Skateboards vs. Cars Revisited

Marty Cagan revisits Henrik Kniberg's skateboard-versus-car MVP article and argues it describes what engineering-only teams can do, not what true cross-functional product teams should do. He contends that a skilled product manager and product designer should run product discovery to test risks in value, usability, feasibility and viability using many kinds of prototypes, rather than having engineers build learning iterations in production. Doing 10-15 iterations in discovery can compress months of elapsed time into days. Discovery also makes delivery faster and more focused, since engineers can then architect and build the right product at scale without repeated detours.

01Key takeaways

• Treat engineers as builders of the final product, not as the main tool for learning what to build.
• Identify value, usability, feasibility and business viability risks before committing engineering time.
• Choose the right prototype type for each risk and test it qualitatively or quantitatively.
• Include engineers in discovery to improve the solution, not just to check feasibility.
• Use discovery to clarify requirements so delivery can be faster and avoid costly rework.

02Key sections

Engineering teams versus product teams

Cagan distinguishes teams made only of engineers with a basic product owner from true cross-functional teams that include a skilled product manager and product designer. He says the skateboard approach is the best option for the former but that the latter should aim higher.

Discovery before engineering

The first principle is to tackle value, usability, feasibility and viability risks before engineers write production code. Teams pick from several prototype types and test them with users, teammates and business stakeholders.

Why engineers should not do the learning iterations

Using engineers to build learning iterations is far slower and wasteful, often by an order of magnitude. Discovery work is led mainly by the PM and designer, with engineers involved throughout to shape the solution.

Time to money and the discovery-delivery model

A medium-to-large product effort may need 5-15 iterations, which can take a week of discovery instead of months of releases. Delivery then has a clearer target and proceeds faster, since the engineers can design a scalable architecture for a known product.

03From the post

“A partnership dedicated to teaching best practices to product teams and product leaders”

04Frameworks mentioned

Summary and takeaways written by PM Atlas; quotes are short excerpts. © the original author.