I used to manage a very popular online diet subscription service. My quick reaction is that you are addressing a pain point that many people have -- it's annoy

I used to manage a very popular online diet subscription service. My quick reaction is that you are addressing a pain point that many people have -- it's annoying to plan meals and map grocery purchases to them, while the "shop first and figure it out later" approach is inefficient. BUT, in reality, people have a lot of foibles about what they do and don't like to eat, allergies, aversions, what's considered acceptable/very good/too fancy, what they will and won't buy, whether and how often they're cooking for 1/2/more, what's in season or easily available where they are, how expensive ingredients are, how long cooking takes, etc. Before subscribing, I personally would need a lot more color on the above and probably some sample recipes. As an MVP, it's a perfectly reasonable experiment to keep it simple, but I would think to be even moderately successful you'll need to generate some custom content based on preference variables -- or at least some profile alternatives (like, say, a vegetarian, simple/balanced meals, and foodie option per week), and some basic substitutions for things people don't like and/or aren't readily available.