It makes me feel extremely overwhelmed knowing how much time it would take to make edited screencaps for the story that I have in mind.
Going for the sandbox-like route will perhaps make you feel less overwhelmed, but it will not lead to less works. But in the end, it's before anything else a question of practice.
I slowly learn to use Daz Studio, and while I don't really have much time to do it, what sometimes lead to months without opening it. Yet I can clearly see that I'm adapting and finding my marks, doing things faster now that when I started. By example, I don't like the default pose and made my own one for my test. I needed more than half a hour for this, and now I can pose a figure in less than 5 minutes ; yet I'm probably too perfectionist, especially for the poor results I still get.
And I'm sure that most dev, whatever it they use Daz Studio, Blender or and Illusion studio, would tell you the same. It's a whole new workflow, it just need time.
Time to memorize how and where each action is done, for it to become instinctive. Time for your brain to fully register what action lead to what result, for it to come faster to the process you've to pass through to reach the result you want. Time also to learn two new skills, "posing a 3D figure" and "composing a 3D scene".
It's like illustrators. They aren't born with that ability to compose their drawing. It's something they learned through time and practice. If at some point they can draw a complex scene without having to think about it's composition, it's because they spent hours looking at the paper sheet, and trashed thousand of sketches, until their brain learned how to do this naturally.
And the same goes for the story progression and its coding. Two others new skills that you are learning, and two others tasks that you'll do faster and faster with the time going.
So perhaps that, instead of switching to a sandbox-like approach, what you need is just more time to adapt to the tools you are using and that new workflow.
The idea I got in mind now is that you play through the VN part, read the story with perhaps a choice or 2.
Then into sandbox in which you can choose which stats to raise, and then, when you reach a certain amount of points on one of the stats, that it then goes to the VN linked to that stat.
So, as I hinted previously, the same amount of story, but with more code, and a more complex one, and with also more CGs.
I think that you choose the sandbox-like approach by using a player point of view. It's basic to witness and to play, so it seemed to you that it would also be easier to do. But it will not be the case.
Like you have a strong idea about your story, and are more interested by its none-con part, you should instead goes for a pure Kinetic Novel. It's less CGs than for an interactive VN, and even more than for a sandbox-like, and it's also less code, and an easier one.
And the time you finish it, you would have learned everything you need, and could do something way more interactive and complex, and do it in less time.
As far as VN's go, I only knew of the romance/anime style VN's and I wanted to combine them. That's when I first came across itch.io and found games similar in style to what you suggested, and that's when ideas started growing.
