When I started to get deeply into the kinds of writing that forced me to learn things I didn’t already know, I thought close enough was good enough. That is, that accuracy wasn’t as important as making something believable. It was fiction, after all. Part of that, I think, was a native laziness, and part of it was worrying about truth getting in the way of a good story. We writers are all, at bottom, liars.

But I started to encounter situations where waving my hands and making something up wasn’t good enough. I could invent a street in Boston to put my bar on, but I couldn’t make up a story why the lions reside in the Boston Public Library.

So I started to shed my lazy ways, as well as I could, and learned how to do research.

The key to using research in fiction, of course, is not to let it overwhelm the story. Just because you’ve done all that work doesn’t mean it belongs in the book. I recently read a 977 page behemoth of a novel about the contemporary South that was larded with research, only thinly disguised, to the detriment of the story’s momentum. Readers can and do skip parts of a book they don’t find compelling.

Starting research may mean going directly to Google, that most tempting seductress. It’s easy to access, returns many results (some of them even appropriate to your need), and yes, it’s always there. And if you Google my address in Oregon, where I haven’t lived for a dozen years, you will see see my cat Jasper, long since gone, sitting in the end of the driveway. Know the limits of generalized online research and act appropriately. It’s also helpful to learn to construct narrow and specific queries to get what you want efficiently. And move on.

Further online research, more specific to your requirements, can be helpful, but there is always a tendency to dive down ratholes in search of that perfect factoid, which will eventually cut into your writing time.

Whenever possible, I go to primary and secondary sources for material. Primary sources are original information, as nearly as possible. In the case of the 977 page novel, the author based some of his research on actual slave documents. Secondary sources are usually based on the primary sources and often interpret the primary sources. Regardless whether these are physical or digital, these kinds of sources give more reliable and accurate information than random Internet sites. And there is a deep thrill in touching a document or object that is directly connected to the time, place, or people you’re writing about.

Far and away the most useful kind of research, though, comes from other people. Who do you know who has information that you want? Is it a police officer, a spy, a nurse, a librarian, and so on ad infinitum. In essence, how big is your Rolodex? Michael Koryta’s book Last Words relies on an intimate knowledge of caving which, he admits, he didn’t know anything about before he started.

The best part of doing research with live sources is that most people love to talk about themselves, their work, their hobbies, their experiences. Even an introvert can get over the hump of discomfort by asking simple leading questions: what was it like to do what you did? How did it feel? And that is the point your story takes on true verisimilitude. Readers are intelligent. They can tell when you know what you’re talking about and when you’re faking it, even if what you’re telling them is believable. Readers love to learn something new, as long as they’re not drowned in it. Give them that and make it accurate and you’ve got a reader beguiled.