Skip to content

Process-based Writing

Principle · All principles
Evidence · 33 claims (17 for, 16 mixed) · 51 studies (16 causal, 10 quant-synthesis, 8 review, 6 theoretical, 5 associational, 4 qualitative, 2 design), q1–q4 · 10 of 51 report an effect size · 23 claims rest on one study

Conditional relationship

This page owns writing taught as a staged, revisable process: planning, drafting, getting a response, revising and editing a text for a stated reader and purpose, rather than writing it once and having it marked. For a developing writer (a novice or intermediate writer in the genre at hand) whose texts show little planning and revisions that change only surface errors, teaching the process, with feedback on drafts and time to act on it before the final version, is expected to produce higher-quality compositions (skill-performance, judged on a writing-quality measure) than instruction that focuses on grammar and the finished product. The goal is a complex-skill with high element interactivity: the writer has to juggle ideas, structure, language and transcription at once.

One claim tests the relationship directly, and it bounds it. A meta-analysis of 29 experimental and quasi-experimental studies in grades 1–12 found that process writing instruction gave general-education students a statistically significant but modest gain in writing quality (weighted ES = 0.34), no significant gain for struggling writers, and no gain in motivation to write (Process Writing Improves Writing Quality [+M]). A second meta-analysis, of adolescent writing instruction, found that explicitly teaching strategies for planning, revising and editing, with modelling, at least three days of instruction and a move toward independent use, had a large effect on writing quality (0.82), larger again when self-regulation procedures were added (1.14) and for struggling writers (1.02), and these effects held against several control conditions, process writing among them (Strategy Instruction Improves Writing Quality [+M]). So the working rule is: stage the process, but teach the moves inside each stage explicitly; do not rely on drafting and feedback cycles alone, least of all for writers who struggle.

What no claim tests: how many drafts to require, the stance a teacher takes when responding (as an interested reader, without taking over the writer's purpose, with the writer negotiating the response), and process instruction for adults or in higher education (the strategy meta-analysis had no students above 10th grade; the process meta-analysis stopped at grade 12). Those recommendations, absorbed from three earlier pages on responding to student writing, are kept below as labelled proposals.

How this sits beside converted pages: Guided Practice holds the practice phase after a model, and here it is the supported first use of a newly taught planning or revising strategy. Immediate Feedback holds feedback timing, and its rule (presence and content matter more than timing; leave a chance to use the correction) applies to draft feedback. Self-Regulated Learning and Goal Setting and Monitoring hold the general plan–monitor–adjust cycle that writing strategy instruction applies to composing. The collaborative evaluation pattern holds judging work against criteria with others before revising one's own, and structured peer review holds the peer-review routine; this page uses both for the response stage and does not repeat them. The Writing Workshop pattern is one reusable configuration of this principle.

Default design

A claim supports the general relationship (process plus explicit strategy teaching, with feedback before the final version). The particular sequence, quantities and response stance below are the earlier pages' guidance made concrete; each step says whether it rests on a claim or is an untested proposal.

