Overwrite Then Curate
Draft three times what you ship, finish every first draft, then judge them together
- Difficulty
- Moderate
- Time to result
- ~months to results
- Steps
- 6
- Confidence
- 90%
Rubin's sequencing rule separates origination from craft. If an idea is new and flowing, take it all the way to a complete first draft but do no revision past that, then move to the next piece. Do this until every intended piece exists in rough form, and deliberately originate far more than you will ship: to release ten, he works on thirty, because the best of three projects merged into one will always beat the first ten. Only when the whole set exists do you look at it holistically, which is when you see which piece is not working, where the connections are, and whether the group should move backwards rather than forwards. The reason for the sequencing is that something discovered in piece eight routinely rewrites the right choice for pieces one and two.
Origin
Extracted from The Tim Ferriss Show
Core principles
- 01Everything good is a best-of, so build the pool before you pick
- 02A first draft finished beats a second draft polished
- 03Ideas made later change what the earlier ideas should be
- 04You cannot see what matters until you see the whole set together
- 05Revision before completion optimises pieces you may end up cutting
How to run it
- 1
Set the ship number and multiply it
Name how many pieces the finished work needs, then commit to originating roughly three times that many. Rubin works on thirty to release ten.
Pro tip Treat the surplus as the point, not as waste; the discards are what make the survivors good.
- 2
Ride the flow to a complete first draft
When an idea is new and moving, get it down as fast as possible and finish the thought all the way through rather than stopping partway.
Watch out An unfinished idea cannot be judged against the others later.
- 3
Stop at draft one and move on
Do no revision past the first draft. Close the piece and start the next one.
Pro tip If you feel pulled to polish, that is usually avoidance of starting the next blank page.
Watch out Polishing now is polishing something you may cut.
- 4
Repeat until the whole set exists
Keep originating until every piece in the enlarged pool has a great first draft, before any second pass begins.
- 5
Lay everything out side by side
View the entire set together. This is the only condition under which you can see what is important, where connections exist, and which single piece is not working.
Pro tip Physically arrange them in one view rather than reviewing sequentially from memory.
Watch out Seeing them one at a time is what produces a Frankenstein result.
- 6
Select, then revise
Cut down to your ship number and only then begin the craft passes on the pieces that made the cut.
Pro tip Be open to the group telling you to go backwards, toward more abstract or less specific, rather than forwards.
In the wild
Asked how much to revise a given album track before moving on, Rubin says he would get a full first draft of the idea, do no revision past it, and move to piece two. If the release needs ten tracks he would be working on thirty to get to ten, because everything shipped is a best-of. Picking the best of three projects and making that one project is certainly going to be better than the first ten you happened to write.
→ The released album is a curated top third rather than an over-polished first attempt.
Ferriss describes working with an illustrator on insignias for fictional clans. After two rounds of revisions on two different insignias, he realised he had to see all of them side by side to get any semantic interconnectedness, and could not keep refining one in isolation. Rubin agrees: seeing them together may tell you to go backwards rather than forwards, perhaps more abstract and less specific, and it is the only way to spot that number six is not working.
→ Refinement decisions are made at the level of the set, avoiding a Frankenstein collection of individually polished mismatches.
Common mistakes
Revising piece one before piece ten exists
Something that happens in the eighth piece routinely informs a choice in the first two. Early polish locks in decisions you have not yet earned the information to make.
Originating exactly what you plan to ship
With no surplus there is nothing to curate, so weak pieces survive on necessity rather than merit.
Judging pieces sequentially instead of together
You only learn which one is not working, and what the group's real identity is, when the whole set is in view at once.
Is it for you?
Best for
Any multi-piece project with selection built in: albums, essay collections, story cycles, design systems, product concepts.
Not ideal for
Single commissioned deliverables where the brief is fixed and there is nothing to select between.
From the transcript
“If I wanted to release 10, I would be working on 30 to get to 10 to release. So, everything's a best of.”
“Get them all to the point where you have a great first draft of everything with the ideas before ever going back”
“It's impossible to know until you see the group. And then you understand what's important.”
From the episode
#649: Rick Rubin, Legendary Producer — Timeless Methods for Unlocking Creativity, Secrets Hidden in Plain Sight, The Future with AI, Helpful Distractions, Working with Strong Personalities, Breaking Out of “The Sameness,” and More