Hello! Here are some tips about how to apply writing advice in a logical and productive manner. (Trust, but verify!)
1. Any “rule” of writing can be broken. Writing advice can be very helpful, but almost every writing tip ever given has been broken by at least one published work. In particular, authors with a history of success get more leeway to publish whatever they want because their editors trust them and can give them the benefit of the doubt.
2. However, it may not help you that another author could publish a book that did something otherwise hard-to-publish. For example, I’d generally recommend against using a character name in a title– most character names aren’t very interesting to prospective readers. However, some authors (including JK Rowling) have gotten such books published anyway. The main question is whether you can pull it off. If you’re an unknown author with no audience that’s submitting an unsolicited manuscript to a publishing house that has a 99.9% rejection rate, you’re facing a very different decision-making process than those authors probably did. The publisher’s assistant is going to send on probably around five manuscripts out of every thousand to her boss. She is looking for any reason to eliminate your manuscript. A lackluster title is probably sufficient. 
3. The publisher’s assistant does not have a rulebook in front of her listing which sorts of manuscripts have to be rejected. ”Oh, this manuscript has a character name in the title, so I have to reject it.” Obviously not. So think less about the advice (don’t use a character’s name!) and more about the goal (write a gripping title). For example, an unsolicited manuscript named Tom Smith is dead on arrival. But Barbara Bloodbath would probably warrant further attention. Even though it uses a character name, it sounds really interesting.
When I offer advice, I don’t want the reader to think that “trying X cannot work,” but rather that “if you want to try X, make sure that it does work by avoiding problems Y and Z.” For example, Barbara Bloodbath tells us enough about the character and plot to interest prospective readers. Tom Smith does not. There are very, very few stylistic choices that cannot work under any circumstances. (For example, having a first-person narrator describe his appearance by gazing at his reflection. But even then I think it would only be an instant-rejection if it happened in the first few chapters).
4. Think critically before acting on any writing advice. For example, there may be logical reasons advice to apply to your situation. Perhaps you’re in a genre where editors will accept a particular type of writing? Perhaps the market is changing or has changed to accept the type of writing in question? Alternately, perhaps the advice is just generally a flawed means to an end. For example, “a lot of editors complain about manuscripts that have too many adverbs, so you should not use adverbs!” Unless the logic behind the advice strikes you as sound– could you imagine an editor really tossing your manuscript because it had adverbs?– I would recommend disregarding it or at least investigating further.
5. If you want to try something unconventional– something that does not get published often–I would recommend caution. For example, in the adult fiction market, I can’t think of too many published books with talking animals or 5+ main characters. I would extrapolate that one is a gross mismatch for an adult audience and the other would probably suffer from massive character-development problems. (Ahem– developing 5+ characters in a single novel is hard). If you’re dead-set on doing something unconventional, I would recommend thinking long and hard about what it adds to the story. If your rationale is something like “it would be neat to try this” or “I couldn’t tell this story any other way,” I would highly recommend going back to the drawing board.
