How to Turn Supervisor Comments into a Revision Plan

Supervisor comments are not a to-do list yet. They are raw feedback: a mix of questions, objections, suggestions, and quick margin notes, often written in a hurry and sometimes pointing at problems the supervisor has not fully named. Turning them into a revision plan means converting that raw feedback into a small set of decisions, a sequenced set of tasks, and a clear definition of what "done" looks like for each one.

This guide walks through that conversion for a course paper, thesis chapter, or first manuscript. It uses a winter-break cleanup of a thesis chapter as a running example, gives you checkpoints along the way, and ends with a concrete next action you can take today.

Before you plan anything, sort the comments by what they ask of you. Most supervisor feedback falls into a few recognizable types, and each type needs a different response.

• Scope and framing comments question what the work is about, what it claims, or how it is positioned. These usually require a decision before any rewriting.
• Evidence and gap comments point at missing prior work, weak support, or unclear knowledge gaps. NIH guidance on scientific rigor treats careful assessment of prior knowledge, strengths, weaknesses, and knowledge gaps as a core part of rigorous practice, so these comments are substantive, not cosmetic.
• Method and transparency comments ask how something was done, validated, or reported. These often require you to check your own records rather than rewrite prose.
• Search and selection comments apply mainly to review-type work: which sources, which criteria, how records were screened.
• Presentation comments cover structure, clarity, figures, and language.

A quick diagnostic question for each comment: Does answering this require a decision, a search, a check of my own data, or only editing? Write the answer next to the comment. That single label tells you who needs to act and how long the task will take.

If a comment does not fit any category, mark it as unclear rather than guessing. Unclear comments are the ones you will bring back to your supervisor, and identifying them early prevents wasted rewriting.

A revision plan fails when it lives only in your head or in scattered margin notes. Create one register that holds every comment in a single place. A spreadsheet or table is enough.

Two habits make this register useful. First, copy comments verbatim; paraphrasing too early loses the supervisor's emphasis. Second, fill in the definition of done for every row before you start work. "Address methods comment" is not a definition of done. "Add a paragraph explaining how the sample was validated, with a citation to the method source" is.

This register is also your checkpoint artifact. When you next meet your supervisor, you can show which comments are resolved, which are in progress, and which you need to discuss.

The most common mistake is treating every comment as equally urgent. Sort the register into three tiers.

1. Plan-changing comments. These alter scope, argument, or method. They must be resolved before you rewrite affected sections, because rewriting first wastes effort.
2. Evidence-gathering comments. These require you to find, verify, or organize sources. They can run in parallel with writing but must finish before the affected section is declared done.
3. Editing comments. These are local fixes that can wait until the substance is settled.

Work top-down. Resolve plan-changing comments first, because they can invalidate editing work done earlier. If a scope comment and a presentation comment conflict, the scope decision wins.

For each plan-changing comment, write a one-sentence decision: What will the chapter now claim, include, or exclude? If you cannot write that sentence, the comment is still unclear and belongs on your escalation list.

Many supervisor comments amount to "where is the support for this?" or "you have missed work on this topic." For a review or literature-heavy chapter, answering these well means running a documented search rather than grabbing a few familiar papers.

The Cochrane Handbook's guidance on searching for studies is a useful model even outside systematic reviews: plan the search from explicit eligibility criteria, use multiple sources, combine free-text terms with subject headings, document the process, deduplicate records, and screen in stages. For a thesis chapter, you can scale this down while keeping the logic.

A practical sequence:

1. Write your eligibility criteria in one or two sentences. What counts as relevant prior work for this comment?
2. Draft search terms from those criteria, including synonyms and field-specific headings.
3. Run the search across more than one source.
4. Record what you searched, where, and when, so you can report it if asked.
5. Deduplicate and screen in stages: title and abstract first, then full text.
6. Link multiple reports of the same study so you do not double-count.
7. Rerun the search before you finalize, so recent work is not missed.

If your chapter reports a systematic review, PRISMA 2020 provides a reporting checklist, expanded explanations, an abstract checklist, and flow-diagram templates that cover transparent reporting of search, selection, and synthesis methods. Use it as a reporting scaffold for the methods and results of the search, not as a substitute for doing the search.

This is where a scholarly search tool earns its place in the workflow. When a comment requires finding academic papers, you can run the search through WisPaper's scholarly search interface and keep the terms and criteria you used in your register. For readers who want a public entry point to finding academic papers first, the WisPaper Scholar Search page explains that starting point. The point is not the tool; it is that every evidence comment ends with a documented search and a shortlist you can defend.

Once you have a shortlist, the revision plan needs a place where sources live alongside the comments they answer. Otherwise you will rediscover the same papers twice and lose track of which comment each source supports.

Save or upload the papers you intend to cite into My Library and organize them by comment ID or section. Then, when you need to check what a saved paper actually says before you cite it, you can ask questions based on papers in your library through Library QA. This keeps the evidence check close to the writing task instead of sending you back to a general web search.

