My sideproject, fantasysp.com, cost basically nothing to start around 7 years ago. I found free baseball boxscore data online and bought the domain and ran it

My sideproject, fantasysp.com, cost basically nothing to start around 7 years ago. I found free baseball boxscore data online and bought the domain and ran it on hosting I already had. The main cost was my time involved to code everything. "True" costs didn't arrive until I thought I actually had something worth investing in. Which meant I had to purchase boxscore data for the sports I couldn't find. So to answer your question... ideally your project should cost very little to start with. Once you discover that you might be on to something, then slowly increment costs. Baby steps.