I’ve called myself a discovery writer before, and that’s still true. I don’t work from a locked outline, and I’ve said as much: outline the miscellaneous stuff, let the rest stay loose, let the imagination do the work without hand-holding. But there’s one place where I don’t discover anything. I plan it, in advance, before a word of the draft exists. The ending.
That sounds like a contradiction. It isn’t. It’s just a different kind of preparation than the one most craft advice assumes exists.
Two Different Kinds of Not-Knowing
Most of the pantser-versus-plotter conversation treats “discovery writing” as one thing: you sit down, you don’t know what happens, and you find out by writing it. That’s true for a lot of what I do. Scenes, dialogue, the shape of a chapter, I find those the way a discovery writer is supposed to.
Endings are different, and I think they have to be. An ending is a promise the whole book has been making, whether you meant it to or not. If you don’t know where a character is headed, you can’t tell whether the scene you’re drafting today is moving toward it or wandering away from it. So before I draft, I walk through potential endings with the characters. Not on the page. In my head, sometimes on paper, the way you’d rehearse a scene before performing it. I need to be confident they’ll get what they’ve earned before I trust myself to write the two hundred pages that lead there.
Rehearsal, Not Prophecy
This isn’t the same as outlining a plot beat by beat. A rehearsed ending isn’t fixed in stone, and it usually deviates some from the version I started with. What rehearsal gives me isn’t a guaranteed outcome. It’s confidence that the character’s performance, meaning their choices, their capacity for change, what they will and won’t do under pressure, is solid enough that when the draft finally arrives at the ending, it isn’t a surprise to me. It’s a confirmation.
That’s the part that took me a while to articulate even to myself: I’m not planning what happens so much as I’m planning who’s capable of what. Once I trust that, the actual events can still shift without the ending falling apart, because the character underneath it hasn’t changed.
When It Goes Wrong, It’s Never the Character’s Fault
Here’s the part that surprised me when I actually sat down and thought about it. When an ending gives me trouble mid-draft, it’s almost never because a character stopped behaving the way I expected. The rehearsal already caught that problem before it could happen. When I struggle, it’s structural. Pacing that’s built up more tension than the ending has room to spend. A technical knot in the plot that needs untangling before the character can do what they were always going to do.
That’s a useful diagnostic, honestly, and one I didn’t have language for until I had to explain it out loud. If your ending feels wrong, ask which kind of wrong it is. A character problem means you didn’t rehearse them enough, you’re asking them to do something they haven’t earned yet. A structural problem means the character’s fine, but the scaffolding around them needs work. Those get fixed completely differently, and conflating them wastes time solving the wrong one.
The Practical Version
If you want to try this without overhauling how you write: before your next draft session, don’t outline the ending. Rehearse it. Sit with the character, ask what they’d actually do if the story ended today, and see if the answer feels earned or forced. If it feels forced, that’s information. Either the character isn’t ready yet, or the ending you had in mind was never really theirs to begin with.
Keep writing.