  1. Fix the reader, purpose and criteria before anyone writes. Name who will read the text, what it must do for them, and three to five criteria for quality, shown with one or two example texts at different levels. An explicit statement of the task and the standards for success is reported as motivating (Explicit statement of tasks and standards [+W], an argument citing unnamed experiments); the earlier page's "process without quality criteria" caution. Otherwise untested.
  2. Teach one planning strategy and one revising strategy explicitly, by modelling them aloud. Think aloud while planning a text (generate ideas, select, order, outline) and while revising one (reread as the reader, check against the criteria, change content and structure before sentences). Then give supported practice with the strategy and a prompt card, and fade the card. Evidence: Strategy Instruction Improves Writing Quality [+M] (grades 4–10; modelling, at least three days, progression to independent use). See modeling writing and revising.
  3. Give the writer process goals first and an outcome goal later. While a strategy is new, set goals about the moves ("check each paragraph against criterion 2 and rewrite one that fails"), not the final score; once the moves are reliable, add a quality target. Evidence: Process goals lead to better skill acquisition for novices than outcome goals [+M] (its experiment is high-school students learning writing revision; it also suggests the shift to outcome goals after mastery). Add a self-recording step (the writer ticks which moves they used, and where), which is part of the self-regulated version of strategy teaching in the same meta-analysis.
  4. Draft without stopping to correct. Ask for a full first draft in one sitting or two, with surface errors left for later; the earlier page's separation of generation from editing. The draft-then-revise order is untested as instruction. One experiment bears on what drafting does: university writers given only an overall goal (no outline) who produced sentences spontaneously gained in self-rated understanding of the topic but wrote a weaker first text, while the same spontaneous production under detailed outline planning went with a fall in understanding; revising the text's global structure predicted greater understanding in either condition (spontaneous sentence production under minimal planning [~M], one study, 30-minute essays, regression within conditions). So a loose first draft can serve the writer's thinking at the cost of first-draft quality, which the revision stage then has to recover.
  5. Respond to the draft before it is final, as a reader first. Give the first response on meaning and structure: what you understood, what you wanted to know more about, where you lost the thread, with an affective remark where it is honest (surprise, interest). Limit comments to the two or three changes that matter most for the reader. Direct comments at the text and the writer's process, not at the writer as a person, and avoid generic praise. Evidence for task- and process-level comments over person-level ones: Feedback Most Effective At Task And Process Levels [+M] (feedback in general, not writing). Responding as an interested reader is from Shvidko (2020), an essay from practice; untested.
  6. Do not take the text over. Before sending comments, check each one against the writer's purpose, point of view and intended meaning (Goldstein's three features, as Shvidko reports them): a comment that would change any of them is phrased as a question or an option, not an instruction. Corrections of fact, convention and clarity remain the teacher's job. Untested proposal, from the earlier "avoid appropriating student writing" page.
  7. Make revision a negotiation. Ask the writer to answer the response in writing before revising: what they will change, what they will not and why, and what they want help with (a letter-to-the-reviewer memo), or hold a short conference. In a conference, raise the level of help only when the writer cannot make the change at the current level, and lower it when they can: one-to-one tutoring with support contingent on the learner's last response beat fixed or no support, though in a dynamic-assessment meta-analysis explicit strategy training beat contingent scaffolding (contingent scaffolding [~M]; long division and science tutoring, not writing; classroom scale not shown), which is why the strategy is taught first (step 2) and the conference adjusts help with it. Writers may challenge a comment if they give a reason. Untested proposal, from the earlier "involve students in revision" page. Asking writers to explain their revision choices is a form of self-explanation, which improves understanding on immediate tests in other domains (self-explanation [+S], carried to writing).
  8. Use peers as readers once they have criteria and practice. After writers have used the criteria on example texts, add a peer-review round (two or three peers each, with the criteria and a comment frame) before the teacher's response. Evidence: Peer Feedback Improves Writing [+M] (higher education: large gain against no feedback, smaller against self-assessment, not reliably different from teacher feedback); Peer assessment improves academic performance by a small average amount [+M]. Peers flag problems no less accurately when they write less well, but their comments are rated less helpful (peer reviewer accuracy and helpfulness [~M]), so the teacher reads a sample of peer comments. The routine is in structured peer review.
  9. Edit last, against a short list. After content revision, have writers find and fix two or three error types named in advance, chosen from their own drafts. Second-hand evidence: asking students to notice particular errors decreased those errors in the next draft without lowering overall quality (noticing errors [+W], reported in a review).
  10. Close each piece with a short reflection. The writer writes three or four sentences: what changed between drafts, why, and what they will do first next time. Evidence that monitoring performance against a criterion supports regulation is theoretical (Self-monitoring improves self-regulation [+M]); the routine is untested.
  11. Keep the whole text in view. Each stage ends by rereading the whole text for the reader. Not every task needs the full cycle: short pieces can go plan–draft–one revision. Untested; from the earlier page's caution against over-fragmentation.

Dose and progression across sessions (untested proposals, except where marked). A unit of three to four weeks, two or three compositions, each taken through the full cycle, with roughly: planning 1 session (20–40 minutes), drafting 1–2 sessions, response and revision 1–2 sessions, editing and reflection 20–30 minutes. The strategy studies in the meta-analysis taught each strategy over at least three days (strategy instruction [+M]); the rest of these quantities are proposals.

