Lots of partially good ideas here but many of them still skip steps or suggest spending money you shouldn't. There is an MIT-based framework outlined in the book Disciplined Entrepreneurship that combines Lean Startup principles with other proven best practices to tell you exactly how to do this without spending much of anything.

It starts with identifying as many markets as you think might be yours based on what your proposed solution is. If you have a solution for teachers, your end users could be high school teachers, elementary teachers, private school, subject-specific teachers, special needs, etc. To the extent that these people will have different needs or will use the product differently, they are different markets. You want to identify all of the markets you can think of, then narrow it down to the 6-12 most promising.

Then, you want to conduct customer interviews with 5-10 people in each of your top markets. Your goal is to find out what their pain points are, what they currently do to solve their issues, whether they have the budget to pay for a solution, the most important needs for a solution, and more. Your goal is to narrow your markets down to one beachhead market that, if you dominated it, could get your startup to break even. That allows you to build a very specific solution meeting the needs of only one narrow set of users and, if successful, prevent startup death. From there, ideally, the next market only requires mild changes or additional features to expand into.

Once you narrow your markets to a beachhead, you would do more interviews and start finding out where these users aggregate online and offline, what solution they need exactly, what features are the most valuable to them, how they describe those features (specific keyword phrases that resonate with them) and so on. By the time you're done, you should know exactly what to build first, how to describe it in a way that the users will respond to, and exactly where to market to them online and offline.

Next you could design a solution using a prototyping tool and conduct interviews to see if people understand the product, whether it's intuitive and so on. By the time you actually start spending any money on developing the actual product, you should all but know that it's exactly what you need.

Hit me up if you want to chat.