Here are some ideas on putting together a good job talk, based on years of observation, experiments (failed and successful experiences of my own), and discussions with colleagues past and present. Below, I focus on the talk, the Q&A, and the one-on-one/small group sessions. And remember, practice, practice, practice. You only get a few shots a year if you are lucky, so it is better to fall flat on your face in front of a friendly audience with no stakes, rather than when the job is on the line. Finally, some additional thoughts at the end.

The Talk

Make it clear and well organized—the audience should know where you are going at each step—using transparencies or computer projection is useful here

Make the talk simple enough so that somebody who studies (or does) sewage maintenance can follow—not all attendees will be in your field—so a political theorist/Americanist/public administration/whatever should be able to follow your argument.

What is the puzzle, the question, that animates your interest and your research?

Identify the relevance of your question—who cares? The talk should be interesting, not arcane

Be clear about what your argument is and define the limits of its scope—what is it that you seek to explain and what do you think is important.

Identify some, not all, of the counterarguments/conventional wisdoms you are addressing—who says?

Be clear about your variables and methodology—why these cases?

You will not have time to get into the details of your case, nor should you try, as case studies, unless something is really cool, are likely to bore the audience. You could highlight a mini-case within one of the cases.

If using quantitative techniques, explain the setup of one of your analyses (not all of them), and show a few of the most interesting results. Your transparency or slide should use, dare I say it, color, to highlight the most important stuff.

Wrap up with a summary and the implications for future research, highlighting your next project.

In the Q&A, (the most important part in most departments)

Think before answering each question.

Focus on answering each question through the lens of your theory. Use your theory to highlight what is and is not more important.

Many questions will essentially ask how would you do the project if you did it just like the person asking the question might.

Be relatively brief in your answers—most of the audience will have questions (they want to hear themselves speak) so give them a chance to talk.

It is ok to stay “I don’t know” but not to every question. Be aware of what your limitations are.

Do not be defensive—people may find flaws in your work that nobody has seen before. The purpose of this exercise is to see how you think on your feet.

In the one on one or small group sessions,

Don’t ask about money--the chair will tell you.

Don't ask directly about department politics--if there are problems, people will usually tell you, particularly the individuals causing the problems.
Don’t ask about teaching load if you are interviewing at a liberal arts college or other institutions that are avowedly serious about their teaching.
Ask about teaching support—not TAs but rather computer projection, funding for simulations, workshops on teaching, etc.
Ask about the students and show that you are interested in teaching.
You can ask about research support but do not obsess about this either
Ask about the town.
Ask about their research and their teaching.
Ask about the requirements for getting tenure--the response is usually vague but can highlight a variety of interesting dynamics.
Show a genuine curiosity about the place.

Additional Thoughts

Show a sincere interest without appearing desperate or arrogant.
Be aware of the particular school for which you are interviewing.
Figure out its priorities and where you might fit.
Find out what the faculty members study (at least the ones in your field).
Do not fake who you are, but market yourself.
It has been a convention to thank the chair and perhaps others via email upon your return home--will not affect your status, but is nice.
The outcome of the process has something to do with your performance, but there are a bunch of factors that have nothing to do with you. It is like poker, in that you can play your hand as best you can, but sometimes the outcome is beyond your control due to department politics, individual idiosyncrasies, and the position of Mars relative to Jupiter.
After returning home and after you know the outcome, seek out feedback about what you did right and wrong so that you can do better in the future.

Good luck!!