Tim Ferriss — How to Use Writing to Sharpen Your Thinking

Why this is in the vault

The founder’s primary public output is Sanity Check, a written newsletter. Every skill in the vault that touches writing — research-brief, draft-review, voice-match, remix — is downstream of the question “how does this person actually think on paper?” Ferriss argues that writing is not just output of thinking but the substrate that allows thinking to be sharpened at all. The video’s specific gift is naming three protocols (morning pages, two-pages-per-day, three-pass revision) plus three proofreader prompts (highlight confusion, note mind-wandering, 10% rule) that map directly onto the existing draft-review and voice-match skills. It’s a tactical-discipline doc, not a thought-leadership think-piece — vault-worthy because it’s specific enough to actually change a workflow.

Core argument

Writing makes thinking visible to the writer. Without freezing it on paper, you can’t see undefined words, redundant claims, or unsharpened logic. McPhee anecdote (Princeton class) is the proof — Ferriss’s grades in every other class improved because the McPhee class taught extraneous-word removal as a habit.

The protocols:

1. Morning pages (Julia Cameron, The Artist’s Way; popularized by Brian Koppelman in the writer-screenwriter community). 1-3 pages handwritten, stream of consciousness, no editing. Two effects: (a) freeze worries onto paper so they stop occupying working memory, (b) give yourself feedback on where your thinking is currently sharp vs dull.
2. Two crappy pages per day (mantra borrowed from a career ghostwriter). Set a winnable minimum. The pages are allowed to be crap — the point is keeping the game in motion. Most days you’ll exceed two; on bad days, two is the floor. The compounding discipline is what produces voice.
3. Three-pass revision (Neil Strauss model): Round 1 for self (what do you think is good), Round 2 for fans (what will resonate with the people who already love your work), Round 3 for critics (what would the worst-faith reader latch onto). Each round shifts the writer’s audience-instinct.

Proofreader prompts (use lay readers, not just professional writers — Ferriss’s workaround when no pro is available is use a lawyer or paralegal, because legal training enforces precise language):

• Highlight anything confusing. Readers can like or dislike, but should not be confused. Confusion = revision target.
• Note where your mind started wandering. “When in doubt, take it out.” Slow passages are usually cuttable.
• The 10% rule. Tell me the 10% I should keep no matter what (the strongest signal) and the 10% I should cut if forced (the weakest signal). Forces ranking.

Cognition compounds. Writing systematically transfers to thinking systematically — Ferriss’s claim is that the discipline propagates beyond writing into general analytical sharpness.

Mapping against Ray Data Co

• The “morning pages” frame validates the existing /journal pattern, but exposes a missing one. The vault has dated journals, but the founder doesn’t have a daily handwritten morning-pages habit. Worth surfacing as a candidate — not as a Ray task (Ray can’t do handwriting) but as a founder-owned discipline. The vault’s job would be a weekly check on whether the practice is happening.
• “Two crappy pages per day” is the operative model for Sanity Check resurrection. Per the revival strategy doc, the bottleneck is consistency, not quality. Two pages — even rough — beat zero pages every time. Could become a check-in question on /morning-prep: “Did you write at least 2 pages yesterday?”
• Three-pass revision maps directly onto the existing draft-review skill. Currently /draft-review runs once. Splitting it into three passes — self-pass, fan-pass, critic-pass — would mirror Strauss’s discipline. Each pass uses a different evaluation lens. This is a concrete improvement to file as a follow-up.
• The “lawyer as proofreader” workaround is an interesting capability statement for AI. Claude (or Ray) is exactly the kind of trained-on-precise-language reader Ferriss describes. The /draft-review skill is already the Ferriss-lawyer pattern, just automated. Worth making explicit in the skill description.
• The 10% rule is a cleaner forcing function than current /draft-review prompts. Current draft-review surfaces all issues evenly. Forcing the model to say “this is the 10% you must keep, this is the 10% you must cut” creates a sharper signal for the founder. Add to /draft-review skill.
• “Highlight confusing” + “note mind-wandering” overlaps with the existing voice-match skill but extends it. Voice-match checks consistency; this adds clarity and engagement as separate axes. Could become the “reader-experience” pass that complements voice-match (which is the “writer-fidelity” pass).
• Anti-pattern alert. Ferriss recommends handwriting morning pages for the haptic feedback loop. The vault’s founder is already mostly digital. This may be a place where the founder’s personal preference diverges from the recommended discipline; flag it as an open question, don’t auto-prescribe.

Open follow-ups

• Refactor /draft-review into a three-pass pipeline. Pass 1 = self (founder’s own intent), Pass 2 = fan (Sanity Check’s existing audience), Pass 3 = critic (worst-faith reader). Each pass takes the prior pass’s output as input. Concrete skill update.
• Add the 10% rule to /draft-review output format. Force the model to explicitly call out “must keep” and “must cut” sections, not just general feedback. Sharper forcing function.
• Build a “reader-experience” pass parallel to /voice-match. voice-match is writer-fidelity; this would be reader-clarity. Use the Ferriss prompts: confusion-highlight, mind-wandering-note. Could be /clarity-check or roll into /draft-review pass 2.
• Surface morning-pages as a founder-owned discipline candidate. Not a Ray task. Check in once during /morning-prep this week to see if the founder has a writing-practice ritual, and whether morning pages would slot in.
• Cross-reference the writing cluster. The vault now has: this Ferriss video, Perell’s writing wisdom, CopyThat copywriting challenges, the writing-thoughts notes, and the Sanity Check revival strategy. Worth a ~/rdco-vault/05-concepts/writing-as-thinking.md concept page that ties them together.

Related

• ~/rdco-vault/06-reference/transcripts/2026-04-19-tim-ferriss-how-to-use-writing-to-sharpen-thinking-transcript.md — raw transcript
• ~/rdco-vault/06-reference/2026-04-06-writing-thoughts.md — Ray’s writing frameworks notes; Ferriss’s three-pass model would slot directly into this collection
• ~/rdco-vault/06-reference/2026-04-03-david-perell-writing-wisdom.md — Perell on simplicity and clarity; same axis as Ferriss’s “extraneous-word removal” via McPhee
• ~/rdco-vault/06-reference/2026-04-04-copythat-copywriting-challenges.md — CopyThat’s specific revision exercises; tactical companion to Ferriss’s three-pass discipline