By 2006 we were all pretty serious about the “love your help desk” idea, but we were still keeping our consulting gigs as a way to pay our bills and fund this idea. I was doing some business and IT process consultancy on the side. Alex was involved in a bunch of different projects. And Morten was still doing a lot of security work.
We had officially moved “headquarters” from my apartment, which was being taken over by my kids and family life, to Alex's loft, a sunny space with carefully curated pieces. Alex was more comfortable in his own cave, and, like most Danes, took great care in creating his space. It was an old loft that had been remodeled just a few years earlier. New, fabulous windows framed views over the old roofs of Copenhagen. But it was also very much a bachelor pad. It had a semi-functional kitchen with more drinks and condiments than actual food.
The loft was in the old part of Copenhagen, which was turning hip but still had a pregentrified feel. A gay S&M bar was across the street. Every day I'd walk up five flights of rickety uneven stairs to get to work. And though it was a beautiful apartment, you wouldn't ever call it ideal office conditions.
Although Alex had selected Scandinavian furniture that looked clean and modern, it was definitely a setup for one guy only. There was only one real work chair. Whoever got in first (Alex) took it. The others sat on old kitchen chairs. The desk we worked at was an old door on sawhorses. (I'm not sure why. I know Amazon and salesforce.com started with desks built out of doors, but we were unaware that this was a trend happening across the pond; it was more of a utilitarian decision.)
Working together in the small loft, we quickly learned each other's quirks and foibles. It's so intimate to be working in someone's home. You see everything.
Alex, a complete workaholic and night owl, frequently stayed up until after 2 a.m. working. I would ring the bell the next morning with my coffee in hand, ready to start the workday at 9 a.m. I had little kids, so I'd already been up for several hours. Often Alex answered the door in his underwear, hair disheveled. Sometimes he'd go back to bed. Later, we had keys and let ourselves in, only to find Alex still in the bathtub. He'd emerge into the “office” in his robe, his head wrapped in a towel.
We had a lot of respect for one other, but spending eight or more hours a day together, you do get annoyed by each other's mannerisms—you find you need a break. Alex was always pointing out the way I moved my chair or the way I tapped my fingers. He chastised me for clearing my throat. And it sometimes seemed that Morten was annoyed by everything.
Copenhagen is amazing in the summer—the sun stays up all night—but the winters are abysmal. It is still dark when you wake up, and the sun disappears again in the afternoon. And in between, it's gray, windy, and wet. A lot of people in Scandinavia struggle with winter depression. Of the three of us, Morten was probably the most sensitive to these winters, and the uncertainty of our pursuit, in combination with this bleariness, sometimes got to him. There were periods when he disappeared, distancing himself from the stress of building this company, tuning us out. It became his way of recharging.
I knew from years of working with Morten that he was his own worst critic, always being too hard on himself, and always trying to make everything perfect. Anyone who meets Morten recognizes him as good-looking and charming. Women love him. He's had so many great girlfriends, nice girls, beautiful women—I've probably had a crush on every single one of them—but he's never decided to commit to taking the next step. No one is ever perfect.
But while I see this as a challenge in his personal life—the obsession over perfection, finding the right apartment, the perfect girlfriend, what car to buy—it was an incredible benefit for our software. From the very beginning, Morten led the effort to build the infrastructure and tech stack in a way that obsessed over every detail and considered every possibility. Our stack is pristine. It's perfect, and that is what has enabled us to scale. In some ways Morten's nose-down personal journey toward perfection translated into our strength.
We were all on different schedules, and there were times we were together only on nights and weekends, but on the weekdays only if we happened to be off our other jobs at the same time. This made for a slow way to launch a company. We worked on building the product for eighteen months.
Amazingly, no larger industry innovation happened at all in that time. That was mainly a testament to the state of the customer support industry. No one was interested in it. It actually made us paranoid, because no one cared. We wondered if we were missing something.
Because it took us so long to build the service, it took us equally long to make any revenue. We financed our lives and the development of the product through our individual consultancy gigs. The result: we were running out of money all the time (both personally and at the company).
This was trying for all of us. I had a burgeoning family—Mie was pregnant again, and I would soon have three kids and little income. It was expensive to set up a family. I maxed out credit cards and spent my retirement money to finance this period of my life. (I did not tell Mie.) To finance initial operations, I also took out a $50,000 line of credit for which I was personally liable. (Again, I didn't tell Mie.) There was a point when I was going to be out of cash in two weeks. (Definitely did not tell Mie.)
