The seminal moment

For as intangible as the idea of “company culture” is, it can be thought of as the stories that are passed on in a company and the notion of “the way things are done around here.” It seems every successful company has a breakthrough moment, an event that lives in the memories of its first employees as being a mythological “hero story” and a catalyst for later success. In the case of Open Systems, this important catalytic event came for Ann Winblad when her small team was doing consulting in 1977 for CADO computer. She was invited to present her accounting system product to a group of CADO resellers at a conference. She did a demo of the product and had the nerve to offer a “blue light special” as she called it giving a major discount to those companies that would write her a check on the spot. It was a formative, confidence-building experience as she ended up landing twelve different deals that day and walking away with $120k in checks for a company she had started from literally nothing borrowing $500 from her younger brother.

We’re still pre-revenue at JumpBox rapidly approaching sales but one of the formative memories thus far for me was our attendance at the rPath Virtual Appliances Summit last month. There we were, a company of four people along side major vendors, the next biggest one being a $15MM VC-funded, 38-person virtual appliance toolset provider. There were higher-ups from big names like Dell, HP, Sun, AMD, Intel, Parallels and VMware. Eating dinner sandwiched between the CEO of Red Hat Linux and some big dog at VMware was a little surreal but it was a major morale boost and validation to what we’ve been doing.

On teaching as opposed to telling

After Open Systems ran its course, Winblad founded a venture capital firm. When asked about how the business was different in approaching entrepreneurship from the perspective of a VC, she responded, “I think it’s really interesting being a venture capitalist because, when you’ve got 30 years of experience, then our challenge is how to teach and not tell. Because you want people to figure it out. You want to make sure that you can grab them by the coattails if they are falling off a cliff, but you want them to discover the edges by themselves. That’s the biggest challenge of moving from being a business leader to being a business investor. Your job is not to tell, but to teach.”

Winblad’s advice for founders

“We try not to give too much prescriptive advice. “Think like a big dog and then figure out how you find leverage to get there.” You have to have tactics to get to strategies, but you have to have a strategy, and you have to put your strategy up here and then see ‘Where’s my gap’ to get to this aspirational goal. You’re always going to be short of people, you’re always going to be short of money, you’re going to be short of source supply value. So you have to find leverage points, versus working your way up through tiny little rungs and seeing if you get there. Think like a big dog, and find leverage to get there.”

Winblad sold Open Systems in 1982 for $15MM+ and works now as a founding partner at Hummer Winblad Venture Partners.

How Adobe began

Charles Geschke got his start in 1972 at Xerox Palo Alto Research Center (PARC). He developed a technology called Interpress that would become the precursor to Postscript. It was a language that would make it possible for any computer to talk to any printer. In the fall of 1977 he and his team flew to Florida to put on an “internal trade show” for Xerox management. The demo was a success and management was excited but said it would be a seven-year process to commercialize the Interpress technology. Charles and his chief scientist John Warnock didn’t want to wait that long. The two talked with an acquaintance who was a professor at the University of Utah and sat on the board of a venture capital firm. One introduction led to another and they found themselves in discussions with Bill Hambrecht of Hambrecht & Quist Venture Capital. Bill persuaded them to quit their jobs and take a $50k personal loan to start their company. Adobe Systems was born.

Started out creating the wrong product

This theme will seem utterly repetitive by now, but like almost every other startup in the Founders book, Adobe set out to build the wrong thing. They envisioned creating a computer/printer combo package that they would sell as a complete self-publishing solution. They were approached by Digital Equipment and later Apple, who were both very interested in the printing software they had developed. By this point though, Adobe had raised $2.5MM in funding under the story that they would sell the computer/printer combo package and they were stubbornly clinging to the original plan in spite of two large potential customers begging them for just the software. They went back to their board and received the sage advice from the chairman to ditch the plan and listen to their customers. Q.T. Wiles said, “You guys are nuts. Throw out your business plan. Your customers - or potential customers - are telling you what your business should be. The business plan was only used to get you the money. Why don’t you rewrite a business plan that is focused just on providing what your customers want?” They followed the advice resuming talks with DEC and Apple and soon after closed deals with both.

Creating an industry

Geschke and Warnock had the ambitious goal to not just create a company but an entire industry. The Postscript technology they developed brought desktop publishing capabilities to the masses. For a few thousand dollars you could now ditch the tedious analog method of publishing and produce cleaner output more cheaply and more quickly. Geschke says, “As graphic artists and designers began to learn how to use a computer, we brought out products like Adobe Illustrator. All of a sudden, the whole industry began to move, and within less than a decade the entire printing and publishing industry went from the old analog world completely over to the digital world. That was a tremendous thing to see, and of course it was a huge benefit to us.”

One-trick ponies die young

They realized early in the life of Adobe that having a single-product company was a fragile way of life and sought to branch out their product offering. About this time they were looking for a simpler way to generate the Postscript code that was sent to the printer. Geschke says, “John’s wife was a graphic designer, and one we brought out the LaserWriter, she wanted to get some of her design concepts out on that machine. So John was programming in PostScript by hand to get this output to come out and he said, “This is stupid. I need to build a tool that behaves more like what a graphic artist would expect to have in terms of pen and ink and drawing and so forth, and then let the tool write the PostScript code.” So Illustrator wasn’t originally as much a drawing program as it was a visual IDE for generating PostScript code.

On the value of the customer relationship

At the Seybold conference in San Francisco in 1989, a blow was struck to Adobe as its largest customer (Apple) and their largest competitor (Microsoft) announced an alliance around the TrueType technology that would replace PostScript as the standard for publishing. They did an ad-hoc panel later in the conference to explore the audience’s take on the announcement and the resounding feedback was in support of Adobe - the audience did not want them to go away. This development galvanized Adobe to action on quickly bringing a cross-platform version of PostScript to market and making it available at a low enough price point that it could be an easily-justifiable add-on for consumers to purchase. The takeaway from that conference was a clear affirmation of the importance of the relationships they had created with customers and their customer service philosophy. Geschke says, “That reinforced a message that John and I had always preached inside the company about how to treat our customers. Listen to them very carefully. Understand what their requirements are and what their needs are. Not necessarily do what they asked us to do, but to have the vision to do more than they expected…And that event demonstrated it; basically everybody voted for us. In fact, while there was a hiccup in the stock because of the Apple-Microsoft announcement, our business never faltered.

