SaaS
How do you guys actually decide which feature request worth building?
I keep thinking about this because customers can ask for the same thing, but that doesn't always mean it's the right thing to build. Like sometimes one customer asks for something and it sounds like a great idea to build in the saas but then you look at it again and realize they might actually be trying to solve a completely different problem. And then there's also another side where you get 10β20 people asking for something, so it feels obvious that you should build it, but maybe those peop...
Read the full discussion on Reddit
This article was aggregated from Reddit. Click to join the conversation.
View on Reddit