Alex had his own financial woes, due to taxes owed on estimated income that was never realized from his previous company, Araneum, which had folded and therefore never earned what he expected. Morten, caught up in the real estate bubble, also had monetary concerns. He had left a high-income job for one with no pay and at the same time was trying to pay a mortgage on an apartment that was under water thanks to the housing crisis. He talked with urgency about the importance of paying off this mortgage as fast as possible. It may have been self-imposed pressure, but to him it was very real and very pressing.
We had initially agreed that we would forgo any sort of salary while we were building the product, but at a certain point Morten decided he couldn't continue this way. He needed the certainty of having a regular salary, and although it was an extremely modest salary, it was still a lot for a company with no money. We weren't crazy about Morten's attitude. After all, we were all in the same position. Alex, who was always focused on things being equitable, was probably especially perturbed. I didn't want to think about it too much. It was an annoying discussion to have, and impossibly untimely, considering we didn't have any cash flow, but I knew we had to deal with it to keep Morten focused and undistracted.
The thing about money is, it's happening in your head. Everyone processes it differently. People can live with different risk profiles. You have to accept that and not let that come between you. You have to be realistic.
When you are building something, ultimately you have to make all sorts of sacrifices to accommodate people's needs. Morten needed certainty about a salary; he had an apartment he wanted to pay off. I needed to balance work and home life, and I'm sure that at times my absence was more tolerated than appreciated. Alex needed artistic freedom and to feel there was a good upside for him, which required patience on our part as he meandered through his process of determining that everything was fair and square.
But because of our different personalities and approaches, we were able to contribute different pieces that were all of value. We needed each other: Morten on the back end, with his talent for engineering and drive for perfection; Alex on the front end, with his sixth sense for product and attention to detail; and me with my understanding of the customer support domain and an impatience to get things done. We were different, but in a way we were like the three necessary legs of a tripod—dependent on one another and strong only if we were together.
It was difficult to stay committed to working together for no pay when there were so many temptations along the way. At one particularly precarious point I was offered a consulting gig that paid well. It was safe, and it was even interesting, but it would have taken six months out of my life. I couldn't do that. I wasn't ready to go back to the old way of working forever and turn my back on this opportunity—maybe my last one before it was too late. My personal responsibilities were only growing. And I was turning thirty-six. It felt like it was now or never.
I passed on that job, and though I tend not to regret things, and I never look back, I had to admit it made things hard in the moment. I had to continue balancing gigs to make money and pursue this company. It was challenging to be supporting a family with this precariousness. There was no predictability at all.
Mie was always supportive, but the truth is I was never super transparent about the state of our finances. I tried to keep the family sheltered so they wouldn't feel the impact. Admittedly, it was easier to do this in Denmark, which has a huge social welfare system; daycare and school are free. There's a safety net that cushions you and gives you the opportunity to live closer to the edge than you might feel comfortable with in a different economy. Denmark is not like the United States, where you can quickly go from living a decent life to destitution. There are no homeless people in Denmark. Jobless, but not homeless.
So although I knew we would never be on the street, I was always aware of our finances and always thinking ahead, trying to figure out how to make ends meet. My life was like a chess match; I was always mapping out the next consultancy move to enable me to survive another round.
An unpredictable financial state is not really something you talk about when you get together with friends for a social evening, so I mostly kept quiet about this, but with Alex and Morten I could be open. I remember telling them I had only a few more weeks before I'd have to go out and make some money. “Let me know if you hear about anything,” I told them.
They were in the same situation. And both Alex and Morten had plenty of opportunities to consider, including some that offered them ownership and real money. One of Morten's offers would pay $30,000 a month, solving his financial woes. With so many temptations—and such a difficult reality—keeping everyone committed was a nightmare. This was the biggest challenge of doing a startup—and, I later learned, the point where most of them fail.
Truly, it is a miracle that we stayed together and remained focused on an effort where everything was an unknown and there was no stability and no salary. Somehow I think it's luck that we all loved our baby enough to not give up. But it could just as well have gone south so many times.
Although we had built businesses before, that didn't make it any easier in terms of coming up with a formula to do it right. There was no formula. Our first focus was on the product—and on thinking about how to do it differently. All we knew was that it had to be lean and elegant.
