Try working backwards. Step 1: Pick your customers. Are they small business, enterprise or consumers? Are they in a specific vertical or are you addressing a c

Try working backwards. Step 1: Pick your customers. Are they small business, enterprise or consumers? Are they in a specific vertical or are you addressing a cross-cutting concern? Step 2: Start learning. Set up interviews and learn their business inside and out. Perhaps volunteer to be unpaid labor as a way to get an inside look at how their businesses run and where the inefficiencies are. Step 3: Find the right inefficiency to address. It should be one that people are willing to spend to address, not just one where people should be willing to spend on. Also filter based on whether your team is best equipped to tackle the problem (i.e. don't try solving a really hard problem that, if you're successful, Google or Facebook could bang out in an afternoon with small modifications to an existing product.) Also look for products/services that give you a durable advantage, whether it's because you're the only one with the right data or because switching is particularly painful or a host of other reasons...much has been written on the subject, so read up for more examples. Step 4: Go back and try to work up a demo of a product idea. It doesn't have to actually work (read do things that don't scale), it just has to show your vision. Step 5: More interviews with potential customers. These aren't sales calls, they're a chance to gauge interest and get feedback on features and potential pricing. You'll have a good idea whether you're ready to move forward or back to step 4, 3 or even 2 at that point, but if you hit on something, then your tech starts coding and your non-tech starts trying to find launch customers using the demo. At this point, you've probably read enough about the next steps to take. Good luck!