You might find this guide useful: https://tomprof.stanford.edu/posting/1085
The key organizing questions are as follows:
• What problem have I worked on?
• Why would anyone work on this problem?
• What is significant about what I have done?
• How has my work made progress on the problem?
Here's a sample structure for a 45-minute research job talk:
Content * Time * Target Audience * Detail Level / Purpose
Background * 15 * Everyone present * Your parents would understand it
Your approach * 10 * People in related fields * Show you know the field
Your results * 10 * People who work in your field * Show that you are the world expert on something
Summary * 10 * Everyone in the room * Relate your results to the big picture
I found this structure particularly useful because it's easy to lose sight of how specialized one's work is. You need to make the talk relatable, and to speak to a broader audience than, for example, a conference presentation or a journal article. Even though these faculty might be in the same discipline and have phds, there is every chance they don't follow your subfield. Only part of your talk needs to demonstrate the cutting edge aspects of your research. Most of it should be broader.
Of course, your mileage may vary, but I found this advice convincing (and it worked for me!)