Curiously enough, we didn't think about much else—even the basics that would have been in an instruction manual. (There was no instruction manual!) We never thought about “call center systems.” We never thought about a “target market.” We never thought about competition. All we thought about was trying a different approach to existing customer service software. There was no business plan. There was no plan.
For a long time, we didn't have a name for our project. Morten had originally baptized it “Project Eisenhut,” which is German for Iron Hat. A truly terrible name, and none of us remember its genesis. Later we talked about Help Desk Factory, which Alex hated very deeply. One of the smarter names we came up with was Tagtix. We wanted to convey how we had replaced traditional categorization with tagging. And it was all about tickets. Get it? Tagging tickets. Tagtix. Too smart and too lame.
Finally we settled on Thank You Machine. It was Morten's invention, and we all agreed it was super cool. We originally used it together with a famous Internet meme photo that is a mashup of President's Truman's visit to a nuclear submarine and a lot of old-school IT equipment. It was so hipster smart.
About one month before launch, Alex said, “Let's find a real name. Thank You Machine will not work. It's too weird. We need something clean and simple.” Alex forced us to consider other options, which started yet another discussion.
We argued forever. We needed alternatives. At the same time, we had had so many alternatives! The real problem was making a decision. Every name sounded weird. It's like thinking of a name for your child—you don't know what kind of person she will be, and a person's name carries such weight.
Ultimately, we wanted something that showed we were a customer service desk but that also signaled something different. Our philosophy was to be elegant and bring peace of mind to customer support in an enlightened way. Zen was enlightenment—Zendesk.
Of course, the first time I heard it I thought it was awful. Again, we argued forever. Then we just made a decision and went with it. Later we discovered we could build a whole story and universe around the name, and it started to make sense. The fact that the domain name could be acquired for $1,000 was also a factor—we were on tight budget. (The domain name had previously belonged to a website in Korea. We never learned what the original Zendesk in Korea did, but we did receive a lot of its mail.)
We understood how important it was to build a brand identity around the name. Back when we were still Thank You Machine, Alex had asked Toke Nygaard, a former Araneum employee and an epically talented designer, for help with the logo. Toke worked with a Swedish designer, and they created the first version over the weekend. It looked like a cog in a wheel, a nod to the machine in Thank You Machine, with a heart in the center. It conveyed what we needed.
The logo made perfect sense when we were Thank You Machine, but now we needed to make it more Zen. We reshaped the cog in the wheel to be a heart in a lotus, to better fit our name. With that, we really started to establish our identity. Everything began to come together.
At a later point Alex talked to Toke about having an illustration for Zendesk. The last thing he wanted was the traditional image of a smiling woman with a headset. Toke came up with this fantastic image that we still use today. Inspired by the Laughing Buddha (or Budai, if you want to be technical), he created something that we call the Mentor. He's a big, bright, fun-loving, smiling character in a yoga pose—wearing a headset!
Right before we launched in late 2007, we applied online to be part of the TechCrunch20. This award, which became the TechCrunch40, and later the TechCrunch50, identified the hottest startups worldwide with unreleased products ready to launch.
Amazingly, we made the top one hundred. We were ecstatic. Some seven hundred companies had applied, and now we were among the final one hundred selected for additional interviews and due diligence.
We couldn't believe it. We read TechCrunch religiously and saw it as the gravitational center of the startup world. We were intimidated by its influence; we had this impression of a large, well-oiled, and super-professional organization. (Of course, we were still working in a kitchen and still subsisting on Alex's only recipe—for spaghetti.)
We tried to schedule meetings with the contest judges, only to have it degenerate into confusion. TechCrunch scheduled our interview during a two-hour window in the middle of the night in Copenhagen. We stayed up (easier for nocturnal Alex than it was for me) and called in to the conference bridge, only to find no one there. We rescheduled, and the same thing happened again.
And then, one night, a night when no time had been blocked on the calendar, TechCrunch called us without any warning. We were available, although that is not to say we were ready. Jason Calacanis—founder of Weblogs, Inc. (which was acquired by AOL), a cofounder of the TechCrunch50 conference, and an entrepreneur extraordinaire—was on the line, firing away questions. He was very polite, but not what I would call passionate about our product. The conversation was relatively short. And in the end, Jason summed it up like this:
“Well, so basically you've built a help desk with a blog and some RSS.”
