Indeed, they are. The authorization and approval process can be daunting, but if you can manage those roadblocks, Amazon will give you a free pipe into the data.
But beyond access, there are server costs associated with pulling and storing your data which depending on how much data you pull, can be expensive, especially if you are looking to keep your data refreshed (and thus accurate).
Another cost to consider is one of the most burdensome aspects of leveraging API data, maintaining the pipeline. Amazon’s data structure, reports, and API ‘versions’ are always changing, and staying on top of these changes requires a dedicated resource for ongoing maintenance.
So, despite the API’s being free-to-use, we’ve found a role for ourselves by aggregating the demand for data flow and spreading the associated costs across a large user base.