Adobe IPO’d in 1986 and is still today the undisputed leader in quality publishing and design software.

The birth of Alexa and the Internet Archive

Brewster Kahle had founded the Wide Area Information Servers (WAIS) search engine out of his company Thinking Machines an sold WAIS to AOL in 1995. Alexa and the Internet Archive were his next endeavors on a path to his dream of creating the world’s largest library of published information. Alexa provides the equivalent of the “Nielsen ratings” through tracking a sample of all Internet surfers’ bahavior via a toolbar installed in their browser. Alexa takes a snapshot of each web site it indexes and donates it to the non-profit Internet Archive to create a giant scrapbook of every web as it evolves over time. I’ve been using the Internet Archive’s “wayback machine” throughout this series to display what the founders’ web sites looked like at the time they were created.

Choose a setting that inspires you

Kahle’s advice to startup founders: pick a location for your office which will inspire you every day. Kahle says, “If you’re trying to get your company to think differently - to do something interesting - pick your setting carefully. Thinking Machines was set in an 1800’s Victorian mansion on 100 acres of forest just outside of Boston. It was a park, basically. Working in an environment where, if you got stuck, you’d go for a long walk is very different than trying to do a startup and think differently if you’re in Suite 201 in some major office complex. That was a lesson I’ve used every startup since.”

I completely agree with Kahle on this- the location and design of our office is non-traditional and we love it. We’re in the heart of Tempe a few blocks off famous Mill Ave on the bottom floor of what’s called a “live-work space,” basically a residential condo that’s been re-zoned for commercial use. The vibe of being close to ASU campus and within walking distance of so many great restaurants is key. We’re also in this odd junction of various transportation being next to a railroad, the new light rail, in the flight path of Sky Harbor airport and on the Tempe Flash bus route so there’s this constant feeling of motion, development and progress. It’s no substitute for being in an environment like Palo Alto or Mountain View but for Arizona we lucked out in being in about the best spot you could pick to do a startup.

The importance of controlling your own distribution and pricing

Kahle talks about how his company put a bunch of newspapers online and the lessons he learned through that experience. “You take it for granted, but all the newspapers are pretty much online now. They control their own distribution. They have their own websites. It doesn’t all funnel in through an iTunes. The music guys, I’m not sure why they did this, but they sold their souls. Somebody else controls not only the distribution of their product, but they control the pricing. What do you have if somebody else controls the distribution and pricing of your product?”

Loud and clear on this one. With JumpBox we’ve been heavily dependent on traffic coming from VMware’s VMTN marketplace for virtual appliances. Until recently they supplied the majority of our traffic. We’re a VMware partner and thankful for their existence but as Kahle says it’s dangerous to have an external entity become your exclusive means for distribution. We now have presence in the Parallels equivalent of VMTN and are starting to see traffic from other sources, which is a healthy thing. Ideally when our library of applications becomes large enough, it will become a destination itself and we’ll control our own fate as far as distribution. Until that point we continue to rely upon other vendors for our traffic.

The Amazon acquisition

Amazon acquired Alexa in ‘99 for $250 million in Amazon stock. Like Paul Graham and the Viaweb team, Kahle told Jeff Bezos of Amazon that he didn’t do well assimilating into a bigger corporate culture in his last acquisition experience. Bezos gave him the latitude to keep Alexa running as an independent company and they went they went forward under that plan. Having the freedom to run things his way kept Kahle with Alexa for three years after the acquisition. He eventually left and moved across the street to work at the sister non-profit company Internet Archive where he works today pursuing his goal of creating a free public library of everything ever published.

No surprise: they started building the wrong thing

Flickr was no different from the rest of the FAW stories in that they started building something completely different and an offshoot of that became the product that caught on. Co-founders Caterina Fake, Stewart Butterfield and Jason Classon began building an online game called “Game Neverending” in which players formed chat groups and interacted via instant messenger. When they added a feature that allowed players to share and discuss photos, the feature became the focal point and the game was forgotten. They decided to put game development on hold and focus on what the users were clamoring for which was enhancements around the photo sharing feature. They spun off this feature into a standalone web product called “Flickr.”

What made it tip

Fake says, “Flickr started off as a feature. It wasn’t really a product. It was kind of IM in which you could drag and drop photos onto people’s desktops and show them what you were looking at. We built it really fast; we had a lot of the technology already from the game but we built the first instance of Flickr in eight weeks…The response was positive but it didn’t end up being a compelling product mainly because it was a feature. It had a critical mass problem…It still grew, slowly. But it really started getting traction when we added the ability to put your photographs on a web page.” “Social networking got people used to this idea that they could make an online digital identity…And social networking as social networking pretty briskly showed itself to be a fairly pointless activity…But when you tied it to a very specific, very connective activity like photo sharing, it really flourished.” Flickr exposed an API that further opened up their service and allowed developers to create add-on components and offerings that directly tied into the Flickr back end. This further strengthened their position.

Only made possible by their naïveté

Had the founders been more thorough in doing market research, they might have tainted their creative approach to Flickr and imitated the existing photo sharing sites. Fake says, “This is weird, but one of the things that enabled us to innovate within this space was that we hadn’t done our research…We were naive and optimistic. What we did was just start building stuff. And I think if we had sat down and done the research, we would have looked at the companies that had actually made businesses in this area, like Ofoto, Shutterfly, and Snapfish. Basically their model was that photo sharing was a loss leader for photo finishing services…Photo sharing wasn’t seen as a valuable enough activity that people would pay for that itself. So I think that our naïveté was what made the whole thing possible.”

In the same vein, we’ve purposely kept our engineers from using the other virtual appliances on the market so as not to taint their idea of how to best deliver the experience. In the same effect of “seeing the movie before reading the book” in that it makes an indelible mark on your imagination and your ability to form your own vision. This has been a good strategy for us thus far as we’ve come up with some innovative ways to think about the problem of delivering pre-packaged applications. Had we done thorough analysis of everything on the market and exposed our engineers to the designs, our engineers surely would have been limited their ability to think creatively about the problem and JumpBox would not have the revolutionary design that it does.

