My friend Christopher Bibby sent me this article over the weekend. It's by Brian Potter -- who writes an excellent newsletter on Substack about construction things -- and it's about why it's so hard to innovate in construction.

To explain this, he starts by showing that the distribution of cost outcomes in construction projects tend to be both skewed toward the right and "fat tailed."

What does this mean? It means that construction projects have a tendency to run over a budget. And that they are much more likely to be over budget than under budget (right-skewed distribution). According to some data from the US Navy, the difference in likelihood is 10x.

At the same time, there are also instances where projects don't just run over budget, they run really over budget (fat tail). All of this is different from your normal distribution where you have a symmetrical curve and thin tails. I guess construction isn't normal.

One of the reasons for this abnormal distribution is the fact that construction suffers from what Brian calls "cascading failures." This is kind of intuitive, but it is everything in construction: In order to complete Y, you need to complete X. If X is delayed, then everything is delayed.

Because of these dynamics, changes to the construction process are perceived as incredibly risky. This has created a bias toward incremental rather than fundamental innovations.

For the full article, click here. It's worth a read.

My friend Christopher Bibby sent me this article over the weekend. It's by Brian Potter -- who writes an excellent newsletter on Substack about construction things -- and it's about why it's so hard to innovate in construction.

To explain this, he starts by showing that the distribution of cost outcomes in construction projects tend to be both skewed toward the right and "fat tailed."

What does this mean? It means that construction projects have a tendency to run over a budget. And that they are much more likely to be over budget than under budget (right-skewed distribution). According to some data from the US Navy, the difference in likelihood is 10x.

At the same time, there are also instances where projects don't just run over budget, they run really over budget (fat tail). All of this is different from your normal distribution where you have a symmetrical curve and thin tails. I guess construction isn't normal.

One of the reasons for this abnormal distribution is the fact that construction suffers from what Brian calls "cascading failures." This is kind of intuitive, but it is everything in construction: In order to complete Y, you need to complete X. If X is delayed, then everything is delayed.

Because of these dynamics, changes to the construction process are perceived as incredibly risky. This has created a bias toward incremental rather than fundamental innovations.

For the full article, click here. It's worth a read.