First, a little alliteration, just for the fun of it. It’s a common concern for the conscientious communicator: when have you collected enough information to call your creation complete?

I’d love to tell you there’s a hard-and-fast rule for these sorts of things, but the honest answer is, even for an English major working among rocket scientists, “It depends.” Generally, the shorter the document, the easier it is to lay your hands on the proper information or acquire it through a short interview with a subject-matter expert (SME). Longer documents, of course, require you to access a bit more information from multiple sources. The best advice I can give, having learned the hard way sometimes, is a mixed bag:

• Start with references directly relevant to your topic: meeting minutes, technical reports, or topical interviews.
• Depending on the scope of the work, you might need to provide some context for your document. For instance, while my initial topic on some papers might have been the Ares I-X flight test, it was sometimes useful and necessary to discuss first the larger program of which Ares I-X was a part—the Constellation Program—and then delve down to a brief discussion of the Ares I Crew Launch Vehicle, which the flight test was designed to help prove. A technical paper might require a sentence or paragraph about each of the above—a full report might entail a full page.
• If you are dealing with a highly complex topic—say, nanotechnology or astrophysics—it is sometimes necessary to pull back the camera and go for a much wider view, talking about stars and galaxies or atoms first before talking about dark matter or molecular machinery.
• Much of your research will be driven by the two most important questions you ask at the front end of the assignment: who is your audience and what are they using the information for?
• When I was writing long academic papers on tricky topics for my undergrad and graduate degrees, I would get my hands on whatever seemed relevant and then stop when the sources were only tangentially touching on my topic or when they started referring back to articles I’d already collected. That worked pre- and post-internet, so I find that that’s still a valid strategy.
• Naturally if you’ve got to really get into the minutiae to get some details correct or to clarify what a specific detail means, then go back and find the best source you can.

I’ll write a separate blog on which sources to trust. In the meantime, happy researching!