If you think reengineering belongs only to the cluttered bureaucracies of big business, it may be time to take a closer look at how things are being done at your company

This year my 11-year-old company, Racing Strollers Inc., went on a different sort of odyssey. We stumbled willy-nilly into reengineering, or process analysis. Now we're like the crew of a little spaceship: we've blasted off, we're too far out to turn back, and we're praying that we'll land OK.

The trip started when I met Spike at a Tom Peters seminar last year. Spike told me that at his job at a major U.S. corporation, he and his team roved from factory to factory "like a quiet storm." Just what, we all wanted to know, was this quiet-storm business all about?

My company invited Spike to come up and teach us. The first morning, over breakfast I explained that we had originated the three-wheeled stroller for joggers, our Baby Jogger, and were the market leader, but I was deeply concerned about losing ground. My competitors were manufacturing offshore, were selling their strollers for half my price, and were superb marketers. We'd improved our margins and profitability, but I was sure there was no way I could make one of my fine strollers for anything near the costs that I figured my best competition had.

Spike then blew me out of the water. "Wow," he said, "you'll never have a chance of competing on price as long as the president has given up." We decided, with Spike's help, to reengineer how we did things. Spike went around and talked to people in the company (which we'd done before) and started mapping out work flow (which we thought we'd done before). But we had never had someone do it in as much detail as Spike did, and we'd never tried asking, "Why?" and, "Why?" when someone said something had to be done a certain way.

After the first day, Spike said he had a few things to show me. He took me to our production area and had me watch our people staple a large box (for the stroller) closed. Then he took me 20 feet away, where a worker was pulling the staples out to open the box so one of our colored seats could be thrown in. And he asked a very simple question: "Why do you put staples in, only to make the poor guy take them out again?" It turned out that our boxes would collapse in a stack if the ends were left open, and we needed to put in our bagged seats at shipping time. Now we're experimenting with tabbed boxes to find out just how little we can do and still have the box hold up.

It was just as bad, and even worse, in our office. Take faxes. Pretty simple, right? Not at our place! We had 15 steps just to get a fax from the fax machine to the computers, where an order would be entered into our accounting program. (Laugh all you want, but you probably do something similar at your place.) A fax would come in. We'd log it into a fax book. (Someone once told us to do that for legal reasons.) A copy of each fax would be saved for me. (Someone once told me I should glance at all faxes.) It all added up to those 15 steps. Only after a lot of work did we reduce the number to 4.

That was just one area, and it's typical of what we found throughout the company. If you were to look at all those steps for a fax, every one originally had a good purpose. The process was very much like a tribal superstition, though, since the original reason for each step had disappeared long ago. What was even worse was that no one could remember who started many of the steps. I had no idea we were doing them, and everybody else thought they were what I wanted done!

What really hurt were all the dumb things we were doing in the name of quality. We were obsessed with quality, as we should have been. But our quality efforts had begun to run the company. We had weekly quality meetings, at which we gave employees door prizes and paid time off for good suggestions. We documented and charted quality stuff like crazy. And that's where we got really screwed up. Everything about quality says, "Document, document. Chart it. Get numbers on problems; if you can get a number and a cost, you'll be able to improve." And to a certain extent, that's true. But that obsession hurt us because we were still building flaws, accidentally, into our production process.

Here's an example: Someone would bring up a problem, like scratched tubing on some of the strollers. We would add that to the list the inspection people used. The line people would also be asked to watch for it. Jennifer in sales would be told that we had a problem, and sales would gear up to sell off the scratched strollers as factory seconds. Perhaps someone would start charting that problem, and we would add it to our checklist. And the problem would stay on the list of items to be discussed at the quality meetings until we felt our customers were protected from it. But we might never get to the root of a problem -- in this case, it was that we produced subcomponents (like our handle assembly) in batches, which meant they could get scratched when they were set on a rack with other parts.

In our passion to get a perfect stroller out, and because we followed fairly standard quality procedures, we added, oh, say, five steps to various processes, both in the office and in the production area. And in our gusto, we did the same thing almost every time a problem came up. Now, our people are really bright. If the root cause of a problem was easy to see, we'd root it out. But a lot of times the extra steps would remain long after the problem had disappeared.

Now, you're probably just like us. We said, "Aw, shucks, Spike, there's no way we'd have 100 steps in a process!" We didn't have 100 steps. To build a stroller we had 237. That's build. That doesn't count the ordering processes, the accounting processes, the purchasing processes, and the customer-service processes. We didn't have those steps in some sort of policy manual, either. They were just the little bits of folklore that we accumulated and taught new people in our training programs. And over the years we added a little here, put a few personal touches in there, and never took a damn step out. A French pastry chef had nothing on us! We were the Cordon Bleu of complicated processes.

Here's what Spike taught us, which is almost the opposite of quality dogma: Let's say it takes 100 steps to accomplish a certain process (like manufacturing a stroller). If 80 of those steps are unnecessary, what has happened to the time involved in the process? It has increased by 80%! And to get a perfect product, you need to do each step perfectly. So if 80 out of 100 steps are unnecessary, your chances of causing an error have gone up dramatically.

Spike helped us chart the steps in each process. He'd take a big piece of paper and put yellow Post-its on it, each Post-it representing a step. We'd bring in the people who worked in that particular area and make sure we'd gotten the steps right. (The Post-its let us move steps around on a page.) Then we'd ask questions to find out how essential each step was and put a big X on steps we could eliminate. That's where we ran into big problems. People thought it was a fabulous idea to dig into our processes until we got around to their section. Then the kicking, the yelling, and the screaming started.

One of the best points Spike taught us is that people are working as hard and as fast as they can. They're giving their best effort. It's the system, all those processes, that holds them back. Great people cannot beat a bad system. On the other hand, people really do hate change. All those good people worried that if they gave up control of a step (if, for example, the salespeople were allowed to check on credit status and make a decision), we would completely lose control and also screw up that part of the process. We had to argue, explain, and coax, and occasionally we had to insist that they try the change.

We're only halfway through all our reengineering functions; the biggies coming up are retooling the accounting processes and reassessing the steps to make a Baby Jogger. Our spring tidal wave of orders is just about to start, and we're nervously watching to see if our pared-down ordering process is going to work.

The best part is that we have hope. We've watched our overhead per employee like hawks. We've tracked that number for years, but we couldn't get it to budge. Now we see how we can keep the same workers, increase our job security (profits have a way of doing that), and cut out half the work our employees are doing. If we follow the process evaluation to its limit, we will have freed up our energy and our people in the best possible way -- by making more profits! What more could a company want?

* * *

Mary Baechler is president of Racing Strollers, a $7-million manufacturer in Yakima, Wash.