Bart on the Road

In episode 20, season 7, of The Simpsons, Bart and a few of his friends get their hands on a car and go on a long road trip. At Milhouse's suggestion, they decide to travel to the World's Fair in Knoxville, Tennessee. Only after the long drive do they realize that the fair was held years prior. All that wasted time and effort, only to arrive at a destination that wasn't what they hoped for.

And that is exactly what is wrong with the sequential model of product delivery I wrote about in last week's post. All that planning and deliberating and building and testing, and when do we find out if users actually want the thing? At the very end.

The sequential (so-called waterfall) model has many other issues, but this is the biggest one: The most important risk, the value risk, gets resolved only at the very end.

"But didn't we do a business case at the beginning?"

Sure. But how can you know how much money the "feature" will make until it's out in the market? And how can you know how expensive it will be to build (and sustain) it until your engineers have actually built it? And, more importantly, how can you know what the "effort vs payoff" curve looks like? Because for any feature, its success depends on how well you build it.

The way they're typically applied, with all their numbers of purported rigour hiding the uncomfortable truth that you just cannot know, they're a waste of time. There are good ways to do them, but we’ll leave that for another post.

Okay, I get it, we'll start with an MVP

Good idea, but also a much misunderstood concept, so stay tuned for more.