I think it's especially important to make sure that you interview customers and find out what their problems are before you even think about writing a single li

I think it's especially important to make sure that you interview customers and find out what their problems are before you even think about writing a single line of code. You should strive to get a mix of qualitative (because the stories you hear will let you empathize with the customer problems) as well as quantitative via surveys / small ad experiments. Once you have some signal, then proceed to start exploring solutions (and again, this isn't about writing code, you can do this entirely with mockups/napkin drawings, whatever it takes to get your point across) and test those solutions with the customers you interviewed earlier. Once you have some evidence that a) you have a problem that customers will pay to solve, and b) you have a solution that you have validated THEN you proceed to start writing the code. The biggest change at my company is doing product development this way: understand the customer problem and the solution before writing code. Hope this helps.