Big capital projects – hydroelectric facilities, subway systems, massive ERP installations – have in common that they are often built in a monolithic and customized way. Huge upfront investment, and no benefit until the thing is complete. It doesn’t have to be that way, according to research conducted by my guest Bent Flyvbjerg. Even with mega-projects, the principles of replicable, modular design and speed in iteration, work to increase the odds of success. With so much of the future depending on the quality of huge programs, this is a message we all need to hear.