The world’s best authors often suggest that rewriting is more important than writing. Hemingway’s classic line: “The first draft of anything is shit.” He rewrote the ending to A Farewell to Arms “39 times before I was satisfied.”

I believe this kind of thinking inspires quotes of lesser-known authors today, like this one on Goodreads: “I'm writing a first draft and reminding myself that I'm simply shoveling sand into a box so that later I can build castles.”

Nice metaphor, but when renowned authors rave about rewriting, is “shoveling sand” what they mean—getting enough words down, even in an undifferentiated mass, to allow you to reshape them into a first-class manuscript?

At least for books written by experts in business, medicine, law, policy, and other professions, I don’t think so. With people I coach, I insist on a first draft that is at least half-decent, or as I recommended in an earlier post, your “crappy best.”

If you shoot for a first draft that’s “70 percent” on target—something I’ve advised before—the second draft should move you to, say, 85 percent. The third should hit the 95-100 percent range—the A+ territory.

For me, three drafts are the charm. To be sure, on your computer, you make edits as you go. Sometimes you may feel like these amount to 39 drafts. But I advise you to plan on only two full rewrites, creating just three “printable” versions of each chapter.

And for three reasons. “Conservation of freshness” above all. If you want great writing, you can’t edit it into shape if you’ve half memorized what you’ve written. It’s too familiar. You can’t see the flaws. Getting it right would be like trying to “edit” to improve your children’s loving faces. Not possible!

You need to keep distance. Become a stranger between drafts. The trick is, first, to minimize the number of drafts to reduce memorizing, and second, set them aside for weeks or months to put distance between you and your loving words. When you come back after a healthy break, your text doesn’t have a friendly look and familiar ring. You see the flaws instantly.

The second reason for limiting drafts is to speed completion of your manuscript. You can edit your life away on just the initial chapters. (I’ve done it.) That’s what happens if you don’t choose to draw a line on tinkering. Choose to stay on schedule instead. Doing five or ten drafts of early chapters will delay work on later ones—delay that forces you to break deadlines.

The third reason is that, in a book, you don’t really refine your message until you’ve written the whole thing. Your thinking evolves with every chapter. So why polish the first few chapters if you’ll have to edit them substantially later to align them with your final message flow? Your endless tinkering early on will be an utter waste.

For all three reasons, it’s better to write 70-percent drafts of all chapters. After time passes, rewrite all of them fresh as if you’re editing someone else’s prose.

As novelist and Pulitzer winner Bernard Malamud said, “I would write a book, or a short story, at least three times—once to understand it, the second time to improve the prose, and a third to compel it to say what it still must say.”

I’m with Malamud. In drafting a new chapter, don’t start out by producing a mass of poorly shaped ideas. Go for more than a sand pile. Produce a prototype castle that shows you understand your direction. Then, with just two rounds of refinements, with time to partly forget the specifics in between, turn this crude model into a castle worth an A+.

[Revised January 2020. Originally published October 16, 2014]