I didn't learn this from books, but instead from working with skilled senior designers, creative directors, and design executives. I try to put as many objective frameworks around aesthetics. Here is where having a tight Design Language System comes in really handy. As a consultant I've been doing a lot of this kind of work for clients, I'm working on an article about how they work for core77.... just haven't had time to finish it up.
Even without a DLS, you can start to hang some of those objective frameworks yourself. The goal of this is to create language to talk about form that is more intellectual and less reactive (I like it, I don't like, yay, yuk). When subjective language dominates the conversation, the HIPPO always wins (Highest Paid Person's Opinion). You want to avoid this at all cost because in almost all situations the HIPPO is not aesthetically trained or sensitive.
Different frameworks work in different situations, here are a few I typically use:
1) establish agreed upon brand principles.
What are the priorities and strengths of the brand and its product portfolio? Is it fun? What does fun look like across industries. Is it safety? What does safety look like across all industries. Is it Fun + Safety? What other things in the world have combined those attributes. To do this you will need to do a series of internal stakeholder interviews with execs, product managers, marketing folks and engineers. Once you establish and get alignment on that, you can have a conversation about what forms achieve and represent those principles most effectively. You can do simple force rank exercises, or set up 2x2's to rank different principles and objectively get to an agreed upon solution.
2) establish agreed upon user personas.
I usually break this into at least 2 but no more then 5 personas. Typically an "aspirational persona" representing the tip of the spear user and a "target persona" representing the higher sales volume. Fill out the persona with as much detail as reasonable, and build a "closet" of other items in that persona's life. Once alignment is established with the other decision makers in the process the conversation should be about what designs fit the persona's best. Use first names with personas as much as possible. Instead of saying "I like this one" the conversation should feel more like "I think Tony would gravitate toward solutions like 1 and 4 and not 2 and 3"
3) establish gaps in retail
This one is really dependent on the market and the organization's risk tolerance, but sometimes I'll put together a document showing the similarities in all of the competitive product. I did this a few years ago with sound bars which was easy, they are all black rectangles, so the opportunity was in making anything that was not purely rectangular and not purely black. Then the challenge is to what degree do we need to break that to be successful but still successful. It turned out that just a rounded front view and a grey cloth with a metallic weave was enough to move sales, but at least we could have a conversation about vs just throwing out crazy shapes or making another tee too black rectangle.
Those are the simplest ones I like to use. There are a few more complicated methods that require more dedicated time, resources, and money like foundational ethnographic interviews. I tend to avoid user testing of aesthetics as much as possible. I have never seen that lead to anything interesting. When I pitch this as aDLS project to clients I tend to use a combination of ethnography, retailer interview, and internal stakeholder interviews to build up a shared knowledge set and break that into three lenses (internal, retail, and user), this usually helps to ensure as much internal buy in as possible... maybe I should write a book about this...