Last week I had a chance to sit down with Andrew Cox from FreightWaves TV to chat about Amazon's early days and what learnings from that period can be applied today for everyone else who isn't Amazon.

➜ You can watch the replay of the interview here.

The short explanation: Keep things simple.

We kept a lot of experiments simple on purpose. Manual work helped us test and understand improvements before implementing a fully automated solution.

In the same spirit, we simplified the physical aspect of fulfillment by moving complexity to software components. The laws of physics make it hard to test, experiment, and scale automations. Software, meanwhile, can always be changed with enough resources available.

You don't need 100 warehouses and an insanely complex fulfillment network to profitably meet customer delivery expectations because you can handle a lot of the complexity via software.