Ouch! That hurt. While that was one of our capabilities—you could subscribe to articles and tickets via RSS—that was definitely not the whole story. I knew what we were doing was not sexy. I understood that the help desk industry wasn't exactly hot. I got it; it was boring. But although we thought there was room and reason for change in this tired area, TechCrunch didn't see the potential of our offering the way we did: as the must-have, cost-effective, and convenient help desk alternative for thousands of organizations all over the world.
In truth, we probably did a poor job conveying the big idea. We found it hard to pitch our customer support system. We thought that it was great, but we weren't ready to say it would change the world. How could we? Of course it wasn't changing the world! (In Silicon Valley everything is about changing the world; at the time I did not know this.)
If you weren't using the service—and customers weren't using it yet, as this was prelaunch—it was hard to make someone imagine the kind of success that could be had on it. We thought that in order to appreciate it you really needed to experience the pain we were trying to solve for. And admittedly, for an outsider who had zero interest in help desk improvement, this was not the most compelling company in technology. This was not Twitter, something we could all use and relate to, and it was not Google, which so obviously makes life easier. We just didn't have that same kind of broad appeal. It took many years before enterprise software became sexy again, but it did ultimately enjoy a comeback, and, ironically, in 2013 we won the TechCrunch award for sexiest enterprise startup.
It wasn't that surprising when, some time shortly after that middle-of-the-night interview, we received an email informing us that we had not made the final cut. Though we didn't think Jason Calacanis “got us” during that awkward call, we also had to admit that we would have to hone our pitch. We were not naturally great at promoting our product. Our plan was to let the product speak for itself. Now we had to hope that it would be much better at selling itself than we were at selling it. And we would also have to improve our story, just to be safe.
We learned something important from that nixing, and even though we might have made a costly mistake, this contest was an incredible coup for us. It created an initial bit of awareness about us. It was our first interaction with TechCrunch—maker of all startups—and now we knew its key players: Jason Calacanis, CEO Heather Harde, and founder and editor-in-chief Michael Arrington. We lost, but at the same time we sort of felt as if we had won.
After much discussion, and number crunching, we decided that once we launched the service we would quit our other gigs and focus on Zendesk full time. We would go all in.
And finally, eighteen months after starting to work on this idea, it happened.
On October 1, 2007, Morten sent me a note: “We're live.”
I immediately sent it out to hundreds of people.
“I meant to say, check this out,” said Morten, a bit taken aback that the site was now out of our control.
“Welcome to the rest of your life,” I said.
I don't think he found it funny.
We needed to create use cases to demonstrate how the software worked, and we built product demo sites and presentations that potential customers could see on the website. We took it seriously; we wanted it to show what we could do. But we didn't take it too seriously.
In one example, we named the “demo” company Talladega, and all of its employees were characters from the Will Ferrell movie Talladega Nights: The Ballad of Ricky Bobby. (Relatively inappropriate.) Another company demo site was called Glengarry. (From the Glengarry Glen Ross movie; I guess we were subconsciously focused on sales.) We needed pictures of people to go with these names, so we used random photos we found on the Internet to represent these pretend users, with names inspired by the respective movies. I picked up these photos online, and I no idea who these people were. But sometimes others did. In one case, someone wrote us, “Why do you have a picture of Georgina Beyer, the transsexual mayor of Carterton, New Zealand, on your website?” And whoa, that was a good question. I don't think he ever got an answer from me. I couldn't really tell him that Georgina Beyer was a female version of George Aaronow, a character from Glengarry Glen Ross, and that it was just a photo that came up first on a Google Images search for that name.
Whereas enterprise software was sold through a tedious process that was extremely costly and complicated, our model was extremely basic, functioning just like the consumer software model. The idea was to make it as seamless as possible—we would deliver a service that immediately provided 80 percent of what a support person needed to do (in contrast to traditional software—out of the gate, it would not do anything they needed), tailoring it to their needs and immediately getting them hooked on the product.
In what was an unusual model at the time, everything was self-service; there were no salespeople. Prospective customers went to the website, signed up, did a free trial, and, if they wanted it, bought the product with their credit card. It was like downloading an app from the App Store (although that didn't arrive until 2008). It was so straightforward and provided instant gratification.
By now you have heard of this model; it is a routine practice today. But when we started it was not common for enterprise software. I didn't know it was a real business model. In fact, I didn't think we had a business model!
