Although I haven't used it myself, I've read this approach a few times on HN: If you can come up with a good solution to an existing problem, perhaps an app of

Although I haven't used it myself, I've read this approach a few times on HN: If you can come up with a good solution to an existing problem, perhaps an app of some sort, create a landing page for it using some mock-ups (e.g. using Sketch, another thing I haven't used!) along with a mailing list subscription to indicate their interest... if it picks up, you know it's worth dedicating resources to the problem. Then when you're ready to release the software you know you have some potential users/customers - they could even have a trial of early versions and give valuable feedback. There are downsides, of course, e.g. giving away your idea to people who may be able to implement it quicker than you do - but they won't have your mailing list!