I believe there's some sort of misconception here. How good a design is is not in correlation with how expensive or premium something seems.
A design can't be "too good" for its users. In the package of design (the art of applying knowledge and best practices into a solution) knowing one's target group and speaking to that target group accordingly is also included. In the package they can't be separated.
Certain aspects, like interaction design, IA, color picking, etc. are practices based on cognitive science and platform guidelines. They are universal and can't be "too good" for any target group (small disclaimer here for market variations, what works and what doesn't eg. color wise in specific markets).
Target group aspects, such as visual expression, communicative tone, stock imagery etc. are target group oriented and WILL differ between products depending on what feelings you want to evoke in your users/customers. Here it's easy to to form misconceptions if you don't know your customers/users well.
To illustrate: A user browsing for some product has never thought "This seems a bit too reliable to me.. I feel like I understand what I'm getting here a bit too well.. If there were just a few elements of uncertainty I'm sure I would definitely buy this product..". Ie, making a design worse will never convert a user. However, if you're packaging the trampoline you're selling like it's the next Rolls Royce then customers will think you're dishonest, unreliable and false and will find business elsewhere.
So in summary, a design communicating to your users in an inappropriate fashion isn't a good design. And again: How good a design is is not in correlation with how expensive or premium something seems.