Emergent behavior via tagging

Like del.icio.us, Flickr employed the concept of “tagging” for organizing one’s photos. Tagging produces interesting results because people, in the pursuit of their own self interests for organizing their photos, will use similar tags. If the photos are public this yields a unique taxonomy across users of what’s important to people. There are interesting situations that emerge. An example Fake referenced in the interview: “There’s a guy who was on vacation in Maine and got an alarming phone call from one of his neighbors saying that his apartment building in Atalanta, Georgia was on fire. So he immediately went on line to Flickr and typed the name of his apartment complex, ‘Atlantic Landing Georgia,’ and found all of these pictures of his apartment complex on fire. He was able to see that the fire was on the opposite side of the building and that his apartment wasn’t affected, so he didn’t have to panic and call his insurance company; he could continue on his vacation.” Other eccentric, random groups like “Squared Circle” emerged where people would frame a circular object and crop it into a square photo. Viewed as a slide show it would produce a strange visual effect to see the transition between these objects. Most photo sharing apps had been private with the ability to expose photos to friends but they found that people when given the opportunity we’re happy to expose their photos publicly.

The perfect storm of market trends

Flickr had perfect timing to intersect several important market trends. Fake says, “We could not have timed it better. All of these things were in the air: blogging, social networking, camera phones, the ubiquity of network, suddenly more people were on broadband. All these things converged at the same time and we were really well-positioned to ride that wave.” The moons were in alignment and the scene was primed for the emergence of a service like Flickr that allowed people to manage their photos and publish them to their blogs. I’ve been using Flickr myself for the past two years as the sole mechanism for managing my photos. For many of my photos I don’t have a local copy - everything resides on Flickr. When a company is able to entrench themselves this deeply with so many people relying upon their services for something as personal and important as your authoritative photo album, they win.

Likewise it’s our hope with JumpBox to provide applications that allow people to collaborate and manage their important data in such an easy way that they come to rely deeply upon the JumpBox offering.

Flickr was acquired by Yahoo in March of 2005 for an undisclosed sum. Yahoo recently announced the abandonment of their Yahoo Photos service in order to focus fully on Flickr as their main photo product. Flickr embodies many qualities of the Web 2.0 revolution and is frequently used in conversation as the poster child for a Web 2.0 startup.

The birth of Craigslist

Craigslist began as a weekly email newsletter that Craig Newmark would send to friends notifying them of interesting upcoming cultural events in San Francisco. Newmark kept the quality of the emails high and gradually more and more people asked to be added to the email. Mid-’95 his pine email program was unable to accommodate the number of recipients and he migrated it to a donated listserv and called it “Craig’s List.” He was using the listserv as a database of sorts and wrote some perl scripts that would extract the emails and publish them as web pages and Craigslist the web site was born.

On the philosophy of not charging

Newmark made a decision that he wanted to keep the free of cluttered advertising and when Microsoft Sidewalk approached him in ‘97 to run ads for their local cityguide, Newmark turned them down. He felt he was making enough money through his consulting and was willing to sacrifice the revenue to keep Craigslist a pristine ad-free system for handling classifieds. In 1999 Newmark brought Jim Buckmaster on board as CTO in order to pursue it as a full business. Jim later took the role of CEO and helped put business structure to the hobby site that was Craigslist at the time. They’ve managed to keep 99% of the site free to use. Newmark said, “‘The principle is: charge people who would otherwise be paying more money for less effective ads.’ [Users] specifically said, ‘It’s cool to charge for job ads and to charge landlords or apartment brokers.’ Beyond that, there was some mix of opinion, but we stuck with that.”

We made a similar decision that we wanted to keep JumpBox free in it’s most basic version for people. We considered making all JumpBoxes run under a time-limited constraint until registered but ultimately decided that by giving away something of value and not crippling the usage by deactivating after a certain time period, we’d see more loyalty and eventually convert more people to paying customers as they grew to rely upon it and realize the value. The power users that need features like automated backups and shell access are theoretically using it frequently and are either making or saving money with it, and therefore can justify the nominal license fee.

On keeping lean

Craigslist has grown to over five billion pageviews per month and has a staff of twenty people to handle operations. It is still orders of magnitude smaller in staff than sites that have less traffic. The way they were able to scale things effectively with so few people was by pushing much of the duty of policing against spammers out to the users. They developed both the tools for the site as well as the right culture amongst users to be able to make the site self-policing. The concept of “flagging” has been instrumental in bubbling up abuse from listings so that they can be dealt with by a staff member. Newmark says, “[Flagging] works great in all sorts of ways, and it’s also an expression of our values. Mutual trust. This is kind of democracy in real life. Everyone wins, except for the bad guys.”

On the general nature of humankind

Newmark says, “What surprises me, in a way, is how almost universally people are trustworthy and god. There are problems, and sometimes people bicker, which is a pain in the ass, but people are good. No matter what your religious background, we share pretty much the same values. There are some minor differences that we disagree on, but the differences are at the 5 percent level. That’s pretty good.” Throughout his interview, Newmark constantly refers to the “moral compass” that guides every activity with how they run the company. “I’m not implicitly judging anyone here. We’re not anti-traditional by any means. We just made a specific decision based on our specific values and followed through.”

In 2004 eBay took a 25% stake in the company via a stock purchase from an ex-employee. Craigslist is still privately-held though and has massively disrupted the classifieds business for local newspapers. It’s currently in the top ten most trafficked sites in the US.

How Bloglines got started

Mark Fletcher had created the service ONElist which later became eGroups and then was acquired by Yahoo. After a sabbatical of travel to clear his head, he was looking for what to do next. He started a company called Trustic which was focused on developing an anti-spam product. Like so many others, the tool he had created for himself to solve another problem during development became more important than the product he was developing. Fletcher came to the realization that making anti-spam measures is not a fun business to be in- “because everybody hates you. You’re never perfect. You either don’t block enough spam or you block somebody’s favorite emails. I quickly got out of that.” He took the tool he had developed for himself and shifted that to be the main product development focus of Trustic and Bloglines was born.

