At what point does self-hosting stop being worth it for a SaaS team?
retained text
I’m curious how other SaaS teams think about this. Self-hosting often looks extremely cheap at first. You spin up a $5–20 VPS, deploy something with Docker, connect a domain, and you’re done. But over time, somebody still has to deal with: * backups and restores * updates * SSL/DNS issues * monitoring * security patches * disk usage * broken containers * database maintenance * random downtime For a technical founder, some of this is easy. But even if the server costs $10/month, spending 2–3 engineering hours maintaining it can make the real cost completely different. I’m interested in where people here draw the line. Do you generally: 1. Self-host as much as possible 2. Self-host only non-critical/internal tools 3. Pay for managed versions once the business grows 4. Avoid self-hosting entirely What actually changed your mind one way or the other?
document fetched 2026-09-27 00:29 · canonical https://reddit.com/r/SaaS/comments/1wr1uj9/at_what_point_does_selfhosting_stop_being_worth