When giving design feedback, don't solve the problem for the designer. Tell them the goal that's NOT being met, and let them figure out how to achieve it.
This is a simple switch to make in how you give design feedback, but it makes a huge difference.
Another way to think about it is: give designers goal-based feedback, not design-based feedback.
For instance:
β’ [Design-based] Can you change this button color to a darker blue?
β’ [Goal-based] I feel like the site is too informal, and one of our goals is "Users should feel that we are professional and trustworthy"
The first is the epitome of cringe-worthy design feedback (no, strike that β it would be "bigger logo" or "make it pop"). The second will MAKE THE DESIGNER LOVE YOU. You're letting them do their job (which is, for the record, translating a business goal into an interface).
Another example:
β’ [Design-based] Let's use a carousel to show multiple images here.
β’ [Goal-based] There are 3 primary messages we want to communicate with this page, and right now, only one of them is coming through.
Again, the first is awful. Not only does it prescribe the solution, but that "solution" happens to be "let's make users wait 10,000 milliseconds without scrolling or looking at anything else, in order to read 3 sentences that they can only navigate between by clicking tiny half-invisible dots."
Carousels are a stain on digital interfaces. Seriously.
One more:
β’ [Design-based] Let's move this button to the right side of the screen and make it bigger.
β’ [Goal-based] The primary goal of this page is to get people to sign up, and right now, very few people are clicking the signup button.
When you give a designer feedback like "move this button", they ultimately have to figure out why you want them to move the button anyways. Better to just give the why in the first place and save you both a meeting next week.