1. Not validating the problem statement enough. Is this really a problem?
2. For a B2B product, I think its important to think through early on whether this is a problem they are willing to pay for. Often times, this is an after thought and expensive to pivot.
3. Giving up too soon. Its easier said than done to validate the problem statement. Sometimes this take iterations where you get live feedback from real users. So you might be dancing around the problem space for a bit and that's okay.