ChatGPT and other Generative AI tools are now changing the world, offering great ways to enhance products and increase productivity. It’s therefore a wise choice to think if such tools could be integrated in your product. If you find a use case for Generative AI use, you will need to use an API like OpenAI’s GPT API, which costs money per token.

I have the feeling that some users may actually use those tools far too much, that could cost you more than other users. The question is, how would you make sure you have a predictable net income, and you don’t make a loss on some users using these systems extensively? Would you put a cap per user or would you try to do something more intelligent like a user-base budget that would dynamically change per user depending on the availability of that budget? Meaning, if 90% of your users are not using the AI tools much, you have plenty of resources to allow those users who do need them to use them with virtually no limit.

  • founderscurve@alien.topBEnglish
    1·
    1 year ago

    I’ve had a few companies that have had this kind of dilemma - one where the cost of a part of their product e.g. a specific feature has different unit economics that scale differently to their overall product, and can become an issue at scale and heavy use.

    a comparable example would be power users of a VPN or cloud storage solution.

    so the examples you use are good solutions, most commonly i see tiered pricing based on usage and variation on that same unit economics dynamic. eg. top up credit for more usage.

    i also see ‘reverse UX’ e.g. once you pass the threshold the experience starts to become poorer, downloads become slower for example. … until you pay more.