The art and science of e-commerce pricing at scale

As a loyal Prime customer, I couldn’t turn down the opportunity to answer a Quora question about Amazon’s pricing systems. With only a couple minutes of searching LinkedIn and Amazon’s career site, it was easy to see that their approach to pricing—no surprise here—combines human and algorithmic elements.

When the goal is market share growth (e.g. Kindle with tablets, e-books) prices are set to reflect competitive concerns. In other cases, algorithms drive decision making. To quote from a retail pricing software engineer’s LinkedIn profile:

Led the grounds-up design and implementation of pricing systems for Amazon’s retail business and evolved them from proof of concept prototypes to large-scale distributed systems that are the brains behind every price on every item sold by Amazon worldwide. Drove automation and algorithmic sophistication across various pricing use cases ranging from competitive strategy to optimal inventory management.

The ability to change prices in real-time based on customer demand, competitor rates, etc. is part of Amazon’s e-commerce edge. It’s a combination of art and science that, for the most part, stays behind closed doors. But recent coverage from the WSJ on yield management in e-commerce and the NYT on Wal-Mart vs. Amazon price wars shows the topic spilling over into the public arena.