How Innovation Led HTC to the Dream

Last week's unveiling of the new HTC Dream, the first handset powered by Google's Android platform, represents more than another innovation by the mighty Google. It also marks an important chapter for HTC, the Taiwanese company that manufactured the first Google phone and has quietly built itself into one of the world's leading manufacturers of mobile phones.

How did HTC come to build the first phone powered by Google software?

Andy Rubin (Google's director of mobile platforms) and Peter Chou, our CEO, actually knew each other for years. The collaboration happened very organically. The first step was to create the Android platform. For a mobile phone, it's not only a software effort, nor is it only a hardware effort. It truly took the collaboration of both Google and HTC to create what you see today. Together we had been working on the Android platform and the first Google phone for almost three years now. We had people collaborating with Google engineers on the Google campus. We also had a big team within HTC working on the Google phone.

How would you describe your company's model for innovation and the innovation team known as Magic Labs?

HTC already had innovation DNA for many, many years. About three years ago, I started Magic Labs inside HTC more to formalize the innovation process. Magic Labs today has about 60 magicians.

You really call them magicians?

We actually do. We have business cards that say software magicians, chemical wizard and so on. I'm the chief marketing officer of HTC, but if you look at my business card it says "Chief Innovation Wizard."

Actually business cards are not that important. What is important is that this is the only group within HTC that does not have product ship date deadlines. They are not chartered to ship the product over the Christmas season, but rather to think beyond Christmas, think beyond tomorrow, and basically think about the future.

Inside Magic Labs there are people from a diversity of different backgrounds. We of course have software magicians, hardware magicians, electrical people, people with mechanical engineering backgrounds, graphic designers, usability experts. One of the magicians used to design jewelry in New York. The whole organization is designed to fail.

What do you mean by designed to fail?

The way to get a great idea is to have many ideas. By definition, most of your ideas will fail. You want to be able to generate ideas very fast, very cheaply and fail very often but at very low cost. Magic Labs is optimized for the efficiency of failure. Among the many ideas, there will be great ideas that bubble up and then we will invest R&D efforts to cultivate the great ideas.

Can you describe an example of this process at work?

Let's dial back to about three years ago. This was a time when all the phones were getting more and more features and phones were getting too complicated. Everybody was talking about simplicity, everybody was talking about usability. What did people do? Well, they rearranged the menu and called that improving usability. That's not simplicity, that's rearranging menu items. People added graphics -- that's pretty icons, but not simplicity.

Inside Magic Labs we tried to tackle this problem as well. We also tried rearranging menus, making the screen easier to view, so on and so forth. But it did not work. One day, inside Magic Labs, there was an epiphany -- that was a day I still remember -- and the key can be described in a single word: baby.

We recognized that we had been going about simplicity in the wrong way. This happened in a brainstorming session. The true mission is not to reduce learning, but to eliminate learning. There needs to be zero learning, not very little learning.

The baby is probably the best expression for zero learning because the baby has not learned anything yet. If she wants to see the monkey on the other side of the block, she simply reaches out and turns the cube. I don't think she would read a user menu. There is something that is innate to living beings that you just simply reach out intuitively and turn the object. In HTC Touch, there is a slide out cube and you literally just turn the cube. That design was actually motivated by the recognition that people don't read user manuals, they just want to interact with objects intuitively.

We have been abusing end users for many, many years by forcing people to push the up, down, left and right buttons. That defined the direction of Touch Flo and, subsequently, HTC Touch. On the HTC Touch, when you are viewing a web page to move it around, you just reach out and move a virtual sheet of paper around, just like what a baby would do.