Play #1: Have a Big Dream

I saw an opportunity to deliver business software applications in a new way. My vision was to make software easier to purchase, simpler to use, and more democratic without the complexities of installation, maintenance, and constant upgrades. Rather than selling multimillion-dollar CD-ROM software packages that took six to eighteen months for companies to install and required hefty investments in hardware and networking, we would sell Software-as-a-Service through a model known as cloud computing. Companies could pay per-user, per-month fees for the services they used, and those services would be delivered to them immediately via the Internet, in the cloud.

If we hosted it ourselves and used the Internet as a delivery platform, customers wouldn't have to shut down their operations as their programs were installed. The software would be on a Web site that they could access from any device anywhere in the world, 24/7. This model made software similar to a utility, akin to paying a monthly electric bill. Why couldn't customers pay a monthly bill for a service that would run business applications whenever and wherever?

This delivery model seems so obvious now. Today we call it on-demand, Software-as-a-Service (SaaS), multitenant (shared infrastructure), or cloud computing. In fact, Nicholas Carr, former executive editor of the Harvard Business Review and one of the most influential thinkers in the IT industry, has since written two best-selling books validating this idea. Carr has even suggested that “utility-supplied” computing will have economic and social impacts as profound as the ones that took place one hundred years ago, when companies “stopped generating their own power with steam engines and dynamos and plugged into the newly built electric grid.”

The industry has come a long way, but consider that when we started, we didn't have these industry supporters, or even these words, to describe the computing revolution we believed was beginning. Although there was yet to be any kind of SaaS industry, I believed that all software would eventually be delivered in the cloud. I would soon find that in order to pursue my dream, I had to believe in it passionately and be ready to constantly defend it. This lesson learned during our earliest days still guides us today.

Play #2: Pursue Top Talent as If Your Success Depended on It

I met with Parker Harris as soon as possible. “So, are you guys good?” I asked.

“We're some of the best people you'll find in the Valley.”

I liked that confidence, especially considering that it was bolstered by what I had already heard. Still, I prepared myself for a very short meeting. Although Parker seemed like a promising technical candidate, I wasn't sure that this was the next move he had envisioned for himself. I'd heard that Parker had recently returned from a six-week trek in Nepal and told his business partners that he wanted to do something more meaningful than helping salespeople sell more. I was concerned that Parker would be fundamentally opposed to SFA and that he would think it boring because he had done it before.

I also thought that enterprise software was boring, but my vision was to do something much bigger. My vision was “the end of the software business and technology models” as we knew it. I believed that this was a great story and would appeal to Parker, who had majored in English literature at Middlebury College. Building this service also provided an intellectual challenge inasmuch as it had to be highly scalable, reliable, and secure; the service had to be something every customer could use simultaneously. I knew that the scaling test would be compelling to any great developer. I also had a trump card: Parker wanted to be in San Francisco. Every day, he endured a long commute from his house in the city to the Saba offices in Redwood Shores. “I have the same problem,” I told him. “Salesforce.com will be in the city.”

Parker was sold, but he had to get his business partners, especially the more pessimistic Dave Moellenhoff, to see the light.

Play #3: Define Your Values and Culture Up Front

On March 8, 1999, Parker Harris, Frank Dominguez, and Dave Moellenhoff began working in a one-bedroom apartment I'd rented at 1449 Montgomery, next door to my house. We didn't have office furniture, so we used card tables and folding chairs. What we lacked in furnishings, we made up for with an amazing view of the San Francisco Bay Bridge. I hung a picture of the Dalai Lama over the fireplace and another of Albert Einstein on the wall. Both were part of Apple's new ad campaign, and each said, “Think Different.”

My summers at Apple had taught me that the secret to encouraging creativity and producing the best possible product was to keep people fulfilled and happy. I wanted the people who built salesforce.com to be inspired and to feel valued.

That wasn't to say there was anything glamorous about those early days. (The original server room was the bedroom closet, which also held Frank's clothes because he was flying down from Portland for the workweek and sleeping on a futon under his desk.) We built a culture simply by doing what we enjoyed. We wore Hawaiian shirts to instill the aloha spirit in the company. We ate late breakfasts at one of my favorite restaurants, Mama's, just down the street on Washington Square. Dave brought his dog to work. I got a dog too, a golden retriever named Koa, who also joined us in the office and soon got promoted to CLO (chief love officer).