Serendipity

They put the Bloglines product online in June of 2003 and received immediate press coverage. It was perfect timing given the exploding popularity in blogs in RSS. Though RSS was not yet mainstream and popular more amongst technical circles, it held a huge advantage for anyone who had to keep up with a constant flow of news. For reporters, it meant not having to check a hundred different web sites every day. They saw the immediate benefit and the ease of use of the web-based aggregator made it a frictionless process for them to adopt it (ie. no installation). “I think we got really lucky because blogs in general started to become really big and the downturn was ending, so you had all of these people looking for the next big thing. Also a lot of reporters used Bloglines. They like to talk about things they use, so we got really fortunate in that regard.”

We’re counting on a high transmissibility factor for JumpBox among technical circles as well. Bloglines shaved hours of the daily task for a reporter to visit a ton of web sites and read the new developments. JumpBox has a similar promise in terms of time savings for the average IT administrator- the tedious task of setting up a new Open Source server application can take anywhere from a couple hours to a couple days. This task is now reduced to a thirty-second setup process with JumpBox. Beyond neutralizing the setup process, the IT admin has reduced maintenance concerns and gets the comfort of knowing that he/she can effortlessly move the application to faster hardware if it starts getting pummeled and avoid re-installing components and migrating data manually. Our expectation is that, like the reporters that sneeze the advice of helpful tools to their colleagues and ultimately their readers, IT admins for whom we save a significant amount of pain will go and tell their friends in the industry about the “secret advantage” of using JumpBox.

How does Bloglines make money?

This is the one question I still had after reading this interview. Fletcher said, “My philosophy on these types of companies - consumer-based Internet companies - is that you don’t need to worry about the business model initially. If you get users, then everything else follows. Basically any technology can be copied, any concept can be copied. In my opinion, what makes one of these companies valuable is the users. That can’t be copied.”

Hey- no arguments that the users and the brand are the value with any company; they and their loyalty to your product cannot be duplicated. But at the end of the day there still needs to be some business model. I get the idea of “just concentrate on making something people want” and clearly Bloglines is something that a lot of people use and love. But nowhere on their site is there an opportunity to pay for a premium-level service or be subject to advertising. I still don’t understand how it makes money. Ask Jeeves acquired them so perhaps the value of having the Ask brand on the home page and their partner destinations listed in the navigation is enough. They undoubtedly have a strategic plan for how it makes them money but that model is anything but apparent.

On how startups are changing

Under a decade later the landscape for doing a tech startup has completely changed. Virtual hosting options reduces the hosting fees by an order of magnitude. Off-shore development is becoming more practical and there is now an abundance of information via blogs on practices like negotiating a term sheet with a VC that were previously privy to only an elite few. Fletcher says, “The whole VC process in general has been very closed and oftentimes by design by the VC’s. Because they don’t want to be negotiating against other VCs they don’t want terms of the deal to get out, so it’s in their best interest to keep things secret. But it is very nice that things are starting to open up now, whether they like it or not… The great thing for entrepreneurs these days is that there is so much more information out there than there was ain the ’90’s. Any number of people that have gone through this are blogging. All sorts of VC’s are blogging now. There are a lot more books out now. You can just do a search and find sample term sheet, for example…the other thing, doing startups like this is so cheap that it just doesn’t require a lot of money. I think I put in a total of $200,000. And I didn’t do it nearly as smartly as I could have. I ended up buying all the computers. My recommendation would be: don’t buy any computers. Just use the virtual dedicated hosting services.”

I agree with these points- I’m subscribed to the blogs of about eight different VC’s and there is a transparency that did not exist even five years ago regarding how the mechanics of entrepreneurship work. Hosting-wise- utility computing services like S3 and EC2 are making it so anyone has access to powerful computing resources and can pay for only what they need without having to buy hardware, rent the space at a colo and hire IT staff to tend the servers. We managed to bootstrap JumpBox on $108k to the point at which we raised some angel investment and (knock on wood) this should comfortably carry us through to customer revenue at which point we’ll be in a position to decide the best way to grow the company. But Fletcher is right, the barriers to do a startup have come way down and anyone who can scrape together six months of living expense reserves and develop something useful has the ability to do their own.

On the psychological hurdle of shipping

So true so true… Fletcher says, “Just get something out there. If you find really early versions of ONElist or Bloglines on archive.org, the websites are horrible. They are crap, they don’t have any features, they just try to do one thing. And you just iterate because users are going to tell you what they want, and they’re your best feedback. It’s critical just to get something out quickly. Just ot start shipping and then you can iterate. Because shipping is just this huge hurdle. I’ve been a part of companies that have had big problems shipping- they just can’t ship. It’s a psychological thing.”

We’re itching to ship a production release of JumpBox and with an imminent final release candidate tomorrow, our next release should bring us through that psychological barrier that Fletcher describes. We’ve had 10k+ downloads of our appliances to date and a fair amount of user feedback in our forums and via our private usage surveys. There’s still no substitute for accepting payment for what you’ve made- we recognize it as an important milestone morale-wise and we’re looking forward to hitting that shortly.

Another interesting byproduct of shipping early and intentionally leaving out important features is the beneficial effect on customer loyalty when they inevitably request the missing feature and you add it in. Fletcher says, “We were getting 50 to 100 emails a day at Bloglines and most of them were feature suggestions. Once you start acting on those feature suggestions, the suers see that you are actually listening to them and they become more loyal to your site. Because they see they are able to participate in this and it’s just kind of like a virtuous cycle. So it is not a disadvantage - it may even be an advantage - to ship without all your features initially, for that reason, because you get all of this going and you get out there sooner.”

On the importance of an MBA

According to Fletcher, you don’t need it. The rationale is that you won’t be focused on biz dev early in a startup, you’ll be focused on engineering the product to be useful enough to attract users. “You either stand on your own or fail on your own, initially. If you build momentum, if you get users, then deals may come your way, but a lot of times most deals don’t make sense, so you don’t need hardcore MBAs.” I have a friend who founded a tech company and has grown it slowly over the past six years. He recently completed his MBA at Thunderbird School of Global Management after having started the company because it gives him the theory to better grow and manage his company at this point. But the idea of a would-be entrepreneur facing a crossroads at the beginning deciding whether to go secure an MBA before starting a company is silly- there is no quicker education than that of just doing it.

