For me, it depends; I'd first take a step back and think about my target market. I'd think about whether or not the market you're trying to make this product f

For me, it depends; I'd first take a step back and think about my target market. I'd think about whether or not the market you're trying to make this product for both sees the problems you're addressing as problems (or can be convinced to) and is large enough to support the costs to build it and generate revenue. While I get that ycombinator itself looks more toward the people than the ideas if you're alone on this and jumping into it it's something you need to think about right away. If you spend your time building a product for some niche market you may be wasting your time going that route as well (although the experience gained and the connections made through doing so may be beneficial). Does your target market already have a product that does x? If so what are you doing that is better than than the competitors? If so I'd consider putting aside my work and plunging into it if I could afford the dip in my income as you mentioned you could above. Otherwise I'd either work on my idea as a hobby/side project or over weekends (and some very long nights). The actual business related activities like advertising and considering how you'd generate revenue itself can wait until you've shipped but if you're at it alone it's a lot harder to gauge when you need to re-evaluate and change your idea if it's not working or there isn't enough interest. We drink our own koolaid per se. Note: I don't actually need you to answer these questions I more posted them as what I'd suggest you internally inquire about. Also I get that there is a culture of build first ask questions later so my opinion may be of the minority. I do think though that calculated risks are the best kind and "No wind is favorable to a ship without a destination."