The reality is that probability * (work + cleverness) = success.
The size of the probability makes a big difference. If it's too low you're likely wasting your time.
You can improve it with good business/sales techniques, but if it's tilted the wrong way you're always going to be fighting against it.
Objectively, you'd need to know the hit rate for specific business ideas/approaches/fields.
Realistically, some fields are more likely to be successful than others. Analytics, sales aids, and off-shore arbitrage are far more likely to make money than - say - software support for the disabled, or starting a band.