On making a graceful exit

In the same way that Paul Graham stayed on with Viaweb to ensure a smooth transition after Yahoo acquired them, Fletcher stayed on with Ask after Bloglines acquired them. “With Ask, I stuck around for 14 months. Not that it was contingent on the acquisition in any way, it was just the right thing to do… I’ve learned that your reputation is very important, as an entrepreneur, as a tech guy in the Valley, and it’s a good thing to worry about your reputation. I was very concerned about that, and so, when only two people are coming over - and most of the knowledge was still in my head - it wouldn’t have been right for me to leave.”

Fletcher’s interview was one of the ones that most resonated with me personally. Whether it was his direct conversational style, him talking about apps he had homegrown to manage a server from his Treo or the fact they used the same law firm as us, I felt like this is somebody I would want to sit down and buy a beer. I had been reading his blog Winged Pig for the past year so perhaps contributed.

Ask Jeeves acquired Bloglines in February, 2005 for an undisclosed sum. It’s the second most popular web-based feed reader today behind Google Reader (though having tried both I still prefer it myself to Google Reader).

Itch-scratching at its finest

Joshua Schachter had 20,000 bookmarks he was tracking in a single text file. He knew his homegrown system was breaking down and decided to develop a web-based system to manage his bookmarks. In 2001 he had a working version that published his bookmarks to his site and he was getting 10k visitors per day. By late 2003 he had reworked the system to be a multi-player version which we now know as del.icio.us. By the end of 2004 del.icio.us had 30k users.

Being over-capitalized would have killed it

Schachter attributes the success of del.icio.us as a product to the fact that it emerged slowly as a hobby site rather than a rocket-fueled, VC-funded company. It was constructed using best practices from everything he had learned in building his own bookmark tracking system so it incorporated the set of essential features that had been validated through personal usage. Schachter said, “I think in general being overcapitalized is a path to failure. The VC’s want you to spend. There are general ills with being over-funded…There was no point at which I said, ‘I’m inventing this wonderful new thing.’ I just sort of realized that I had evolved my own filing system, and it worked for me. I’d used it for a long time before del.icio.us even showed up. This was the codification of that practice.”

How to triage feature requests

Schachter said, “I think people ask for features - they want to do something, but they don’t say, ‘I want to do that something.’ They translate it into some feature that typically they’ve seen somewhere else and ask for that instead…So, stuff that people ask for, I tend to try and dig to the root cause, before reducing to practice.”

I recently wrote a piece called “How to tap the knowledge you don’t know you don’t know“- it covers this very scenario. There is a rift in translating customer needs into product features. The bridge-building that occurs in order to understand the desired feature set is typically done by the product development team soliciting feature requests from users. Schachter’s advice is to instead seek the root cause behind the features they’re requesting. For instance, in our realm, users of traditional installable desktop software typically aren’t aware of the capability to ship a mini virtual computer that is fully preconfigured and self-contained- they don’t know what they don’t know about virtual appliances and therefore the features they request are related to usability failings with their current, known offering. It’s up to product development team to cross that divide and help “build the bridge of understanding” from the customer end by understanding their needs and requirements rather than seeking implementation-level feedback.

More 37signals philosophy

Schachter echoes two of the core 37s tenets of “Embrace Constraints” and “Less Mass.” Schachter found that through working on del.icio.us during his full-time job at Morgan Stanley he had to work in bursts of fifteen-minute windows where he would be under hard time constraints and therefore severely limited in what features he could add. These constraints forced him to squeeze ultra productivity from the limited time he had and meant the product had less mass in terms of extraneous features. He says, “Reduce. Do as little as possible to get what you have to get done. Do less of it; get it done. If you’ve got two things that you want to put together, take away until they go together. Don’t add another thing. Because you can understand it better, you can analyze it more cleanly. The UI will be easier. Doing less is so important.”

On explaining a difficult value proposition

Schachter said, “It is a challenging product to do conceptually. It’s not something like, ‘Lets you file your taxes better.’ There’s no clear value proposition here. It is valuable, but hard to understand. You will be able to remember more things this way, and with that, people don’t even realize there’s a problem. So that’s a challenging value proposition to explain or get across.” I sympathize with Schachter on this because it’s a service that until you use it, you don’t fully comprehend what it does for you. Similar to the StumbleUpon service, it’s just not something that you need but it’s something that’s certainly useful once you use it.

The freedom paradox: “a robot on rails”

For all the apparent freedom that founders have in building their startups, once you’re in motion down a path there is not as much latitude as one might think. Schachter said, “It’s a combination of sudden freedom to run things as you please and crushing responsibility in which you know you have to do certain things in a certain way at a certain time. That eradicates all of that freedom. You become a robot on rails…every step was sort of the inevitable, inexorable progress due to the previous steps in the path. It’s not like I had no choice, but everything I did was the only choice because it was the only thing that made sense at the time.”

I would differ on this point from our perspective. During periods you’ll operate “on rails” but there are absolutely “switch ties” where a crossroads decision presents itself and the right decision isn’t apparent. For us these moments usually involve an afternoon of talking through the pro’s and con’s, doing a SWOT analysis on the different scenarios, examining the implications from every possible perspective internal and external to the company against the backdrop of our vision for where we’d like things to go. The dots make perfect sense in hindsight but not necessarily at the time looking forward.

Yahoo acquired del.icio.us in December of 2005 for an undisclosed sum rumored to be about $30MM. It continues to be the most popular social bookmarking service in existence today.

A powerhouse entrepreneur

If you are a programmer or entrepreneur, you have heard the name Paul Graham. This is a guy who wasn’t satisfied creating just one successful company, he needed to establish an incubator to grow many successful tech startups simultaneously. I had read his book Hackers and Painters before finding his story in Founders so I already knew the scoop on the creation of Viaweb (fyi: many of the essays from that book are available on his web site). We had actually applied to his Y Combinator summer program a year ago when we were first getting JumpBox started. This guy is a mentor to many tech entrepreneurs through his essays and his talks - he has a no-BS approach to everything and writes with a powerful, crisp style. It’s clear that he’s in it for the love of the game and simply enjoys working with smart people to develop game-altering technologies. MacGyver used to be my hero growing up- I believe now it’s Paul Graham.

