I think it is difficult to judge based on a % only situation.
If you only have 100 users and you're only get 5 users coming back in the last two days, you might not have hit a market fit.
If you've got 10k users, that's a different story.
But I wouldn't use a 2 day window, as I think that is too short.
I'd use monthly active users (how many users login once a month) or weekly. That should be a more valuable metric.
Again, this depends on exactly what your service does, your target market, how often you expect them to login, but 5% doesn't sound bad to me for a 2 day window.
What happens when you extend the window?
