by Christopher Lindquist

When Good Enough Is Good Enough

Mar 04, 20054 mins

I come from a family of perfectionists—people for whom no level of quality is ever quite high enough. Every project is bound to demonstrate some flaw (even a tiny defect that no one else would ever notice) that will drive the maker crazy until the end of time. It’s just the way we’re wired about some things.

But for me, that trait doesn’t apply to the software I use. Now I’m not talking bugs; I have no tolerance for bugs. I’m talking features. Give me stable, reliable software that installs easily and does 90 percent of what I need in a simple, elegant manner, and I’ll take that immediately over something that will deliver everything I could possibly want but only at extraordinary effort or cost.

I didn’t used to think that way. And I confess, sometimes I’m still tempted by the 100 percent equation. I recently tried to install a version of Gentoo Linux on a server, lured by the promise of ending up with a distribution tailored to my exact requirements—nothing more; nothing less. Two days and half a dozen failed boots later, I tossed my Gentoo disc in the trash and downloaded ProMepis. Does ProMepis let me choose, tweak and optimize every little detail of my installation? No. But it installed in 20 minutes and came preconfigured with the Web server and database I needed. (If you’re a Gentoo fan, don’t flame me. I know it’s a great distribution for a lot of folks—just not for me.)

Corporations are beginning to take the 90 percent rule to heart, too, often as a result of deciding to dump massive roll-in-the-cots-and-build-a-new-wing monolithic software implementations in favor of hosted options. Software as service providers such as, Rearden Commerce, and JRG Systems will gladly give you 90 percent of what you need—from CRM to travel planning to supply chain management—quickly, relatively easily and nexpensively—if only you’re willing to pry your claws off that dream of a 100 percent solution. (Rearden has bigger plans, of course, but I’ll let this Infoworld article tell you all about them.)

The software as service guys aren’t apologetic about what they offer, either (although’s sforce customization initiative comes dangerously close to a mea culpa in my opinion–but they wouldn’t have offered it if customers weren’t creating the demand.) Their strategy is all about leveraging a 90 percent solution across as many customers as possible. Custom code just gets in the way of profits—the exact opposite of what your average monolithic software company is thinking, where the base code is often seen as the toehold that leads to the lifelong engagement.

The big problem for these providers until recently has been mindset—software as a service just sounds dangerous to a lot of companies that have built themselves on foundations of millions of lines of customized code wrapped around software from a handful of vendors. But even some of those companies are beginning to loosen the straps that have tied them down for so long—and those that aren’t will soon find themselves falling well behind the curve as their competition discovers that 90 percent now means a more agile, aggressive enterprise that doesn’t waste endless energy in trying to attain IT perfection. And heck, 90 percent now means more resources available to address a few of those high-value, high-cost percentage points that the generic solutions won’t provide.

Really, even the Swiss figured it out. Sometimes you need a Patek Philippe; sometimes you just need a Swatch.