  • Early (first composition): the teacher models each strategy aloud; the class plans one text together; writers use the prompt cards; the teacher gives the main response; goals are process goals.
  • Middle (second composition): writers plan alone with the card available; peers give the first response with the criteria; the teacher responds to the revised draft; writers write a reviewer memo.
  • Late (third composition): no prompt card unless asked for; writers set their own process goal and a quality target; writers review their own draft against the criteria before any peer or teacher response; the teacher responds only to what the writer asks about plus one priority.
  • When to move on: a writer moves from the supported to the independent version of a strategy when, on two consecutive pieces, their revisions change content or structure (not only spelling and grammar) and their self-check against the criteria roughly agrees with the teacher's reading. A writer whose revisions stay at the surface after two pieces goes back to modelled revising on an example text before the next piece.

When a condition is not met: if there is time for only one response per piece, give it on the draft, not the final version; if no one can respond to drafts, have writers review their own drafts against the criteria and an example text, then revise.

Fitting the design to a situation

The three facts that most change the decision:

  • How the writer currently plans and revises. A writer who already plans and makes global revisions needs a lighter cycle and their own choice of focus; one who writes once and corrects spelling needs explicit strategy teaching, not only more drafts. If unstated, ask for one draft and its revision, written under ordinary conditions, and look at what changed between them.
  • Who can read drafts, and how often. Teacher response is the scarce resource; peer and self-review stretch it only with criteria and practice. If unstated, ask: who will read each draft before the final version, how long after it is written, and how many drafts per writer per week can they read?
  • The text, reader and horizon that count. A timed exam essay, a report for a real client and a portfolio piece need different cycles. If unstated, ask: what must the writer produce unaided, for whom, under what conditions, and when?
If the brief says… Then change… Basis
Writers are struggling, have a learning disability, or cannot yet write fluently by hand or keyboard Teach planning and revising strategies explicitly with self-regulation steps (goal setting, self-recording, self-instruction) over several weeks; do not rely on drafting and feedback cycles alone; reduce transcription load (dictation, scribe, keyboard) while strategies are learned Process Writing Improves Writing Quality [~M] (no significant gain for struggling writers); Strategy Instruction Improves Writing Quality [+M] (struggling writers 1.02); transcription supports untested
Writers are advanced in the genre (they already plan and revise globally) Drop the scripted strategy and prompt cards; let the writer set the revision focus; keep one outside reader's response per piece Instructional guidance that helps novices can become redundant or counterproductive as expertise grows. [~M] (cognitive-load review, not writing); untested for writing
Learners are child in the early primary grades Shorter texts; plan orally or with drawings; co-write with a partner in structured pairs (the more able writer helping the less able); revise one thing per piece Collaborative Writing Improves Text Quality [+M] (grades 4–12, and one randomised study of 10–11-year-olds that found larger gains in later individual writing); younger children untested
Learners are adult, including in work settings (reports, case notes, proposals) Use the writer's own work documents as the texts; take the reader from the real setting (the manager, the client, the colleague); keep the strategy modelling but shorten it; respond as that reader first untested; neither meta-analysis covers adults (strategy studies stop at grade 10)
Writers are writing in a second language, or have low literacy Respond to meaning and structure first, in plain language; correct language separately, on two or three error types the writer can recognise, and ask the writer to find further instances; allow planning in the first language Review: language proficiency and feedback literacy moderate uptake of written corrective feedback [~W] (second-hand); noticing errors [+W] (second-hand, EFL college writing)
The goal includes motivation to write or confidence as a writer Do not expect process instruction alone to raise it: add a real reader and purpose, explicit task and standards, and reader-response comments; measure motivation separately Process Writing Improves Writing Quality [~M] (no gain in motivation); Explicit statement of tasks and standards [+W] (argument only)
Writers have low confidence, or have had harsh feedback before Open every response as a reader (what you understood and found interesting); direct criticism at the text and the next move, never at the writer; give a specific positive comment tied to a criterion, not generic praise Feedback Most Effective At Task And Process Levels [+M]; Praise for intelligence, rather than for effort, after success undermines children's motivation and performance after a later failure [~M] (person praise; children, not writing)
One teacher with a large class or many sections Peers give the first response with criteria; the teacher reads a sample of peer comments and responds to each writer on two or three priorities only; rotate 5-minute conferences so each writer gets one per composition; do not grade drafts Peer Feedback Improves Writing [+M] (peer vs teacher feedback not reliably different); Students report teacher feedback on writing is insufficient [~W] (interviews; teachers' load)
Peers are novices as reviewers Train them on example texts with the criteria first; ask them to point to the place in the text and say what a reader would experience; teacher checks a sample peer reviewer accuracy and helpfulness [~M] (secondary writers, correlational); rubrics and peer reviewers [~M] (online, two quasi-experiments; one found less reflective comments)
online-self-paced or asynchronous course, no teacher reading every draft Build in a rubric-guided peer review with practice on rated examples, a reviewer memo with each revision, and a self-check against the criteria before submission rubrics and peer reviewers [~M] (one MOOC writing course, one online graduate course); Peer assessment improves academic performance by a small average amount [+M] (online and offline did not differ)
Mixed modes: some writers remote, some in the room, or some on phones only Put drafts in one shared document with comment threads so remote and in-room writers review the same way; pair a remote writer with an in-room one for peer review; for phone-only writers keep pieces short, accept voice comments and spoken memos untested proposal
Only one session is available Plan for 10 minutes against a model text, draft, then one revision against a single criterion with a partner's response; collect the draft and the revision; do not claim a gain in writing ability untested; the strategy studies taught over at least three days (strategy instruction [+M])
The text goes to a real audience (publication, a client, a public performance), or the audience should see only the final text Keep the writer as the owner of the decisions: teacher and peers comment as readers and pose options; the writer's memo records what they accepted; add a final edit for conventions before release; keep drafts and comments out of the published version untested proposal, from the earlier pages on appropriation and responding as a reader (Shvidko 2020, an essay from practice)
The goal is writing to learn: developing the writer's own understanding of a topic, not a polished text Plan with one overall goal rather than a detailed outline; let the writer draft spontaneously; then ask for a revision of the global structure (reorder, cut, add sections) and a short note on what they now think; judge the first draft on ideas, not quality spontaneous sentence production under minimal planning [~M] (one experiment, university writers, self-rated understanding; associations within assigned conditions)
Assessment is a timed, unaided essay Practise planning and revising under time limits (for example 5 minutes planning, 5 minutes rereading in a 40-minute task); keep the multi-draft cycle for learning the strategies, and the timed form for applying them untested proposal
Drafts are graded, or writers care only about the score Withhold the grade until the final version; require the reviewer memo before revision; give credit for the revision itself Students report teacher feedback on writing is insufficient, and they value scores over feedback [~W] (one interview study); grade withholding untested

