Don't complicate it. Your customers should dictate what you spend your time on, not directly, but you should filter their feedback into features that fit the w

Don't complicate it. Your customers should dictate what you spend your time on, not directly, but you should filter their feedback into features that fit the whole. If there's nothing filtering out of feedback yet, just work on what's interesting and what you think they'll find useful. I use GitHub issues to track the backlog. But I typically just work on what I think is important or interesting right now, which doesn't even really require a backlog. If it's not important, or if it's uninteresting, it'll probably just sit in the backlog until there's enough demand to make it interesting to work on. It helps to dogfood your own product too, then you get a direct feedback loop on what's needed right now.