The essence of Viaweb’s innovation

Like many others, founders Paul Graham and Robert Morris started out building the wrong thing. They were seeking to put art galleries online with their startup Artix. They quickly realized that art galleries had little desire to be online. They changed up their plan and decided to do make store builder software to produce online stores. It was originally intended to be desktop software that would generate files that would then be uploaded to a server but early in development they had an epiphany in realizing that they could write the administration piece as an online application itself. This was an exciting proposition for a couple reasons, namely as Graham says, “I was pretty excited, because it meant we could start a company without having to learn Windows. Neither of us knew how to write Windows software and we didn’t want to learn. It seemed like this huge steaming turd that was best just avoided.”

“Terrified of customers”

With $10k of seed investment and a summer of working out of an apartment in Cambridge, Massachusetts they were able to get a prototype of their online store builder working. When they realized it was going to take more than a month or two to make it happen they hired another engineer to distribute some of the programming load. They realized they were going to need more money and they faced a decision whether to focus on raising money or getting revenue via customers. Graham admitted that being more technical in nature, they were afraid of having customers: “Being a sales guy and being a hacker are two very different kinds of work. We were very comfortable dealing with hacking, but dealing with customers seemed like this terrifying unknown. If it seems strange to you that we were afraid of customers, imagine how the average sales guy would feel about modifying the software running on his laptop.”

The advice I hear repeatedly from entrepreneurs is “get to revenue as quickly as possible as this puts you in a stronger position when you are talking with investors.” While I wouldn’t say we’re “terrified of customers,” we’ve grappled with this dilemma with JumpBox- on one hand we try to live extremely lean and limit sales and PR until we’re completely satisfied with the quality of our offering. But on the other hand you hear the advice of people that just say “go, go, go- it will come together along the way. It’s more important that you’re out there evangelizing and getting customers.” We saw Reid Hoffman, the founder of LinkedIn, speak at a Churchill Startup panel in Palo Alto and he said, “if you’re not utterly embarrassed by your first release, you’re waiting too long.” We’ve tried to strike a healthy balance (if not erring a bit too much on the conservative side of moving slower than we should). But we’re loosening up the coffers a little, attending more events and in fact decided just today to pull the trigger on ISPcon next week in Florida.

The 37signals philosophy of staggered development windows

Paul Graham seems to share a similar philosophy to the 37s crew in many respects (btw, 37s put their entire book “Getting Real” online for free - we bought it when it first came out but it’s well worth a read at that price ;-). One similarity I noticed in how both companies developed their products was this notion of “staggered development” and “alone time.” When Graham and Morris were writing the LISP code behind Viaweb, they had opposite schedules- Graham was nocturnal while Morris was an early-riser. This actually proved beneficial though allowing them alternating windows of focused work. Paul said, “Robert used to get up early, whereas I stayed up ’til four and got up at noon. So we would kind of work a 24-hour schedule. I would write some new code during the night and send [Robert] an email saying, ‘OK, we’ve got all these new features in my part of the code.’ Then he would write the corresponding stuff in his part. So we got code written very fast.”

Misidentifying the customer

Viaweb’s initial thought was that their customer would be catalog companies looking to extend their distribution online. It proved a false assumption though and turned out that catalog companies were more defensive with the perspective that the online channel would invade their existing mail order territory- they just wanted to close their eyes and ignore it. They had an existing mass distribution channel and therefore were not super motivated to explore the online store option. This is a classic instance of the “Innvoator’s Dilemma” and pushing a disruptive technology on someone whose incentives are not properly aligned. Once Viaweb realized their customer was the individual merchant looking to take their single point-of-presence retail outfit and gain access to a ton of new eyeballs, they crossed the threshold from selling uphill to taking orders.

“eCommerce isn’t about transaction processing”

Paul said, “Our secret weapon was that we knew that e-commerce was really about graphic design, not transaction processing. Unless you had a site that could convince people to buy, you didn’t have a transaction to process, and what convinced people to buy was how good the site looked. So we made sure that our software made great-looking sites - not just better than our competitors, but better than most of the sites that big companies paid web consultants half a million dollars to make for them…We didn’t even process credit card transactions ’til about two years in. We would just forward the order to the merchant, and they’d process it like a phone order.”

In our world the equivalent to this statement is that virtual appliances aren’t about hypervisors and virtualization platforms, it’s about simplicity of the experience and intuitiveness of use. We spend a great deal of energy ensuring that our stuff is dead-simple to deploy. The litmus test for success here is the day that I can point my mother to our web site and she can figure out how to download an Open Source CRM system and install it on her computer using JumpBox- when we achieve that level of usability, we’ve done our job.

Investors are like horses

While I can’t personally validate this statement yet, I found it very interesting. Graham said, “You know, in retrospect I think the big problem with our investors was that we weren’t forceful enough with them. I think investors like to be bossed around, like horses. It reassures them when you’re in control. But these guys were much older than us and had given us huge sums of their money, so it was hard for us to boss them around.”

Our situation is a bit different in that the mean age of our investors is about 27yrs old. Investor relations thus far has been a pleasant monthly update - we were fortunate with getting investors who trust us thoroughly and haven’t encountered the “skittish investor syndrome.” But I can see how being wishy-washy would potential create more problems than being resolute when the outcome is uncertain. The risk is always there, that’s the nature of the game.

Going back to “high school”

Unless you’re working towards an IPO with your startup, the most realistic exist will be an eventual acquisition. While we have way bigger fish to fry with JumpBox before we worry about this day, it’s something that will ultimately happen. Paul said, “It was kind of weird that when the deal closed, we all became Yahoo employees. It was like one of those dreams where you have to go back to high school. Up till that point we’d been independent, and then suddenly we were employees, with bosses. And the weirdest thing was, we, or I at least, actually started to think of them as bosses. Now whatever I did was either submitting or rebelling, whereas before it had been just doing.”

Having answered to middle management dorks at a 2000-person company for a few years, the thought of having to answer to people other than our customers and shareholders is… well… unattractive. Yahoo gave excellent terms to Viaweb employees- they didn’t have to vest at all, they were free to leave on day one. Consequently, Viaweb people stuck around to ensure the acquisition went smoothly. It’s our hope that one day we find a company that we mesh well with that has the massive resources to help fuel the JumpBox revolution we see for deploying all linux software in this fashion.

