Of course you start with the usual questions you must ask of yourself. For instance, who is your audience, how is the video going to be delivered to that audience, what end result do you expect from the video?

But, once that is done, you must identify the pivot points, or change points that will hopefully happen due to the activities of the event. What kind of action would accompany that change?

Next, what is your budget?

Then start looking at samples. You may not see coverage that exactly duplicates what you are after. But, from what you can see, are the producers effectively covering all of the components, especially those that we mentioned above, the change points? Are you seeing the goals, the change points happening with the participants? Is what you are seeing saying "this event is evolving in the direction that it should be? Or, are you just watching odd and end shots of the event as it ambles along to its end?

After many, many years of shooting events, we have built a complete and unique production process around coverage of events - especially far flung events like marathons. Camera people work independently, but with GPS trackers that tell the others where they are and what they are covering. This allows an individual camera person to find a story (or stories) and track them while preventing the others from wasting time and effort duplicating work that is already in process. It also permits communication of a moving story that others can pick up on if it is coming in their direction. Finally, if need be, a video frame can be passed on to the others to reference a specific individual or issue that has possible on-going value during the event. The process frees each individual camera person to identify an unfolding story, pursue it or pass it on to the team for further coverage, on-the-spot. The only other way to get that kind of coverage on a 26 mile long event, for example, is to have at least 30 cameras going. And, even then, most of the stories will be missed.