IIRC, the Jupiter Litterbox accumulated close to 130M JUP tokens — roughly $30M+. While we all obviously hope to see Pyth DAO revenue and the treasury grow to that level as quickly as possible, I expect the scale here to be much smaller at first and to increase gradually over time.
That said, I personally agree with you that the Pyth DAO should start thinking about the end use sooner rather than later, to avoid the risk of a future supply overhang.
My suggestion would still be to kickstart the reserve as soon as possible. Then, during Q1, once we have more data on buybacks and a better understanding of how revenue is growing, DAO members can spend more time defining what to do with the accumulated tokens — whether that’s burning (and at what percentage), funding growth initiatives, or a combination of the options you outlined in your post.
Yes, absolutely — this won’t be easy. I generally believe fees should be handled conservatively at first and increased gradually over time. It’s better to move to something like $0.001 and preserve usage rather than jump straight to $0.10 and lose most of it.
Some early churn is just part of the process. What really matters is building something that can last. Having paying users from the start helps align incentives, pushes everyone to care about data quality, and gives the network real resources to grow. Relying only on massive growth with no revenue is risky — sustainable traction comes from people finding enough value to actually pay.