A useful register addition at this stage: a source-to-comment map. For each comment that required evidence, list the sources you will cite and the sentence or section they support. If a source supports no comment and no claim, it probably does not belong in the revision.

With comments sorted and evidence gathered, you can sequence the actual rewriting. Order matters because some tasks depend on others.

1. Resolve scope decisions. Finalize what the chapter claims and covers.
2. Restructure if needed. Move, merge, or cut sections to match the new scope.
3. Insert evidence and method explanations. Write the paragraphs that answer evidence and method comments, using your source-to-comment map.
4. Revise the argument. Make sure the new material actually supports the claims.
5. Edit for clarity and consistency. Fix presentation comments last.
6. Update the register. Mark each comment resolved only when its definition of done is met.

Set a checkpoint after steps 1 and 2, before you write new prose. At that checkpoint, confirm that every plan-changing comment has a written decision. If any does not, stop and resolve it before continuing.

For longer or multi-part revisions, keeping the work in a structured project space helps. Research Projects is where you can create a new project and keep the revision's materials together. If the revision also involves reproducing a method from a paper your supervisor flagged, PaperClaw supports research-paper reproduction planning from an uploaded PDF and reproduction instructions.

Supervisor comments often include a version of "check your references." Before you close evidence-related comments, verify that every reference in the revised section is real and correctly recorded. TrueCite checks BibTeX against real academic databases to flag hallucinated references, which is a concrete way to screen your bibliography before submission.

Make this a checkpoint, not an afterthought: after inserting new citations, run the bibliography check, fix flagged entries, and only then mark the related comments resolved. A section with unverified references is not done, no matter how polished the prose.

Suppose it is late December. Your supervisor returned your literature review chapter with eleven comments before the winter break. You have two weeks and want to start January with a clean chapter and a plan for the next one.

Day 1 — Register and diagnose. You copy all eleven comments into the register, label each by type, and mark two as unclear. You write a definition of done for the nine clear ones. Checkpoint: nine rows have a definition of done; two are flagged for your supervisor.

Day 2 — Triage. Three comments are plan-changing: the supervisor wants the chapter's scope narrowed to one mechanism and wants the knowledge gap stated explicitly. You write a one-sentence decision for each. Checkpoint: all plan-changing comments have decisions.

Days 3–5 — Evidence. Four comments require sources. You write eligibility criteria, draft terms, search across more than one source, record the searches, deduplicate, and screen in stages. You save the shortlist in My Library and build the source-to-comment map. Checkpoint: every evidence comment has at least one mapped source.

Days 6–9 — Rewrite. You restructure to the narrowed scope, insert the evidence and a short methods paragraph, and revise the argument. You leave presentation comments for later. Checkpoint: the chapter's claims match the narrowed scope.

Days 10–11 — Citations and editing. You run the bibliography check, fix flagged entries, then address presentation comments. Checkpoint: no unresolved evidence or citation comments remain.

Day 12 — Register review. You update every status, write a short summary of what changed, and list the two unclear comments as questions for your supervisor. Checkpoint: the register shows only escalated items as open.

January — Next project. With the chapter closed, you use the same register structure for the next chapter and open a project space to hold its materials.

Use this before, during, and after any revision round.

• Every comment is copied verbatim into the register
• Every comment is typed (scope, evidence, method, search, presentation, unclear)
• Every clear comment has a definition of done
• Plan-changing comments have a written one-sentence decision
• Evidence comments have documented searches (criteria, terms, sources, dates)
• Sources are saved and mapped to the comments they answer
• The rewrite is sequenced: decisions, structure, evidence, argument, editing
• Citations are checked before sections are marked done
• The register status is updated after each work session
• Unclear comments are listed as questions for the supervisor

The revision is ready to hand back when three conditions hold: every comment in the register has a status, every resolved comment meets its definition of done, and every evidence comment has a documented search and mapped sources behind it. If you are reporting a systematic review, the PRISMA 2020 checklist gives you a reporting scaffold to confirm that search, selection, and synthesis methods are transparently described.

Stop and escalate rather than guess when:

• A comment is unclear after one honest attempt to interpret it. Add it to your questions list; do not rewrite on a guess.
• Two comments conflict. For example, one asks for more scope and another for less. Bring the conflict to your supervisor with both comments quoted.
• A plan-changing decision is outside your authority. If narrowing scope would remove a chapter your supervisor expects, confirm before cutting.
• An evidence comment cannot be answered with available sources. Report what you searched and what you found rather than padding with weak citations.
• A citation cannot be verified. Flag it and resolve it before submission rather than leaving it in place.

Escalation is not failure; it is the correct output of a register that is working. A short list of well-framed questions is more useful to your supervisor than a rewritten chapter built on guesses.

Open a blank table with the eight register fields above and copy in every comment from your most recent feedback, verbatim. Then label each one by type and write a definition of done for the clear ones. That single sitting gives you the skeleton of your revision plan, a list of questions for your supervisor, and a clear starting point for the next work session.