Play #4: Work Only on What Is Important

We built the first prototype within a month. It didn't take very long because the developers knew sales force automation from their previous experiences. Further, it was a lot easier to build a Web site than to create complicated enterprise software. Our overarching goal was, as the developers said, to “do it fast, simple, and right the first time.” The user interface was bare bones almost to a fault, but we wanted the service to be extremely easy to use. It had only the necessary information fields, such as contacts, accounts, and opportunities, which were initially organized by green tabs at the top of the screen. “No fluff,” one of our first developers, Paul Nakada, used to say. Exactly like Amazon, I thought.

Our focus was directed at developing the best possible and easiest to use product, and this is where we invested our time. Realize that you won't be able to bring the same focus to everything in the beginning. There won't be enough people or enough hours in the day. So focus on the 20 percent that makes 80 percent of the difference.

Play #5: Listen to Your Prospective Customers

I invited friends and colleagues to visit the apartment, which I called the Laboratory, and asked them to test the prototype and offer feedback. Michelle Pohndorf Forbes, a family friend who was in sales, was one of the first people we invited to cycle through the prototype. She constantly reminded us to make the site easy to navigate with as few clicks as possible. My friends who worked at Cisco shared everything they hated about using traditional enterprise software products, and they walked us through what wasn't working for them. We listened and then responded by designing salesforce.com to be all the things that traditional software wasn't.

Unlike the way software had traditionally been developed — in secret — everyone was welcome at the Laboratory. When a group of Japanese businessmen were in town, they came to see what we were creating. We eventually became a stop on a tour for visiting Korean businesspeople who were interested in seeing an American start-up. Being inclusive of potential users from large and small companies across the world helped us gain valuable insight. After all, our goal was to build something that could serve as a global CRM solution for the masses.

In addition to asking dozens of people to cycle through the application, we hired Usability Sciences in Texas to test the product. The company provided feedback and videotapes of people using the site so that we could see what else needed tweaking. One problem we discovered, for example, was that our “create a new account” button was in the wrong place. It was on the right-hand side, and it disappeared on some monitors. By simply moving it to the left side, mirroring the way people read, we saw a huge improvement in the way people used the site. This experience proved the value of involving prospective users in order to build a user interface that was intuitive.

Play #6: Defy Convention

Asking users for feedback so that you can fine-tune a product or service to their needs is common sense. Yet this practice was completely counterintuitive to the way the software industry worked. Don't be afraid to ignore rules of your industry that have become obsolete or that defy common sense.

Creating an attractive user interface that people enjoy using is the key to building a truly great product. This seems so obvious, but it wasn't the way in which software design was customarily approached.

Steve Jobs was the master of building computer products that create customer excitement and loyalty. It's also no coincidence that his products look like nothing else out there. Think differently in everything you do.

Play #7: Think Bigger

In summer 1999 we had ten employees and a two-page Web site (a home page and a recruiting page requesting that resumes be sent to cooljobs@salesforce.com). Everywhere we looked, Internet companies were growing wildly, and financial deals were heating up. We were constantly talking about the deals of the past few years, such as Hotmail, which had sold for $400 million.

“That's a lot of money, Marc. Don't you think that's a lot of money?” asked Frank Dominguez, one of the salesforce.com developers and cofounders, referring to the Hotmail deal.

“No, I would never sell for that. They left a lot of money on the table,” I replied. Frank couldn't believe that I could think so big when we were still so small. Although the other founders were initially leery about our move to the Rincon Center, they quickly grew to like the new larger space. They drove golf balls down the length of the office and flew remote-control helium blimps. We had no office furniture, so we put tables by the outlets that were already there. Everyone had to set up his or her own desks (we bought sawhorses and doors at Home Depot), and employees received their computers in boxes and put them together themselves. It was an archetypal California start-up scene with a dog in the office and a mass of young and energetic people wearing Hawaiian shirts, working hard, and subsisting on pretzels, Red Vines licorice, and beef jerky.

In typical dot-com style, we exploded. By the time co-founder Dave Moellenhoff returned from his three-week honeymoon in November 1999, the staff had doubled. As I had promised Parker, about one year after we moved into the Rincon Center, we were bursting out of the space. Three salespeople had desks in a hallway, and five IT specialists had taken over the conference room. Our next move, in November 2000, was to shiny new offices at One Market Street. It was only a block away, so we put the servers on office chairs and rolled them across the street. Although we were not going any great distance geographically, the leap ushered in an entirely new era for our company.

These are excerpts from Marc Benioff's book "Behind the Cloud."