The house I live in is almost 30 years old. My wife wanted to freshen up the outside of our house, so we started on a few projects. Little did I know I was about to go down a very slippery slope.
It all started with a new front door. We decided to do it right and get a new door, framing, side lights, storm door, professional installation, the works. We even got a little crazy and bought a red door so we would stand out in the neighborhood. We got the new door, and it looked fabulous.
But the new front door was overshadowed by our 30 year old sagging garage door. So we decided to replace that also. After garage door was in, it was obvious we also needed to repaint all the wood trim on the house, so we did that also.
My freshening project was quickly expanding, and I couldn't stop myself. The next problem was the siding. It was in decent shape, but 30 year old aluminum siding with oxidized paint isn't exactly the sharp appearance we were looking for. Then I got some quotes to replace the siding -- $18,000. Whoa! I had found my tipping point.
I stopped to think about it, and I didn't really need to replace the siding. Our original goal was to freshen up the outside appearance of our house, and we unquestionably did that. But after I got started I wanted to fix every single thing about the house and make it look like new. I almost turned it into a Cadillac of home project, when all I needed was a Chevrolet.
The funny thing is, I see my customers struggle with this very same issue on almost every software project. Most projects start out with reasonable (Chevy) goals. For example, to reach out to customers on the Internet. But after the project gets rolling it quickly gets loaded with attempts to address every conceivable scenario that may ever arise. The Chevy quickly turns into a Cadillac.
When you build (or buy) business software, resist the temptation to buy a Cadillac. In the software world, Chevys are generally better than Caddies, because:
- Most of the business value in software is addressing big picture issues, not fixing every single problem you have.
- Cadillac software is more complex, is more error prone, adds delays, and is more expensive (just like my siding).
Bottom line: Focus on your project's key goals and build (or buy) a Chevy system that meets those goals. When you succeed, you can add the Cadillac features later.
If you have any comments, I would love to hear them. Please let me know what you think by replying to this message or contacting me at bpowell@jx2services.com.
