I consider that I lack a lot of information at this point on the data and the structure (unless I missed it), that I would normally have asked before doing a design. It doesn't fit a specific process but all of them.
Product Family and Product: How many family are we talking, how many products ? Do most account buy most products ?
Are accounts attached to countries ? can they change countries ? How many accounts in total ?
Are sales rep attached to countries ? How many sales rep in total ?
Finally, we always design with constraints. How much space do we have ? Are we designing a 20GB model or we have 100 GB ? It might impact some choices (+ it would be a good constraint for the exercise)
Indeed, when we start with Schema we lack a lot of the information we need to build the perfect schema. But, I assume that's the whole point of the exercise, to create a schema with the details at hand. For example, my initial schema was partially reassembling the one at the end of the Level 3 but both of them were presenting the same data flow. I think you shouldn't worry too much with thinking about the exact hierarchy or the size of the model at this point. You'll deep-dive into those at a later stage.
If I have to recommend something, focus on a higher level schema and by following the best practices (DISCO / PLANS). Use something like LucidCharts to help plot the structure.
I hope this helps and I wish you all the best with your build!
If you have any questions, feel free to reply here.