Twelve of the sixteen rows cite a claim. Only the first rests on claims that test writing instruction for the learners named (the third rests on one for older children only); the others carry a neighbouring finding (feedback in general, a cognitive-load review, peer review in higher education, second-hand reports from reviews) to this setting, and say so. Four rows are untested proposals.

Observation, state and explanation

What can be observed is a text and its history: the plan (if any), each draft, the comments it received, what the writer changed between drafts (content, structure, sentences, surface), what they said about why, and the time spent. "Lacks revision skill", "is dependent on the teacher" and "has lost ownership of the text" are inferred states. A better final draft shows what the writer could do with this response; it does not show what they can do on the next piece alone.

Response Plausible explanations Observation that could change the interpretation
Revisions change only spelling and grammar Does not know how to revise globally; did not understand the comments; believes revision means correction; no time was left Give an example text and ask for one change to content or order with a reason; check whether the comments were read (ask the writer to restate one).
Accepts every comment, including ones that change the point of the text Treats the teacher as the owner of the text; does not have a purpose of their own; fears the grade Ask the writer to state the text's purpose before and after revision; offer a comment that conflicts with that purpose and see whether they challenge it.
Ignores most comments Comments too many or unclear; the writer disagrees; the grade is already fixed; low feedback literacy in a second language Ask for the reviewer memo: which comments they used, which not, and why; compare uptake of two or three focused comments with uptake of a full marking.
Strong plan, weak draft Transcription or language load; the plan was copied from a model; time ran out Have the writer dictate a section from the plan; compare with a timed and an untimed draft.
Peer comments praise everything Reviewers lack criteria or practice; social pressure; reviewers cannot see the problems Give the reviewer a weak example text with a known problem; compare comments made anonymously and with names.
Good revisions with prompts, none without The strategy is not yet internalised; the writer does not believe it is worth the effort Ask for a self-check against the criteria without the card; ask what the strategy is for.

These rows are proposals, untested with learners as diagnostics. A probe (a conference, a memo) is also a response that changes the next draft, so record each one, and keep the explanations that remain open.

