We are not split testing.
That is when you set a price and see how that price performs vs a different price. There are different ways to do this:
You could set a price on a group of products and compare against another group of products with price unchanged (treatment vs control).
You could expose different prices to different groups of buyers (A/B testing)
You could change price one day and then reset it back to its original price the next, repeated in sequence (switchback experiments).
Although split testing is the correct method to use when performing price experiments, it is a tedious and error-prone process on Amazon. The best approach is to compare price performance with how it performed in the past. This is all made possible with our AI, which we've taken to the next level. Our advanced computer algorithm teaches itself through trial and error, experimenting against past data and optimizing pricing at scale. With each iteration, it gets smarter and more efficient, allowing Amazon brands to compound those learnings and stay way ahead.