Things can always be better. We all know that. Often there is a real temptation to continue to tinker with a product to make it perfect. This is a dangerous road because it can lead to continuous requirements creep and if taken to an extreme, can actually make a product worse. In a prior life I had a unit that came back for a warranty claim. When we opened the unit it looked like a something straight out of a prototyping lab. When I looked into it, it turned out the customer couldn't make up their mind about the communications interface and we had to add a second data bus as an afterthought. As it turned out the attachment method made these boxes far less reliable but it was the only way we could fit the extra components in within the customer's time frame. You need to know when to say NO to change.
When building a new product it's important to have as good an understanding of the market requirements. More importantly before you start to move out with the detail design work these requirements should be reviewed and signed off by the executive team which needs to have representation from engineering, marketing and operations. Further, there needs to be a change process where any new requirement that come in are evaluated to understand their benefit and impact before they are accepted. The change management process is important and it is extremely important that pros and cons are thoroughly understood. Teams will often accept a new requirement thinking it's a just "a day or two". It might be a day or two to get a concept, but then figuring out how to get it into the design, fixing the things it broke in the process and testing it could take weeks. Further, this is tying up resources that should have been doing something else.
Having come from the defense industry one of the main causes of cost growth and schedule slip in programs is requirements creep. I was once asked to compare the cost of a commercial satellite development program with a military satellite development program. A major difference is that commercial programs focus on freezing requirements during the contractual negotiation phase and there is rigorous change control where the customer is allowed to make changes only if they modify the contract to accept the extra cost and schedule. In the old days in the defense world the requirements were in flux up through the conceptual phase. The government and their representatives have approval rights on every aspect of the design and changes come pouring in through this approval process plus through any changes to scope of the mission. The end result is that comparable government and commercial space programs take over twice as long and cost over twice as much.
Having a change management process is one thing. Following it is another. I lived this scenario in a prior life., our CEO was a voracious reader. He spent hours at night reading about the latest technologies and he would send emails out at all hours of the night "suggesting" we think about this idea or that. I would have to constantly go back and spend hours explaining the impacts of going back and modifying the design but often we would be forced to go back and have someone go through an exercise of evaluating a new technology. This churn probably cost us months in unplanned effort and millions of dollars without an appreciable change to the program. In this case we had a good change management process but the CEO was his own worst enemy by circumventing it. Everyone needs to follow the process. Even the boss.
I'm not saying that all change is bad or unnecessary. Sometimes changes can be very valuable. They can offer significant product differentiation, open new market segments or, reduce product cost. It's important that you have a process for evaluating these changes, have an honest and thorough understanding of their impacts and benefits to make sure that you are making an informed choice. Otherwise you can fall into the trap of chasing the next bright, shiny thing. This is money and time you had not planned to spend and therefore you need to be sure the investment will pay off.
No comments:
Post a Comment