Evidence and qualifications

  • Process Writing Improves Writing Quality [+M] / [~M]: Graham and Sandmel (2011), a meta-analysis of 29 experimental and quasi-experimental studies in grades 1–12 (synthesis-mixed). For general-education students, process instruction produced a modest gain in writing quality (weighted ES = 0.34), unrelated to grade, measure reliability, professional development, genre or study quality; it did not significantly raise motivation to write or the quality of struggling writers' compositions. This is the direct test of the page's relationship, and its own limit. Horizon not stated on the claim page.
  • Strategy Instruction Improves Writing Quality [+M]: Graham and Perin (2007), meta-analysis of adolescent writing instruction; 20 effect sizes for explicit teaching of planning, revising and/or editing strategies (grades 4–10), all with modelling, at least three days of instruction and progression to independent use. Weighted mean 0.82, every effect positive; SRSD 1.14 against 0.62 for other strategy instruction; struggling writers 1.02 against 0.70 for the full range. Effects held against process writing and other controls. The same study and numbers sit on Self-regulated learning strategy instruction improves writing outcomes [+M], a duplicate claim. No students above grade 10.
  • Process goals lead to better skill acquisition for novices than outcome goals [+M]: an experiment with high-school students learning writing revision (Zimmerman and Kitsantas 1999): process goals outperformed outcome goals, with a suggestion that shifting to outcome goals after mastery helps; plus a theoretical chapter. No effect size or sample size on the claim page.
  • Peer Feedback Improves Writing [+M]: Huisman et al. (2019), 24 higher-education studies: giving and/or receiving peer feedback improved writing against no feedback (g = 0.91) and self-assessment (g = 0.33); the comparison with teacher feedback (g = 0.46) had a confidence interval crossing zero. The nature of the feedback moderated the effect. Supports a peer round; does not show peers can replace the teacher.
  • Collaborative Writing Improves Text Quality [+M]: seven comparisons (grades 4–12) of students planning, drafting and/or revising together against writing alone (ES = 0.75), and one randomised study of 28 children aged 10–11 in which structured pairs gained more in later individual writing. Small pool.
  • Skilled EFL writers' planning and revising strategies [+W]: interviews with nine skilled EFL university writers (You and Joe 2001), all of whom planned before writing (generating, selecting and ordering ideas, outlining) and spent most of their revision time on logic and coherence for the reader rather than surface errors. It describes the target state the planning and revising strategies aim at (steps 2 and 9, and the move-on criterion); it tests no instruction.
  • Spontaneous sentence production under minimal planning [~M]: Baaijen and Galbraith (2018), 78 university students randomly assigned to outline or goal-only planning before a 30-minute argumentative essay, with keystroke logging. Within goal-only planning, spontaneous sentence production predicted a rise in self-rated understanding and lower text quality; within outline planning it predicted a fall in understanding; global structural revision predicted greater understanding in both. Text quality and understanding were not simply positively related. The planning condition was assigned; the process measures were not, so the within-condition findings are associations, and understanding is a single self-rating. It qualifies this page's working assumption that better planning always serves the writer: what counts as a better process depends on whether the goal is a better text or a better grasp of the topic.
  • contingent scaffolding [~M]: one-to-one tutoring (long division, the circulatory system), a review and a dynamic-assessment meta-analysis in which explicit strategy training was more effective than contingent scaffolding, which was more effective than coaching. It agrees with putting explicit strategy teaching first and supports adjusting help in conferences; none of it is writing, and classroom-scale effects are not shown.
  • Limits and qualifications: peer reviewer accuracy and helpfulness [~M] (two correlational secondary-school studies: the reviewer's own writing performance predicted how helpful comments were rated, not whether real problems were found; with a well-specified rubric, peer ratings were as reliable in lower- as in higher-performing schools); rubrics and peer reviewers [~M] (two online quasi-experiments: a rubric changed what reviewers commented on and helped them rate complex criteria, but one found less reflective comment); Feedback Most Effective At Task And Process Levels [+M] (two meta-analyses of feedback in general: high-information feedback far above reinforcement or punishment; effectiveness declined as feedback drew attention toward the self); Praise for intelligence, rather than for effort, after success undermines children's motivation and performance after a later failure [~M] (praise for intelligence in children; the harm is specific to person-focused praise).

Learners, genres and measures differ across these claims (school writers on researcher-rated essays, university students, online courses, self-rated understanding), so do not rank them by effect size. None tests the reader-response stance, the rule against appropriation, or a number of drafts, and none measures writing ability after a delay.