Today these ideas of cloud computing and Software-as-a-Service have emerged as proven business models—preferred business models—but at the time I never knew how we'd keep customers from one day to the next. The metrics that SaaS companies use to track retention and expansion didn't exist yet. With nothing to reference, it felt like we were charting new territory, which was fascinating and terrifying at the same time.
Our customer acquisition “plan” went something like this: “Just make customers happy.” It was all about just giving our customers and prospects a great, smooth experience trying and using the product. Unlike the situation many new businesses encounter, we weren't moving customers away from an existing vendor in most of the cases. We may originally have thought differently about the market, but most of our initial customers were greenfield opportunities. Simply by existing and by introducing a new easy model, we became their tool of preference.
And for these folks, there was no existing vendor. There wasn't anything to replace. No one served the customers we were going after. A lot of them were online businesses, web apps, e-commerce sites, or just forward-looking businesses that saw Zendesk as a much easier way of organizing support.
We won quite a few customers almost immediately, and they found us from all over the world. The very first was an Irish company, Cubic Telecom. (It is still a customer. It just got acquired, and it is still using Zendesk.) The second customer was a chain of gas and convenience stores in Texas. We also had a content management company in New Zealand. And of course we had Laughing Squid, the founder of which later became our good friend and introduced us in San Francisco.
These businesses needed a service like ours, and although we originally didn't see it this way, we soon realized that we were part of a new generation of tools for a new generation of businesses on the Internet.
We didn't have to think about landing the traditional enterprise customer. There was no need to think about the “top five customers to target and win” as the established players did. Instead, we could focus on this low-hanging fruit all over the world.
Customers were everywhere and involved in everything. The question therefore became: how do you suddenly get the attention of a new, widely scattered audience? It certainly wasn't the same way we'd go after a traditional enterprise audience. We had no experience with this, though, and we had to try all sorts of experimental ways to kick up dust on the Internet.
We thought we had a great idea: our aim was to democratize customer service software and make it easy and straightforward by offering it as a subscription. There was no large up-front fee, and using the service was inexpensive and risk-free. It was far less than the many thousands of dollars the traditional products cost, and it didn't require installation or maintenance. We thought everyone would want this.
And they did. But figuring out how to let them know about our service was sometimes an enigma.
The toughest problem was that there wasn't a natural place online that attracted an audience interested in buying a customer support service. It wasn't that the industry didn't exist—there were experts and thought leaders on customer service theory, and there were Gartner reports on the state of the industry—but there was no real community for customer service. At least there wasn't anything that didn't smell like old carpets and cigars. And based on our early customers, we weren't so sure that was the audience we should really target!
With no clue, we took a creative approach and tried a bit of everything. We started buying AdWords. It might have been popular in the United States at that time, but it was a relatively unexplored strategy in Denmark, and we were figuring it all out on our own.
Alex convinced us that we should get space on a specialized niche blog called The Deck (decknetwork.net). It's more commercial now, promoting itself as the ad network for web, design, and creative professionals, but at the time it was a network of all the really great blogs out there, which attracted a creative audience. There was only one little tiny place on each blog post for an ad, which made it look very clean and attractive, and there were all of these interesting blogs attracting viewers. We created artwork and display banners that looked different from traditional help desk promotions, with very simple messaging.
We spent $5,000 for exposure on this network for one month, and we posted a small ad with a different message, like “Help Desk 2.0” or “Love your help desk.”
Within a few weeks of promoting our ads on the Deck we had thousands of people sign-up to try the product. It was a crazy experience. We received so much feedback, and we discovered many different audiences.
Some of our experiments, like The Deck, worked out well. Others did not. At that time, nobody was strategically thinking about search engine optimization (SEO). Perhaps as a result, for a long time the number one search result for “help desk” on Google was a comic strip called The Help Desk.
Boom. Let's go to the top result, we decided. We wanted to win some of that traffic. We figured out the terrible system that allowed us to buy ads on the comic strip site, and we were up and running. We received immense traffic to our website—yet we never converted those clicks to trials or customers. Not one. Luckily that flop didn't cost much money, and we changed our approach, from creating a presence on the most trafficked site to focusing on finding an audience of people interested enough in a customer service support system that they might actually go to our website and use the service.
We tried all sorts of guerilla marketing tactics. I'm not exactly sure how, but in his travels, Alex stumbled upon something called a Buddha Machine. It was a small plastic box the size of a pack of cigarettes. It had a small speaker, a play button, a track selector, and took two AA batteries. It contained eight soothing instrumental tracks that looped perfectly. If you had more than one Buddha Machine and played them alongside each other, the tracks would complement each other in perfect harmony and create new music.
