Last weekend I watched Amazon’s Jeff Bezos announce on 60 Minutes his plans for 30 minute drone delivery to locations within 10 miles of Amazon warehouses. I’ve been thinking about interview all week, but not because of the futuristic drone delivery approach. What most impressed me was the one minute segment with Amazon VP Dave Clark describing their chaotic storage practices which is “optimized for using available space.”
Amazon does not store physical items by what they are, but rather by where in the warehouse they can best fit to minimize wasted shelf space. For example, if they get a shipment of 500 copies of a specific book, they do not store the 500 copies in one location together. Instead, they distribute the books to different areas of empty shelf space where their location is recorded in a database. A specific bin Clark walked to contained a book on Buddhism and Zen sitting next to a Mrs. Potato Head. According to the 60 Minutes piece, Amazon is now able to store twice as many goods in its centers as it did five years ago.
60 Minutes quickly moved on from talking to Clark, as it was not the crux of the piece, but I haven’t stopped thinking about it since. Storing twice the amount of product in the same amount of space is a freakish accomplishment that was achieved in a relatively pedestrian way. Amazon was able to double storage capacity by changing nothing but the way they arranged products on shelves.
Amazon’s version of the Pareto principle
In reading about business and productivity alike, the Pareto principle is frequently referenced. Focus only on the 20% of your customers that generate 80% of your income. Eliminate the 80% of your workflow that produces only 20% of your work. In theory using the Pareto principle to analyze data is awesome. It is shockingly accurate in describing data sets in many different disciplines. However, it is not as easy to apply to reality as it may seem. Many companies and jobs are not agile in a way where the ability to trim the deadweight exists. And even if you could eliminate or reduce focus on the 80% of x that isn’t as profitable or productive, do you run the risk of a new 80% appearing in the new, leaner population? Is that a never-ending cycle of hyper-specialization?
Amazon took a finite amount of shelf space and optimized it by storing things in a way that likely made sense to nobody at first, but then made sense to everybody after the rationale was explained. The efficiencies this system presents are many and easily attainable. Amazon knows the exact dimensions of every product in its warehouses and the exact dimensions of vacant shelf space. The optimization is easily done mathematically because this is an exact science, at least from a pure storage space perspective. Once the leap of faith was taken to store unrelated SKUs next to each other, the problem become much easier. Accepting that it worked was perhaps the hardest bias to overcome.
I’ve been thinking about how companies in different areas of business could achieve similar results. How can you produce a 2x increase in something while essentially operating with identical constraints? To me, this seems like a more realistic way to approach optimizing a company or a job than using the Pareto principle.
What fascinates me so much about this 2x increase in storage was the simplicity of the practice. Of course there is sophisticated software infrastructure running the operation, but the essence of the solution was basic. What other solutions exist for companies that are so obvious, nobody thinks about them? Sticking an extra plate between two bags of flour in your kitchen cabinet might not make sense just because they fit, but Amazon used that concept at scale to increase storage capacity by astonishing proportions.
What other low-tech solutions exist for companies that are real game changers? While innovation will continue to improve by leaps and bounds due to the usage drones and robotic warehouses, Amazon proved that ‘the next big thing’ might be easier to come across than one might think. As people continue to talk about the future of Amazon’s drone delivery service and what that might mean, for me their use of chaotic storage is what stole the show.