Engineering in the Face of Irreducible Uncertainty

Imagine you are building a parking lot for a new shopping mall. You have a fixed footprint and will use all of it.

But how tall should you build it?

If you knew how popular the shopping center would be, and another dozen factors, you would know.

But without a time machine, you can’t know.

If you build it too tall, you waste money.

If you don’t build it tall enough, there won’t be enough parking space making everyone very unhappy.

You could guess / model / estimate then add a safety margin and on average you would be too tall by your safety margin and occasionally be much too tall and on rare occasions, too short.

What do you?

The common answer is to make better estimates. (Or encourage mass transit.)

But another answer is to build a very short parking lot but to build it in such a way that it is easy to add stories. In the case of parking lots, this can often be done in a manner that adds very little cost to the short parking lot and averts huge expenses or other disasters if the parking lot must grow.

This is engineering with irreducible uncertainty.

What if we could apply this principle to initial products from startup companies?

Reduce the cost of an eventual pivot, should it end up being required.

It is possible and it begins with rich enough customer discovery to know what the most likely changes (or full pivots) might be required.

If you haven’t guessed right about your first product, do you have a good second guess?

And third?

Successful serial entrepreneurs (on their second and subsequent companies) are prepared for changes in case the MVP experience didn’t teach them everything. Nothing stops you from being prepared and successful with your first company.

It’s not just that we can learn from the mistakes of others, we can learn from the brilliance as well. Preparing to prevent the need for a pivot (or outright failure) definitely counts as #SuckageReduction.