I took a startup from zero to 100k MRR as of last month over the last 5 years. I can tell you that cloud billing is the least of your concerns if you pay even the cursory attention to writing good queries and adding indexes in the right places. The real issue is the number of developers who never bother to learn how to structure data in a database for their use case. properly done, you can easily support thousands of paying users on a single write server.

What? No no, to be fast you need the whole database only in RAM! And SQL is hard so just make it a giant KV store. Schemas are also hard so all values are just amorphous JSON blobs. Might as well store images in the database too. Since it's RAM it'll be so fast!

/s

A bit hand wavy. It obviously depends on the business and what "least of concerns" entails.

In most cases businesses justify the cost of managed databases for less risk of downtime. A HA postgres server on crunchy can cost over $500/mo for a measly 4vCPU.

I would agree that it's the least of concerns but for a different reason. Spending all your time optimizing for optimal performance (assuming sensible indexing for what you have) by continuously redesigning your DB structure when you don't even know what your company will be doing next year isn't worth the time for a few hundred a month you might save.

I use CQRS with /dev/null for writes and /dev/random for reads. It's web scale, it's cheap and it's fast.

> I can tell you that cloud billing is the least of your concerns if you pay even the cursory attention to writing good queries and adding indexes in the right places.

I read this as "in building your startup, you should be paranoid about team members never making mistakes". I really try to read otherwise, but can't.