In 2004, with the publishing world racing to embrace digital platforms as its ticket to the future, Eileen Gittins proposed a business plan that moved in the opposite direction. She wanted to give ordinary people the ability to design and produce their own bound, bookstore-quality books. Publishers and investors were skeptical. But seven years later, Blurb, the San Francisco–based company that emerged from Gittins's contrarian vision, has grown into a runaway success. The company rang up $45 million in sales in 2009 and has published hundreds of thousands of titles by authors around the world.

My father was born in England, and after World War II, he came to the U.S. as a crew member on the Queen Elizabeth. He worked his way up to become the president of McKesson, the health care company, here in San Francisco, the classic bootstrap immigrant success story. My father would take us out for lunch at restaurants with white tablecloths. He would have a cocktail before lunch, and we would have Shirley Temples. The emphasis in our family was on education, and the expectation was that we'd all become accomplished in whatever field we chose.

My father brought home the bacon, but he always said that my mother was the brains of the family. My mother was artistic and a great reader. She introduced me to classics like A Tale of Two Cities and All Quiet on the Western Front at a very early age, and this instilled my love of books and storytelling.

I went to UC Berkeley to study English and journalism. It was the early 1970s, and the counterculture movement was at its height. At Berkeley, I discovered photography. I've always been a technology geek. Photography, of course, is extremely geeky, and it's also about telling stories. It married my two passions. If you're fortunate enough to discover the thing that you can't not do, that's the bomb.

After college, I pestered Eastman Kodak to hire me. I was relentless. They finally offered me a job in marketing. This was in 1983. My assignment was to tell the story of digital imaging, which was in its infancy. After a few years working for a big company, I got the entrepreneurial bug. I moved to Seattle for Kodak to start a company developing imaging software for the pharmaceutical industry. It was a steep learning curve. There were only five of us, and we thought we were entrepreneurs, but we really weren't there yet. We had a good idea but were way too early. That was the first significant failure of my career, and it had a big effect on me. I realized your greatest strength can be your greatest curse. You can see things coming, but you see them before they're ready.

I had no desire to return to corporate culture. I went to work for Wall Data. Part of my job was going to the computer science department at the University of Washington to search out new technologies. I found software developers combining Web design with database tools in ways that were just awesome. We built a company called Salsa based on that technology. It became a big success.

I left Salsa to move back to the Bay Area, and a few years later, the same thing happened. Some VCs asked me to check out three software developers who had just left the Stanford Research Institute. They were designing software that followed a user's path through websites in real time, providing companies with new insights into consumer behavior. From that, we built a company called Personify.

Running two successful start-ups taught me two lessons about being a CEO. The first was that you have to raise the money, and the second was learning how the minds of investors work and understanding the dynamics operating within a board of directors.

Then the tech bubble burst, and Personify was acquired by another company. I parachuted out as a consultant and board member for some Silicon Valley companies. I became CEO of one of them, and that helped me realize I'm not the consultant type. I'm a builder. So I started looking around, and, as I searched, I had time to take pictures. Digital cameras were just taking off. This is awesome! the geek in me said.

I set a goal—to make portraits of 40 people with whom I'd built companies. I thought I knew these people, but when I started photographing them in places that were special to them, I realized I didn't know them at all. I never knew this man was an amateur boxer or this woman raised horses.

I wanted to present my portraits of these Bay Area Renaissance people in a format that reflected their depth. I needed something tangible. You can't gift a website. I decided I was going to make a book—a professional-quality bound book. The catch was I wanted only 40 copies.

I talked to printers, but they wouldn't accept an order of less than a thousand copies; otherwise, it wasn't profitable for them. I would have to design and publish this book myself. So I went online, expecting to find a wealth of resources. After all, desktop publishing and Adobe had been around for years. But I couldn't find anything. I was shocked.

I thought I must be missing something. Then the printers explained that they're set up to produce 5,000 copies of one book, not 5,000 orders for separate books. That clarified my thinking. I'm a systems person. How do you build book-designing software for people who aren't book designers? How could printers profitably produce a book of one?

For two years, I worked full time resolving these questions. I had meetings in every Starbucks in San Francisco. I met with venture capitalists who gave me a hearing only because of my track record. This was 2004, and everything was going digital. Here I was, proposing a model for going from digital back to print. They thought I was a lunatic. But I kept the faith. I knew that millions of people all around the world had something to say, something to express in book form. Finally, a VC decided to bet on the horse despite how crazy it sounded.

Our book-designing platform had to be simple enough for any reasonably computer-literate person to use, but it also needed to be fun. Designing your book couldn't be a chore. It should be the thing you want to get to. And that same software had to be clean and elegant from the printer's perspective. That was the secret to making a profit on a book of one: a standard file for printing that came in clean every time.

Our website launched in May 2006. Our customers are everybody from photographers and architects wanting to showcase their portfolios to a couple wanting to commemorate their wedding anniversary. Business is almost doubling every year. At peak volume during the holiday season, we process a book every 1.5 seconds. The files are sent along to the printer, and a week later, the customer gets his book in the mail. We charge from $2.95 for a black-and-white text-only pocket-size book to $206.95 for a four-color 400-plus-page coffee-table-style book.

When building a business, you can't do it all. You have to stay focused. At Blurb, we've built an Internet platform for people to produce their own bookstore-quality books. That's it. That's what we do.

For a full archive of How I Did It features, visit www.inc.com/hidi.