1. Don't sell to techies. i.e. ignore any ideas about software testing, cloud base load testing, etc.

2. Pick a solid business domain you know really well. There is no substitute for domain expertise. Sometimes things that look really simple are actually mindblowingly complex. And vice versa. But it sounds like you don't have that so...

3. Do something in HR Tech. Every single company on earth competes for talent, and the HR software industry is already worth many 10s of billions, and is not remotely built out. There are hundreds, perhaps thousands of startups in this space already. Also, in HR software you're dealing with people, so there will never be any "one right approach" - you can always sell your crazy HR software product to at least a few wide eyed early adopters somewhere.