How Do You Eat An Elephant
The enemy of good is perfect. The effort to go from something that is good enough to something that is perfect can be enormous and costly.
There’s a great YouTube Channel called Nile Blue that publishes different chemistry experiments. In one, the chemist decides to make the world’s purist chocolate chip cookie. By “purist”, he means using ingredients that are devoid of anything else than what’s on the ingredients label. Bug parts, dust, and other unmentionable contaminants are allowed by different food agencies as long as they are below some concentration.
This chemist wanted none of it. The result was the a pure cookie but ultimately, one that wasn’t very tasty. It’s hard for perfection to exist across multiple dimensions.
When we’re designing for perfection, it also expands the requirements gathering process to the point where you add so many layers of polish that the object is just a ball of polish. Instead, there is some asymptote to optimize for where the minimum viable product sits, and to start from there. At some point, there also needs to be a cut off for requirements gathering.
Once you’ve set that point, there is still the process of “eating the elephant” and that should be done one bite at a time. Each bite needs to be cooked and tasty. The metaphor is that even if you’ve set your MVP, you can break it apart into phases that can still be useful and at least ship something that can grow over time. Early testing is better than none.
This doesn’t necessarily apply when you’re looking to launch a new product that requires stealth, or a feature that “crosses the chasm” in functionality. However, for most product improvements and features, this makes sense.