On selling and telling the truth

Graham said, “I found I could actually sell moderately well. I could convince people of stuff. I learned a trick for doing this: to tell the truth. A lot of people think that the way to convince people of things is to be eloquent - to have some bag of tricks for sliding conclusions into their brains. But there’s also a sort of hack that you can use if you are not a very good salesman, which is simply tell people the truth… Another advantage of telling the truth is that you don’t have to remember what you’ve said. You don’t have to keep any state in your head. It’s purely functional business strategy (Hackers will get what I mean).

Indeed we do. Our strategy is to plainly state what a JumpBox does, how we think it can help people and answer their questions as openly as possible using our public forums. We’re admittedly horrible sales people which is why we shifted from our original plan of selling the toolset to enterprises, a path that would have demanded considerable salesmanship. The idea of being anything but straightforward with your offering in a startup selling to consumers who all have blogs of their own and don’t hesitate to digg scandalous stories for attention seems suicidal- you’d get publicly crucified as a company if you tried that. And companies do all the time.

Yahoo purchased Viaweb in June of 1998 for $49MM in stock. The Viaweb software became the core of the Yahoo Store offering which is still offered today.

The original vision

Founders Mike Ramsay and Jim Jermoluk had grand visions of a home media server that would handle one’s music and video assets for home entertainment. As they set out to create this product they realized it was an extremely ambitious undertaking and they stood a better chance of success if they could scale things back to a single component of the home media server. They finally settled on the killer app of a PVR (”personal video recorder” - their term for the DVR). Having raised angel investment under the original premise they had to go back and explain to their investors why they were stripping things down to a simpler offering. The investors were skeptical because it sounded like a glorified VCR but eventually acquiesced under Ramsay’s persuasion that users would be able to do things like pause live TV and record one show while watching another. TiVo was born and they started hiring engineers to build their product.

The challenges

TiVo was able to attract brilliant engineers early on because what they were doing had never been done before and was a difficult and intriguing problem to solve: making a rugged, cheap consumer device that could do video storage and manipulation and had a dead-simple user interface to record television programs of interest. Ramsay explains the two breakthrough innovations that made it valuable: “How to do simultaneous record and playback, pause and fast-forward and rewind and stuff like that cheaply and efficiently was the key attribute. In fact, that idea of how you implemented that through a device called a media switch and sort of managed all that flow of data - that became part of the Time Warp patent, which was one of our most important patents.”

The second breakthrough was the ability to expose local program guide information to the box in a way that you could reliably record the programs you wanted and not miss the beginning or end of the shows. With 65,000 different program schedules due to all the different permutations of zip codes, cable companies and service flavors, it was a tricky proposition to reliably deliver the program guide information. They had to embed a modem in the unit and have it dial back to a central server to retrieve the program schedule. As challenging as these problems were to solve, once they mastered it they had a significant barrier to competitors.

They also learned a lesson while the product was still young that they needed to have a “safe mode” in which a unit that had been corrupted with a bad software update could automatically “phone home” and re-download a fresh install of the software and reformat itself. They had an incident where a handful of units in the field went bad and they had to manually get things working again. Had this occurred after there were five or even one million units in the field it would have been a catastrophe and most likely meant the end of the company.

Show me the money

TiVo raised more money in funding than most of the companies in the FAW series made on their exit. Over the course of multiple funding rounds with VC’s and a massive $200MM round from AOL, TiVo raised half a billion dollars in funding. Ramsay said, “I would say that one of the reasons that TiVo is thriving today is that we were well-capitalized. We were able to power our way through the downturn- that early 2000 period when Replay went away. We were capitalized enough that we knew we could ride through it. While we had to make a few adjustments to the company, there was never a question that we were going to survive. We knew we were going to survive.”

Not that JumpBox is extraordinarily well-funded at this point but we do finally have a comfortable buffer of money in the bank and I would agree that this takes a huge psychological burden off the people involved. In the very early days we were skating on a home equity line I had done on my house moving small chunks of money each time we had a payroll. Each payday hurt as you’d see your debt going up. Having investors put their belief in you and share some of the risk does miracles for your mental well being and also helps validate what you’re doing to others when it’s difficult to convey the value of a technology they don’t yet understand.

David and Golliath

The press ate up the notion that a tiny tech startup was turning the revenue model of large media conglomerates on its ear by empowering consumers with the ability to skip commercials. They also liked the head-to-head combat of Replay vs. TiVo. The press-worthiness of these two aspects earned them a lot of attention. Even the bad reviews they would get from some of the more demanding technical folks were still added visibility. The “Purple Cow” hook that captivated everyone was the ability to pause live TV. TiVo cleverly packaged the capability of its unit in a fashion that dazzled consumers- this marketing genius was critical to its success.

You don’t have to outrun the bear…

There’s a relevant parable (albeit a bit gruesome) of two campers in the woods who are preparing to go to sleep when they hear the noises of a bear rummaging around their campsite. One camper freezes and exclaims “what are we going to do?” while the other calmly puts his running shoes on. “Don’t be silly, you can’t outrun a bear!” says the freaked-out camper. “I don’t need to outrun the bear,” says the calm guy. The bear in the TiVo story was Hollywood and the skittish camper was Replay TV.

Replay and TiVo were locked in a head-to-head battle for delivering essentially the same product. The reason Replay failed and TiVo ultimately succeeded was that Reply was bear food for Hollywood while TiVo skillfully navigated the boundary of what they could get away with in terms of commercial-skipping technology. Replay built in an automated method to skip commercials while TiVo knew that this would infuriate the companies that depended on selling TV commercials. TiVo kept it to a manual “30sec skip” feature. Replay became the “bait” and the networks punished them with lawsuits. Ramsay said, “Replay probably did us a fabulous favor when they stepped across the line. There’s a line in the sand that those media companies think about. You don’t know where it is, but if you step over it, they’re going to get you. Replay stepped over it by doing automatic commercial skipping. You didn’t even have to fast-forward through the commercials. They just found out where they were and they eliminated them. And they let you share programs over the Internet. That crossed the line. They got sued. They were the bad guy; therefore, we were the good guy.”

