If you want to write fiction, do almost no research
He says research is more fun than writing, and an excuse to avoid it.
Freakonomics Radio Network · Harlan Coben Turned His Books Into 12 Netflix Series | Better in PersonHarlan Coben, a novelist who makes his living writing thrillers, has blunt advice for anyone trying to write fiction: do as little research as possible.
Coben calls himself a lazy researcher, and he means it as a method. "I do not believe in research for a book," Coben says, especially for a novel, because the real job is to sit down and write.
His first reason is about avoidance. Coben says "you will find any excuse not to write," and you'll do anything to avoid the moment you start. Research, he says, is "much more fun than writing."
His example is a scene set on Park Avenue in New York. The temptation is to say you first have to go there, smell the hot dog stands, and watch people walk by. Coben answers no. "You've been there. You know it. You have an imagination. You have Google Earth. Write the damn scene now." You can go look later.
His second reason is about pace. He says readers have all hit "that fiction book, the novel, where the writer has done a lot of research and he gets lost in it," slowing the story to share factoids. Coben jokes that is not a problem for him because he does not know anything.
My takeaway from his Park Avenue example is to treat what you already know as enough for a first pass. Start with memory and imagination, get the scene down, and save any checking for after the pages exist.
Questions people ask
Won't readers catch mistakes if I don't research?
This is my view, not Coben's: small errors matter less than momentum, and most readers will forgive a wrong storefront if they care about the characters. Write the scene first, then check the important facts afterward with maps, photos, or a quick visit. A copy editor will also catch much of what you miss.
Does this rule work for historical novels or technical subjects?
Coben was speaking generally about a book, so this is my own caveat, not his: if your story depends on a distant era or a technical profession, you will likely need more upfront checking. Even then, his Park Avenue point can still help. Draft the human scene first, then fill in the specialized details.