The general way to measure value creation would be to sum the difference between what both parties would be willing to transact for something with what they act

The general way to measure value creation would be to sum the difference between what both parties would be willing to transact for something with what they actually transact for it. In your example, the value creation is $0.2 ($0.1+$0.1) due to the existence of a perfect substitute. In other cases the value creation may be 10x the turnover or higher. For example, if there's a single surgeon that can give me a life saving procedure that costs me $50k when I would've happily paid $500k, then the value creation for me is $450k and perhaps $20k for the surgeon. In the case of Amazon, their numbers are probably an exaggeration, yes. They've assumed that the turnover = value creation, which is false reasoning, as we both recognize. It might be higher or lower. The real number might be 0.4x of what they stated, if I was to hazard a guess. But since the whole idea of value creation flies over the heads of so many people (and this misunderstanding of basic economics has profound policy implications), Amazon's exaggeration here is not a point I wish to focus on.