For Zendesk this was the perfect idea for swag. Soothing instrumental music for your customer service operations. Just add Zen and calm and enlightenment to your customer engagement. And it was also somewhat crazy. It was hard to understand and gave tons of ambiguous signals. It turned out the two artists behind the Buddha Machine lived in Hong Kong. We got in touch with them, and they agreed to a Zendesk-branded version of the Buddha Machine that we could hand out at events and to customers.
For a long time that was our only piece of swag. And we built quite an operation out of Hong Kong, from where we would send out the machines to customers and friends all over the world. People either totally loved the concept or were completely perplexed by it. In any case, the folks in Hong Kong spent a lot of time with the swag and with our brand.
It was tricky to navigate this uncharted terrain with undefined customers, but we were lucky to sell to a segment that hadn't been defined up front. There was no obvious way to target this underserved market, but we met this challenge by going very broad. And in that, we found ample, unexpected opportunity.
Although the site launched in English and attracted customers in the United States, Australia, Ireland, Canada, and the United Kingdom, it was not bound only to English speakers. Customers in Brazil, Turkey, and other countries all over the world and spanning all industries were registering at the site. The only thing they had in common was a modern approach to business: they were thinking differently about how they should be interacting with customers and how they could leverage the Internet to do it better and increase customer engagement. These web-based businesses wanted to have a relationship with their customers, but they didn't have the right tools for reaching them.
The existing world of customer service systems was all call center–based and telephony oriented. Everything was done on the phone. But this new generation of companies did not conduct the majority of their interactions on the phone. They didn't want to outsource to a call center and bury customers there. They wanted to chat with them where they lived—on the Web—or get in touch with them by email, but the then-current generation of software just wasn't built to do that.
Our new software helped them set up a multichannel tracking system and also build a community with users. It opened the door to take emails and make sure the senders got a reply. Nothing fell through the cracks. That was different from the way things were at the time: a customer could send an email with an inquiry but could not expect to hear back. We helped make companies all over the world more accessible and responsive.
Soon the service made its way to Silicon Valley—the startup and software epicenter of the world. Scribd started using it. Twitter followed. At the time, Twitter had fewer than twenty employees. It was almost like a little family, which signed on because Crystal, an early Twitter employee, believed we had a shared design and brand philosophy and we thought similarly about how to build an application. It was validating to sign up these companies we read about on TechCrunch. Slowly but surely, things were going in the right direction. We soon had three hundred customers—a number that we were impressed by in 2008. We took bets on when we would have a thousand customers. I thought it would happen soon.
“That is completely ludicrous,” said Morten. When that happened, he promised, “I will run down the streets of Copenhagen naked.” (Today, over fifty thousand customers later, we're still waiting for Morten to streak the streets of Copenhagen.)
We had kept our product under wraps during development—there was no beta testing—so we really didn't know what to expect, but customers pretty instantly showed us that they liked the software. And, even better, now that we were live we started to get a lot of feedback on how to improve it. Users had so many great ideas and comments.
For example, in the very first version we released, customers had to be registered already before they received service. We were influenced by what we had previously experienced. That was how banks and most other organizations worked; you couldn't receive help or advice from a bank without being a customer already. The business relationship had to precede the customer service relationship. But we soon discovered that was not how customer relationships worked for websites and web apps. Often there wasn't any kind of established or existing business relationship at the time someone contacted customer support. The lines between visitor, user, prospect, and customer were blurred.
We realized that the initial relationship had to be created spontaneously. Instead of having a preregistered user, we had to create the user in the system at the first point of contact. Although that sounds pretty basic, it was a big distinction, and it would turn the traditional experience with customer support completely upside down. The delivery, discovery, and learning model was completely different. On the Web, everything was driven at a faster pace.
I look back on this time as the “teenage years,” not just because we were in a young and awkward stage, but also because we really acted like young teenagers. We were opinionated. We believed that we were the center of the universe and that everything we did was unique. Though I didn't know it at the time, I now understand that if we hadn't also had the other, more charming characteristics of teenagers—we were sort of interesting, and fun, and passionate—we would have been unbearable.
During this launch period I thought everything that we did was unique and precious. Only much later did I realize that I was just following in the footsteps of many people before us. Maybe it would have been easier if I had known that all along. Or maybe it would have made it impossible.