Probably a couple of hours on support per day on average, some days less, some days more. The vast majority of users don't ask for support, they prefer to read the docs, read the community forum, goggle stuff or hunt-and-peck options in the app. Roughly 90% use the free version. If everyone used the paid version I could have a very large team, but the reality is the product competes with many free tools.

If you update the app or refine docs in response to previous support questions it does streamline the experience but there are always folks who just don't read docs and there are many who will purchase the app just for access to support so they can figure something out.

I'm sure some apps are more support heavy than others, but ours is aimed at system administrators and with that comes an assumed level of competence (in reality, many people are only in the role because nobody else could/would do it but even they are quite independently resourceful).

The disadvantage of users helping themselves is that you don't get feedback from them or learn about their use cases. Knowing how/why people are using your stuff is really valuable for development, so if I had the team for it then dedicated support engineers would follow up with customers early on even if they don't have issues.