Keith Krach: The making of a technology leader

Cutting losses, inventing a 'category king'

So, you quit before the mission was done?

Krach: Yes, I did. But there was no way to turn that one into a win. So, I thought, 'All right, why not practice what they taught at Harvard Business School? Put together a business plan for myself and cover the five P's in marketing—product, price, promotion, place and people.'

I wrote a gazillion letters, because there was no email back then. The headhunters, the Harvard Business School alums, the Purdue alums -- I left no stone unturned. Finally, I ran across someone I'd never met before. He was an HBS grad and a VC [venture capitalist]. He told me, 'I know these three Ph.D. scientists from the IBM Research Labs. They're developing some interesting mechanical computer-aided engineering software -- some kind of automation. Why don't you go check it out? You're out of the automotive industry -- they'd be big target customers.'

I checked it out and saw what they were doing. I figured if they could actually build this product, it would change the way mechanical engineering is done. They were looking for a business guy to lead it, and I said, 'How long do you think it'll take to develop the product?' They told me, 'Six months.' I said, 'OK, I'll join.'

It ended up taking two-and-a-half years to get our product done. That was another valuable lesson. In software development, it's like a law of physics. Estimate how long you think it's going to take to get done and how much money it's going to cost, and you can usually count on it being about '2x + n.' Double whatever it is and add something. It actually applies to most big, complicated projects that involve teams of people.

So, that's the first rule of software development? 

Krach: Yes -- double and then add something more. But I didn't know that law of physics back then. Ironically, it gave us a lot of time to prepare our market. One thing that I had learned back in the GMF Robotics days is the advantage of creating a category because the objective is to be the category king -- particularly in the software business. The category king gets 80% of the market resources and 80% of the market cap.

But we had to balance that against how fast we were burning through cash. We had some VCs who invested in us, but we tapped them out. We got to the point where we were on the verge of running out of money. We were working hard to raise more, but we kept getting turned down by potential investors.

We had real geniuses on our team, but we hadn't proven ourselves. So, I said, 'Let's go to Japan.' Back then, Japanese companies were doing a lot of investing in Silicon Valley. We found a great company called Kubota; they're like the John Deere of Japan. Our original plan was to ask for just enough to get by -- $1 million. We gave them a three-year business plan.

They looked at it and pushed it back across the table. They said, 'We don't do three-year business plans. We do 20-year business plans.' I said, 'That's good, we like that.' We ended up asking for $10 million and got it. It was $3 million in equity, another $2 million to start a joint venture in Japan, $3 million customer deal, $2 million nonrecurring engineering, and we kept the company alive.

What's does nonrecurring engineering mean?

Krach: That means that they paid us time and material for some things that they wanted specifically for Kubota. So, it was like a one-time expense. It wasn't a product we would sell, although we actually did incorporate some of that functionality into our product.

When the product did come out, we created a new category based upon a new technology. Instead of finite element analysis, it was geometric element analysis, and we created a category called mechanical design synthesis. The company was a bit of a slow start, but it came on really strong, because the value proposition was great. Up until that time, you needed Ph.D.s to do that level of structural analysis, mechanical analysis and thermal analysis. Our technology simplified it to the point where regular design engineers could use it, and it would integrate into their CAD [computer-aided design] systems.

100 goals

The company you started was Rasna. What became of it?

Krach: We were all set to take it public in 1995 with Morgan Stanley. We had partners like Autodesk and Parametric Technologies. One day, we had a meeting with Parametric's CEO, Steve Walske, and COO, Dick Harris. I was talking about doing a deep partnership. They said, 'Let's do a deep partnership.' Then, they said, 'Really deep.'

I was wondering where they were going with this, but I said, 'OK, really deep.' Finally, they said, 'I don't think you understand. We want to buy you. Let's just do it. You don't need bankers for this.' I got out of that room, and the first thing I thought was, 'We'd better get some bankers.' They did buy us. By the time everything wrapped up, it was about a $500 million deal. It was probably the best acquisition they have ever done.

At that point, they wanted me to go back to Boston and be their EVP [executive vice president] of strategy. But, by that time, I had fallen in love with the pace of Silicon Valley. Suddenly, I could afford to do whatever I wanted, so I took six months off and set 100 goals for myself.

Wait a minute. You'd been working on growing this, and you had just sold it -- and then walked away?

 Krach: They bought me out, so I didn't stay.

So, your mission was accomplished.

Krach: The mission was accomplished. The shareholders saw a great return on investment, and the employees got a great return, too.

I called my mom back in Ohio to tell her about it. Usually, whenever I called my mom, my dad would pick up on the second phone upstairs in their bedroom. I could always hear him clicking in. So, I said, 'Hey, ma, we just sold the company, and I don't think I have to work anymore.' My mom said, 'I guess this means you're moving back to Rocky River, Ohio.'

Then, I heard my dad say, 'Elda, I don't think he is ever coming back.' It was a bittersweet moment. My dad and I used to talk about my going to college and then coming back to help him grow the business. And I would have loved to spend more time with my dad, but after GM and Rasna, I was really on a different path.

So, you took six months off.

Krach: I had my 100 goals. I broke them out by family, physical, intellectual, spiritual and friends. I took each one of my three children, at that point, on an individual vacation.

What inspired you to do 100 goals?

Krach: I had read Stephen Covey's book and decided to write down all my goals. I'm a fanatical goal setter, and I knew I just couldn't lay around the house or go golfing every day. But I wasn't ready to go back to work at a company full time, either. I decided it was a great time to sharpen the saw.

Believe it or not, one of my goals was to learn to type. I had never learned before. I always had friends in high school and college who helped type my papers. At General Motors, I used the typing pool at first, and then I got a secretary. But in the era of the PC, everybody started typing for themselves. It was really kind of funny. I went out and bought the Mavis Beacon Teaches Typing software. I also learned how to scuba dive and other things like that.

When I finished all 100 goals, I popped my head up, and Bob Kagle, an old General Motors guy who was now a VC starting Benchmark Capital, said, 'Hey, why don't you come on and be a VC?' I told him I wasn't sure I was ready to do that. I thought maybe I would be a VC when I retired. He persisted, 'Why don't you become our first entrepreneur in residence?'

That sounded like a lot of fun, so I joined them. I sat in on the partner's meetings and learned about venture capital. And that's when I saw a big opportunity to pull together some of the old Rasna team and build a business on the internet, which was gaining traction commercially.

Editor's note: In part two of this extended interview, Keith Krach recounts how Ariba got off the ground.