When to Stop Revising

There is no objective moment when a piece of writing is done. There is no threshold you cross, no quality you achieve, no standard you meet that objectively determines the piece is finished. Every piece of writing could always be different. Most pieces could, with more work, be better in some ways. The question of when to stop revising is therefore not a question about the piece. It's a question about your relationship to the piece and about what finishing means and what it costs and what continuing costs.

This sounds like it should make the question easier to answer, and it doesn't. The absence of an objective standard means the decision is entirely yours, which means it requires a quality of judgment about your own work and your own relationship to it that's genuinely difficult to develop and that most writing advice doesn't address directly.

This piece is an attempt to address it directly.

The Two Kinds of Unfinished

The most important distinction in knowing when to stop revising is the distinction between two different kinds of not-done.

The first kind is genuine incompletion. The piece has problems that a competent reader would notice and that would prevent the piece from doing what it's trying to do. A structural problem that keeps the piece from developing its argument or its narrative coherently. A section that's underdeveloped in ways that leave the reader without what they need to follow what comes next. A line of reasoning that doesn't hold. A character whose behavior doesn't make sense in the context the piece has established. These are real problems with real effects on the reader's experience, and they're the kind that revision needs to address.

The second kind is perfectionism-driven incompletion. The piece is functionally complete, it does what it's trying to do, a competent reader would receive it as finished, and you're finding more to change because finishing is frightening and the revision gives you a reason to not finish. The problems you're finding at this stage are real in the sense that changes could be made, but they're not real in the sense of being problems a reader would experience as problems. They're the voice of the perfectionism that keeps the work safely unfinished because finished means evaluated and evaluated means potentially found wanting.

The difference between these two kinds is felt rather than objectively determinable, and the felt difference requires honesty to access. The genuine incompletion tends to feel like something is missing or not working, a specific sense of a specific problem that you can articulate even if you can't immediately solve it. The perfectionism-driven incompletion tends to feel like ambient dissatisfaction without a specific addressable problem, the sense that it's not quite right without being able to say what not quite right means in terms of the reader's experience.

The Useful Questions

When you're trying to determine whether you're in genuine incompletion or perfectionism-driven incompletion, specific questions tend to produce more honest answers than the general question of whether the piece is done.

What would be better for a reader if I made one more revision pass? Not what would be different, because almost anything could be made different, but what would be specifically better for the experience of reading this piece. If you can answer this concretely and specifically, there may be more revision work to do. If the answer is vague or if you can't quite identify what the improvement would be for the reader, you may be in perfectionism territory.

Have I solved the problems I identified in the last revision pass? If the last revision pass identified specific problems and you've addressed those problems, the piece has improved in the ways you identified as needed. That's a form of completion, not because the piece is objectively perfect but because you've done the work you identified as the work.

Would I recognize this piece as finished if someone else had written it? This question cuts through some of the personal relationship to the work. The piece that someone else had written and that you were encountering as a reader, would you experience it as finished? If yes, the additional revision you're considering may be the perfectionism rather than the piece's actual need.

And what specifically is wrong with it? If you can't answer this with something more concrete than it's not quite right or I think it could be better, the answer may be that the piece is done and the feeling of not-right is the feeling of having finished something rather than the feeling of genuine incompletion.

The Outside Perspective

The most reliable external check on whether you're in genuine incompletion or perfectionism territory is a trusted reader who can tell you what the piece is actually doing.

Not whether it's perfect, because nothing is perfect, but whether it's doing what it's trying to do. Whether a reader encounters it as a finished piece or as something that clearly needs more work. Whether the problems you're finding in the latest revision pass are problems the reader experiences as problems or problems only visible from inside the writer's relationship to the work.

This requires being honest about what question you're asking the reader. Not does this need more work, because the answer to that question is almost always yes, but what is not working for you as a reader. If a trusted reader reads the piece and can't identify what's not working, and reads it as a finished piece, that's meaningful information about where you are in the process.

The writer who is in genuine incompletion tends to receive from trusted readers the same concerns they had themselves, confirmation that the specific problems they sensed were real problems the reader experienced. The writer who is in perfectionism territory tends to receive from trusted readers feedback that the piece works, that it does what it's trying to do, that the problems the writer is finding aren't visible from the outside.

The Cost of Continuing

Most writing advice treats the question of when to stop revising as a question about the piece. This piece is treating it partly as a question about the cost of continuing.

Every pass of revision has a cost that's worth accounting for honestly. The time is the obvious cost, the hours or days or weeks spent on additional revision passes that could be spent on new work. The attachment cost is less obvious but equally real: each additional revision pass deepens your relationship to the piece in ways that make the eventual release of it harder. The piece you've revised thirty times is harder to send out than the piece you've revised three times, not because it's better or worse but because the investment is deeper and the vulnerability of releasing it is proportionally greater.

There's also the opportunity cost of the pieces that aren't being written while additional revision passes are being made on a piece that may already be complete. The revision that goes on past the point where the piece is genuinely improved is using creative resources that would be more productively spent on the next piece.

These costs aren't arguments for stopping revision prematurely. They're arguments for being honest about what the additional revision is actually doing. If it's addressing genuine incompletion, the cost is worth it. If it's serving the perfectionism, the cost is being paid for something that isn't making the piece better.

Calling It Done

At some point the decision to stop revising is a decision rather than a discovery. There's no moment when the piece announces itself as finished. You decide it's finished. The decision is imperfect, made under uncertainty, possibly wrong in either direction. You might stop too soon and send out something that needed more work. You might have continued past the point where the work was improving anything.

The decision gets better with practice, with the accumulated experience of finishing pieces and discovering in retrospect whether you stopped at the right point. The piece you thought needed more work that turned out to be fine. The piece you thought was done that turned out to have a problem you missed. These experiences calibrate the decision-making over time.

A few practices that help with the decision.

Set a revision limit in advance. Not because the limit is necessarily the right number of passes, but because having decided in advance how many revision passes this piece will get prevents the number from expanding indefinitely based on the perfectionism's always-available reasons to continue. The limit creates a container that the decision to stop can be made within.

Wait before the final decision. The draft that feels like it needs one more pass today often feels done after a few days away from it. The distance that reveals what the piece actually is also tends to reveal whether the incompletion is genuine or perfectionism-driven. Revising immediately on the feeling that something isn't done tends to produce less accurate assessment of what the piece needs than revising after some distance.

Use a specific rather than a general criterion for stopping. Not done when it's perfect, which is never. Done when the three problems identified in the last read-through have been addressed. Done when a trusted reader reads it as finished. Done when you can't answer the question of what would be better for a reader concretely and specifically. Specific criteria produce better decisions than the unachievable standard of perfection.

The Piece That's Good Enough

Here's what finishing a piece of writing actually means, stated as directly as possible.

Finishing a piece means deciding that the piece is good enough to release, not that it's perfect, not that it's the best possible version of itself, not that there's nothing more that could be done with it. Good enough to do what it's trying to do for the reader it's trying to reach. Good enough to represent the work fairly without embarrassing you. Good enough that the next revision pass would produce improvement not worth the cost of another revision pass.

Good enough is a standard that can be reached. Perfect is a standard that can't. The decision to stop revising is the decision to apply the achievable standard rather than the unachievable one.

This decision requires the acceptance of imperfection that's one of the hardest things in the creative life, the acknowledgment that the piece is as done as it's going to be and that as done as it's going to be is good enough, and the release of the piece into the world where it will do what it can do.

The piece you finish does something in the world. The piece you keep revising does nothing except keep you from finishing it. Stop when it's good enough.