Don't get me wrong but this sounds to me like wishful thinking. If you design the app to be decentralized, you have to provide a mechanism to incentivize node operators.
Let's assume it would be as you say. Would you focus on incentivizing node running before deploying a sufficiently working prototype and seeing the usage report?
Incentivizing node running is not a priority for first version.
Why? It is not necessary in this case. If there were transactions, then yes, decentralization would depend on number of different nodes. However for chat app, the decentralization is there only to provide redundancy and availability.
If you're the only one who runs the node, the solution is not decentralized and will have a single point of failure.
The solution is decentralizable! That's the correct term. Yes if there is just one node it is not decentralized and that's how things first start off, but it would still still be decentralizable.
With all the respect, I don't think it's possible to create any meaningful chat solution with the proposed budget, especially the one that might compete with Discord.
This project is not meant to compete with Discord, but rather create a chat space so that hive communities do not have to rely on Discord and in turn promote hive ecosystem by creating share links (instead of discord share links), and allow guest/lite accounts to participate (as long as the channel settings allow it) to help with on-boarding.
And there is a huge difference between the POC and the production-ready solution.
In marketing terms, yes this is POC. But in software terms, it is a production-ready solution.