Tivo ultimately made friends with the networks and took investment from Disney, Viacom, Discovery Communications, NBC, Showtime, HBO and Time Warner. TiVo IPO’d in 1999 and continues to hold the dominant position of selling the world’s most popular DVR.

What made WebTV possible

The patented technology that was at the core of WebTV’s offering was the ability to display de-interlace the video feed and send a progressive scan signal to a television set. Steve Perlman had learned a lot about working with video signals during his time at Apple. He was the one responsible for developing the technology that brought color to the Macintosh as well as the Quicktime video format. After he left Apple, Perlman did a startup that developed a modem to link Sega game systems over a phone line so players could play each other remotely. The technology worked but the business was a lukewarm success and after a few years of exhaustive work, Perlman was looking for the next thing. He discovered the web and saw an impedance mismatch with the technical barrier of needing a computer to access web sites. He set out with the vision to make the web accessible to non-technical folks via TV.

We saw the same mismatch with Open Source. You have this vast pool of powerful applications out there but the great majority require technical knowledge in order to use them. Like the web sites full of valuable content that were inaccessible to 95% of the population back in ‘97, we see Open Source as a largely-inaccessible territory for many non-technical people. JumpBox aims to do for Open Source what WebTV did for the Internet- expose it via a simple interface so that non-technical people can appreciate the benefits of these applications which have previously only been available to computer-savvy people.

It was 1995 and Perlman went on one of his three-day non-stop working binges and developed a prototype that could display the visual output from a computer on a television. He showed it to his buddy Bruce Leak, who was a former colleague from Apple, and they knew they were onto something big. They had ties to a money guy named Marvin Davis, an investor for whom they had made a ton of money on a previous startup called Catapult Entertainment. They contacted Marvin and raised $1.5MM and went out and secured office space in an old BMW dealership in Palo Alto. It was a classic scene trying to order bandwidth - their ISP thought it was a practical joke because there was no reason a car dealership needed the kind of bandwidth they were asking for. They started hiring, building out the product and courting manufacturers like Sony and Phillips.

“You go first”

With a team of thirty employees they had a significant burn rate and got squeezed when the second half of their funding was withheld out of skepticism. They had a letter of intent from Sony on a 1yr exclusive manufacturing contract and all the investors they spoke with were interested but nobody wanted to take the first risk. Perlman drained his personal bank account, liquidated all his assets and went all in to keep the company breathing during this time. They were days away from completing a deal with a semiconductor company for investment and were getting bullied with a clause in the contract that would have forced them to exclusively source the chips via the semiconductor. It was a frustrating arrangement and Perlman made the decision to walk because that arrangement would have locked them into a single provider and surrendered the ability to get competitive pricing for a core component of their offering.

Talks with VC’s weren’t going any better. They were skeptical on the adoption of the technology- they felt people wouldn’t want to interact with their TV in this way. Perlman said, “We found that nobody was willing to make that first step. In fact, I think a lot of them were sort of like vultures waiting for us to fail, and then pick up the pieces- because they saw the value of what we were doing for a bargain.” They finally found a VC firm that was prepared to go in for $4.5MM but days before the deal was to close, they received a letter from Sony saying the deal was off. Given that it was 1yr exclusive arrangement WebTV hadn’t pursued any other manufacturers. It was an ugly day and they had to disclose this development to the VC, Jeff Brody. He could have walked away from the deal but instead said, “I believe in you guys, and I think this is going to make it. We’ll still go forward on the same terms.” That ounce of faith was a pivotal moment for the company - another VC followed suit and matched the investment to make a $9MM round.

The midnight run

Things began to fall into place after that. Phillips came back and wanted to do a deal. WebTV hired a consultant who was a personal friend of the CEO for Sony and he revived talks with them to determine why the deal had fallen through. Sony immediately dispatched their CTO to the WebTV offices and suddenly they had to scramble to get the unit in a working condition. The way they developed they had to compile a build and the version they had was completely untested at the time. With only 2 1/2 hrs until the Sony CTO landed they had no choice but to roll the dice with the version they had and hope it worked. Fortunately it happened to be a good build and the demo went off flawlessly. It earned them a subsequent demo with the CEO of Sony and they won him over by doing some last-minute development to make a version of WebTV that could display Japanese web pages.

They inked a deal with Sony but couldn’t promise exclusivity because they were already working with Philips. So the situation actually worked in their favor in the end giving them distribution through two powerful electronics manufacturers. They went back and raised a series C round for $35MM and cranked out a product. Nearly one year to the day of their first angel investment they were selling a working product. The WebTV product solved network challenges of having presence of a local ISP in every city, translating the http content to a format that was viewable on a smaller screen, making an interface that catered to non-technical users and fixing the video signal to display properly on a TV - there was a ton of technical magic that went into delivering the experience.

Politics

In April of 1997 Microsoft acquired WebTV for $425MM in cash and stock. It was arguably a defensive play to ensure that nobody else cornered the market on the living room web surfing experience. They mandated some asinine decisions about not supporting competing technologies like Java and Real Player. Perlman was focused on providing the optimal experience to the end user and found himself against a wall facing idiotic defensive technology decisions from MS. Perlman said, “I began to get to know the other top executives at Microsoft, they were talking about negotiating this and that funding, and cutting back products to the point where they no longer made sense. I said, ‘Look, can’t we all agree on what is the right objective for the whole company and then fund that? I don’t care if it’s in your group or my group or whatever, but we should do the right thing.’ It didn’t work that way and, of course, any big company is like this. People have certain things that they control. That’s why there’s politics in large companies.”

On perseverance

Like every one of the FAW stories, Perlman’s team went through a tunnel of extreme adversity and ultimately emerged triumphant on the other side with a popular product. When asked what he could attribute their perseverance to he explained that the core trio of founders in the company held it together. “The key thing about Phil and Bruce is that they had hearts of gold. They were nice people. They were not in it to get rich. I mean, money certainly is freedom. But they both had a vision of creating something great that people would love. That attitude from the three of us permeated the rest of the organization…And through those really difficult things that I told you about, where we were running out of money, when we got the certified letter of rejection from Sony- all those different things I think would have easily knocked over a weaker triad.”

The WebTV company was profitable in its eighteenth month and subsequently every month ever since. It is now known as MSNTV and has grossed over $1.3BN in it’s eight year lifespan.