The missing ingredient in what you're asking is testing.
When all's said and done, if your client is in business to sell something, and your job is to help them succeed (which it is), you have to consult your audience to know what's going to work. You started with some survey data to guide the development process. Now you need to test.
A logo, a slogan, an ad or an identity program stands or falls on how the intended audience responds to it. You can be sure you know that answer, and you may even be right, but there is absolutely no substitute for testing.
Take your top three or five "most likely" candidates out into the field, show them to some people from the target audience and find out how they react. Don't let people look, ponder, discuss and overthink the thing. You're looking for the reaction, the instant take, not a critique.
Show the person one item for a few seconds, then hide it. "What impression did you get about this company?" or "What impression does that leave you with?" will elicit responses that sometimes make your jaw hit the floor. The "extra credit" question comes at the end, after you've shown all your candidate designs/logos/taglines: "Which one stays in your mind?" can tell you more than a month of brainstorming.
This doesn't have to be done on half the local population. Two or three dozen representative people, in most cases, is enough to let you know if you're in the ballpark.
At the very least, you'll be able to present the data to the client and say "Here's how 80% of our test subjects reacted when we showed them this, which is why we recommend it."
A different technique that helps in the studio is to stop asking "Which one seems best?" Flip the question around and ask, "Which of these is least likely?" It's remarkable how this switch in viewpoint can give you a bit of distance and a fresh look. It also works well on CEO's who are nervous about committing to a final decision.
Ultimately, though, it's about the audience, so that's where you have to take it to know the answer.