Looking at the pricing, a "credit" is dependent on the size and complexity of the content, but you haven't provided me with any idea what size or complexity mean, so it's a complete black box.

if anyone looking to solve this with open-source, try Unstract - https://github.com/Zipstack/unstract

If I understand correctly this tool focuses on ability to create easily extractors from documents. I'm wondering who is your target audience? If it's a company operations (accounting, procurement, etc),you missed it, they want to setup once then use it and access documents. If it's for developers to quickly build extractors for different company ops, then I'm missing API, scoring, integrations.

Do you have an API?

What is the privacy policy?