribution would deal a fatal blow to HBO's business. The engine that turns people into subscribers is not sustainable when HBO is not directly packaged with cable TV affiliates:
We benefit tremendously from the existing ecosystem. […] There are 60, 70, 80,000 customer service agents on the phone every day, and you know what they're talking about? They're talking about HBO. The affiliate covers that cost. The billing systems. That's the affiliates. If you watch HBO 5 minutes a month or 24 hours a day, 7 days a week, that's not a cost we have. In addition, we benefit tremendously from the fact that the cable operator bundles HBO into existing packages. So if they offer double-play or triple-play, you know, they say, get HBO free for three months. The ability to market and bundle with the affiliates is very beneficial to us. So it's very beneficial to us to keep that transactional machinery going.
What you don't want to do is to pursue a distribution channel over here [ed: the internet], where you think, well, let's go around the affiliate and we'll get a couple hundred thousand subs. But the promotional, and packaging support we get over here [ed: the affiliate networks], which, by the way, is the foundation of our 30 million subs and enables us to get 10 million transactions, if that dissipates, and that shrinks, then we will lose a lot of subs over here. Because with 10 million transactions, you have to generate a lot of subs every single day. You can't afford to have that machinery slow down. So we'll gain a little over here, and we'll lose a lot over here, and we think there will not be a net gain, there would be a net loss. So it's really about economics and a business issue.
Whether you agree with HBO or not, it is extremely difficult to argue with this philosophy that is driving the company's decisions. It's a classic problem that all soon-to-be disrupted companies face, and it is one that is impossible to solve elegantly.
Next, Kessler talks in detail about the branding implications that moving to a streaming distribution model has for HBO, and what makes subscribers feel that HBO is worth paying for:
Let me talk about the streaming environment. … let me tell you how we view that. When a consumer is sitting on their couch, and they're sitting there and they're watching their 50" flatscreen TV, and they're watching HBO, the way they evaluate HBO and determine whether this is a good value, the reference set, the comparative set for determining that, is the other networks that are coming on that flatscreen. They say to themselves, they think about the broadcast networks, the basic networks, and the other premium networks, and they say to themselves, is this entertainment experience, is it better? Is it different? Is it worth paying more money for? So they compare the HBO experience to the other networks.
If the consumer is now sitting on that same couch, and they got the iPad in their hand, and they're on HBO Go, the reference set changes. And it changes because the dominant distributors of television content and movies – certainly of television content – on that device is not the networks. The dominant distributors of that content are aggregators. It's Hulu, Hulu Plus, Netflix, Amazon, Apple, all of which have very different business models. It could be an ad-supported model. Hulu Plus is an ad supported model that you subscribe to. Netflix is a standalone subscription service. Amazon is a free service, part of the shipping deal that you used to use to get hard goods. Apple is of course an EST service, we're embedded.
The business model that drives the dominant players in internet streaming today–the players who have the most distribution power, and the ones that would be required as partners to fund HBO–simply is not compatible with the way HBO works as a brand and as a sustainable premium package service.
Basically, HBO has to differentiate at a further abstracted plane than the distribution aggregators, because it doesn't have the existing business infrastructure that normal networks have. Kessler continues:
We look at that comparative set, and we ask ourselves, how do we differentiate? How are we different as the consumer evaluates HBO Go against the aggregators. And there's a couple ways we're different. First of all, those services offer quantity. You know, you go to Netflix or any others, and there are hundreds of thousands of pieces of content that you're streaming. We will never compete in that ballpark. We don't want to compete in that ballpark.
What we compete on is quality. If you want HBO programming, the only place to stream it is through Go or through our affiliate portals. If you want to stream our theatrical movies, in the paid television window, again, only through us. The content on those services, it's basically non-exclusive. So much of that content is available on each other's sites. Our content, exclusive. It's the only place you can get it. And we believe there is value in exclusivity.
Kessler then talks about how HBO wants to do more than just stream content to mobile devices and televisions in the future. They want to enhance the experience of watching television with devices like the iPad. Other networks can't do that when they license their content, because then they don't own the entire experience. This is similar, in some ways to Apple's philosophy of marrying the hardware and software into a single experience. HBO wants to marry the distribution technology and the content to create a single experience.
Lastly, those services are aggregators. They're licensing content. What we do … we are content creators. And we believe in order to continue to differentiate the service … there's a tremendous opportunity to marry this platform, to marry HBO Go and this technology, with the content creation skills of our network.
The entire interview is worth watching if you want to understand the intricacies of how HBO's business works and why Hulu is currently attempting to create its own content (with a business model that is so far unsustainable, by the way, unlike HBO, which is hugely profitable).
When a news organization like Forbes publishes linkbait like this it is bad for everyone. But it is especially bad for journalism, which is already in a questionable state. I expect better from Forbes, especially for business analysis. It's not really HBO's fault that Game of Thrones is pirated so much. It's the fault of the entire ecosystem.
