Perfection is a funny thing. It seems like it should be a good thing if it could be achieved. During my years as a competitive fencer, though, I often noticed that the winners of the tournaments were rarely the fencers with the most perfect technique. Rather, the winner usually had some flaws in their technique. Oddly enough, those flaws didn't seem to matter. What mattered was that their areas of strength were strong enough. More broadly, it turns out that those Olympic athletes who strive to be perfect tend to do less well than those who strive to constantly do better.

In the business world, the desire to build the perfect product is a common one. At one Silicon Valley software startup that I worked with some years ago, one of the VPs regularly insisted on adding "absolutely necessary" features at the last minute. As a result, the last minute got pushed out further and further. The product also became steadily more unstable as it pushed further and further beyond the limits of the design. Although the product did eventually ship, the company did not long survive the event. Customers had simply been kept waiting too long, interest in the product was waning, and other products were entering the market. Ironically, none of those other products were as capable as the more perfect product. For some reason, that just didn't seem to matter, perhaps because those products were available and provided clear and obvious value, whereas the more perfect product provided features that people just couldn't imagine themselves using. More broadly, spending the resources to build the perfect mousetrap is rarely cost-effective: by the time it comes out, someone else has come along with a cat. In other words, they've satisfied enough of your customers' needs in a different, unexpected, and attractive way.

The problem is that perfection is idiosyncratic. Ask ten people what a perfect product would look like, be that a phone, a computer, or a theme park, and you'll likely get at least ten different answers. Ask your customers what it would take to make your product perfect, and you'll do two things: first, you'll get a wish list that's impossible to implement before the heat death of the universe; second, you'll set their expectations so high that nothing you actually do will satisfy them. However, if you ask them how to make the product better, you'll get a set of suggestions at least some of which you can implement. Your customers get a better device, and become more loyal because their input made a difference. Thus, the original iPod becomes the iPod Touch, the iPhone, the iPhone 3G, and so forth. Each iteration is based not on what would make a perfect device, but on what would make a better device.

Another unfortunate consequence of perfection is that the more effort someone makes to create that perfect mousetrap, the more invested they are in that particular style of mousetrap. They become less and less able to recognize that other people, such as potential customers, might have a very different view of what constitutes perfection: the person who has never seen a cat, or who doesn't like cats, may not view cute and cuddly as an important feature of a mousetrap, but there may well be plenty of other people who do want those attributes. For example, Polaroid created a perfect mousetrap: instant photography. They were so invested in that market that they couldn't even see digital photography coming. In the end, Polaroid's perfect mousetrap developed poorly.

So how does a company avoid the perfection trap? It's critically important to understand your own decision making process. How do you define what you're trying to build? How do you decide which features to include, which to push off to another release, and which to ignore?

Do you decide based on personal opinion or the most confident voice in the room? Unfortunately, the most common method of decision making is to go along when someone confidently states, "We must have those features or the customers won't buy the product. I've done something very like this, and I know it'll work."

Instead, start by figuring out exactly what problem you're solving. What, specifically, are you trying to do with your product?

Next, define your criteria for success at the beginning. What actually constitutes victory? How do you know? And how will you know when you get there?

As you start to generate possible solutions, constantly ask yourself if those solutions will actually generate the results you want. Always be aware of the criteria that you are using: opinion, educated guess, testing, or research.

In the end, if you want to make money, it's not about having the perfect product, but the product that's good enough to convince people to buy.