A lot of startup advice says to listen carefully to early users, but in practice the signals can be noisy.

For people here who changed direction early, what was the thing that made it obvious?

Was it weak retention, empty user interviews, bad conversion, inability to explain the value clearly, or something else?

I’m especially interested in signals that are easy to rationalize away at first.