The evidence by question

Further evidence, not yet read against this model

Claims this page cited before it was rewritten, and claims found for it that bear on it more distantly. Claims read against the model (2026-10-15) now sit in the steps, rows and evidence above; what remains here bears only by distant analogy.

Objective and learner-valued goal

The designer's objective is usually a writing capability (skill-performance): unaided texts of a given genre, for a given reader, judged on stated criteria, at a stated time. Elicit separately what the writer wants: to get the grade, to say something that matters to them, to pass a workplace or exam requirement, to stop being corrected. These can pull apart. A writer who values the grade will accept every comment, which looks like good uptake and may be the appropriation the earlier pages warned about; a writer who values their own message may resist comments that would improve the grade. Both are observations, recorded beside the text.

For example, an adult in a vocational course may want to write incident reports their supervisor accepts first time, while the designer's objective is a five-paragraph argumentative essay. Teaching the planning and revising strategies on the essay serves the objective; one cycle on a real incident report, with the supervisor's standards as the criteria, tests the writer's goal. Record both, so that a better essay is not read as the capability the writer came for.

What would revise this model?

The expectation should weaken if well-controlled studies show that adding drafting and feedback cycles to explicit strategy teaching adds nothing to writing quality, or that process instruction's modest gain disappears on texts written alone some weeks later. If studies of adults or university writers find no benefit of explicit strategy teaching, the struggling-writer and adult rows change. If teacher response as a reader, or a rule against appropriation, turns out to change revision quality or writers' persistence when tested against ordinary evaluative marking, those steps move from proposals to evidence; if it changes nothing measured, they remain stances. If peer review with criteria matches teacher response on writing quality, the teacher's time can move almost entirely to strategy modelling. Do not protect the model by calling every null result "the process was not taught properly" after the fact.

Better drafts with help, better final versions, better unaided writing on the next piece, delayed retention, transfer to other genres and the writer's own valued use are separate claims. The present evidence establishes no number of drafts, no number of comments and no criterion for moving a writer from supported to independent use.

Claims

Examples

Key Sources

  • Hayes, J. R., & Flower, L. S. (1980). Identifying the organization of writing processes. In Cognitive processes in writing (pp. 3-30). Lawrence Erlbaum.
  • Wei, Z., Shang, H., & Briody, P. (2012). The relationship between English writing ability levels and EFL learners' metacognitive behaviour in the writing process. International Journal of Academic Research in Progressive Education and Development, 1(4), 154-180. doi:10.6007/ijarped/v1-i4/12012
  • Fernandez, R., Peyton, J. K., & Schaetzel, K. (2017). A survey of writing instruction in adult ESL programs: Are teaching practices meeting adult learner needs? COABE Journal, 6(2), 5.
  • Shvidko, Elena. (2020). Taking into Account Interpersonal Aspects of Teacher Feedback: Principles of Responding to Student Writing. Journal on Empowering Teaching Excellence, Vol. 4, Iss. 2, Article 7. https://doi.org/10.26077/936a-72f7
  • Baaijen, V. M., & Galbraith, D. (2018). Discovery Through Writing: Relationships with Writing Processes and Text Quality. Cognition and Instruction, 36(3), 199-223. https://doi.org/10.1080/07370008.2018.1456431
  • Junhong Ren. (2017). College English Writing Instruction for Non-English Majors in Mainland China: The “Output-Driven, Input-Enabled” Hypothesis Perspective. English Language Teaching, 10(7). https://doi.org/10.5539/elt.v10n7p150
Page metadata
Field Value
type principle
id process-based-writing
aliases involve-students-in-revision-process, avoid-appropriating-student-writing, respond-as-a-reader, teach-writing-based-on-reading
title Process-based Writing
description Teaching writing as a staged process (plan, draft, respond, revise, edit) with feedback on drafts gives developing writers a modest gain in writing quality over product-focused teaching, tested by one meta-analysis in grades 1–12 that found no gain for struggling writers, who gain more from explicit strategy instruction.
status review
moved_from pattern/synthetic-planning-for-discovery-oriented-writing
canonical True
generated by claude/unspecified on 2026-10-15
trust unverified — no one has confirmed this page's content
sources 3 — shvidko-2020, baaijen-galbraith-2018, junhong-ren-2017

Frontmatter is the page's machine-readable record — see the Schema & Guide.