A reflective writing builder is scaffolding for a personal assignment: it splits your reflection into the Gibbs cycle's six stages, gives each stage question prompts and a word budget, and flags what a stage might be missing. It never writes a word for you — a reflection is an account of your own experience, and the tool cannot have any.
A reflection is a first-person account of something that happened to you. This page can structure it, time it and flag the thin stages, but it cannot supply the event or the feelings, and any tool that claims otherwise is offering you a fabricated experience — which is the one thing a reflection cannot be.
That matters for marks, not just ethics. Rubrics award the specific detail only the writer has: the exact moment the plan split from reality, the feeling that surprised you, the reading that reframed it afterwards. Invented entries collapse at the first follow-up question in a seminar or a viva.
The demo below is invented and labelled as illustrative — use it to see how the tool behaves, never as text to submit. And if your course restricts AI assistance, prompts plus your own sentences is the compliant way to use this page.
Build the reflection stage by stage
Pick the model your brief names, set the total length, and work the stages in order. Each stage card carries four things: question prompts you can insert as lead-ins, a word budget shown as a live meter, gentle strength flags that react as you type, and — on the Feelings and So What? cards — a bank of twenty-four emotion words you can insert with a click. Nothing is generated: a prompt inserts a question, and answering it is your job. Your draft saves in this browser only (key csw-rwb-v1), never uploaded, and Reset wipes it along with the stage texts.
Reflective writing builder
100% private — runs in your browser0 of 6 stages filled — start with Description.
Strength flags react as you type in each stage card — gentle nudges, never a score. Each model keeps its own text for this session; the browser saves the one you are looking at.
Description
—Feelings
—Inserting adds the word, not the experience — keep only the feelings that were real.
Evaluation
—Analysis
—Conclusion
—Action Plan
—What?
—So What?
—Inserting adds the word, not the experience — keep only the feelings that were real.
Now What?
—The demo is an invented, illustrative scenario (a nursing student's first-placement handover) — never submit it. Your draft saves in this browser only, key csw-rwb-v1; nothing leaves your device.
What the demo shows
The demo reflection is an invented, illustrative scenario: a first-year nursing student asked to give a ward handover on day five of a first placement, who misses a charted medication change and works through the aftermath. It fills all six Gibbs stages to roughly seventy per cent of budget — deliberately under, because a first draft that lands at exactly one hundred per cent is usually padded. Notice what each stage does: Description keeps to the scene and the sequence, Feelings names four real reactions without analysing them, Evaluation separates what worked from what did not, Analysis carries the reasoning, and the Action Plan runs on verbs and dates. If you need the clinical-documentary side of placement paperwork, the nursing care plan template is the sibling page.
Copy, print and reset
Copy full draft builds plain text: a header line with the model and totals, then each stage as a numbered heading with its word count beside the text. Paste it into your working document — not into the submission portal directly, since your department's template, font and reference style still apply. Print view fills a print-only sheet and opens your browser's print dialogue. Reset wipes every stage immediately, no confirmation — the draft lives in your document, not here — and restores the default budgets.
The six stages of the Gibbs cycle: what each one asks
The Gibbs cycle (Gibbs, 1988) walks through one experience in six moves: what happened, how you felt, what was good and bad, why it happened, what you learned, and what you will do. The order matters, because each stage feeds the next — a feelings stage written before the description is invented, and an action plan with no analysis behind it is a guess. What each stage asks of you, and what a grader looks for:
| Stage | What it asks of you | What a grader looks for |
|---|---|---|
| Description | Keep to the facts: who was there, when and where it happened, what you did, what was said. | A scene a reader could re-enter — people, time, place — told without evaluation sneaking in. |
| Feelings | Name your reactions at the time, honestly, before you judge them. | Two or three genuine emotions, including the uncomfortable ones — not a performance of appropriate feeling. |
| Evaluation | Split the experience: what went well, what did not, and how you know. | Judgement with evidence attached — a comment, an outcome, a missed cue — not a verdict with no support. |
| Analysis | Make sense of it: why did it go that way? Bring in module concepts or literature here. | Connectives doing real work (because, however, therefore) and at least one link to a reading or principle. |
| Conclusion | Step back: what does it add up to, and what could you have done differently? | Learning stated plainly and sized honestly — neither a slogan nor self-flagellation. |
| Action Plan | Commit to specific actions with timeframes. | Action verbs, a named deadline, and a way to tell the plan worked. |
The two heaviest stages are Description and Analysis — hence the default 25/25 split — but drafts routinely starve the middle. Feelings gets a shrug ("I felt nervous"), Evaluation becomes a list with no evidence, and the essay loses the band descriptors that reward honest processing. A working rule: if a stage's text could be pasted under a different incident with no changes, it is not specific enough yet. For the essay-shaped version of this model — introduction, thesis, references — the how to write a reflective essay guide walks the same cycle at document level, and the reflective essay examples page shows finished drafts to compare your proportions against.
What? So What? Now What?: the three-stage alternative
What? So What? Now What? is the short form. Often traced to Borton (1970) and popularised in nursing education through Driscoll's adaptation, it compresses the same logic into three moves: What? describes the event, So What? interprets it — feelings, evaluation and analysis bundled together — and Now What? commits to change. Choose it when the word count is small (300–600 words), when the incident is simple, or when your brief names Driscoll or Borton directly. At 1,000 words or more, the six-stage cycle usually earns its keep, because Feelings, Evaluation and Analysis each get their own space instead of sharing one paragraph.
The builder runs both models on the same engine: switch tabs and the prompts, budgets and flags re-map to the new stages. So What? carries the heaviest budget (forty per cent) because it does the work of three Gibbs stages at once, and it inherits both checks that matter — name feelings, and reason with connectives. One caution: the short model is not a licence to skip the uncomfortable middle. A What? So What? Now What? reflection that spends forty per cent on narrative and two sentences on meaning has just moved the summary-only problem into a smaller container.
How the stage prompts work (and why they are questions)
Every prompt on this page is a question, and that is deliberate: reflection answers questions, it does not fill templates. Click a prompt and it inserts at your cursor as a lead-in — Why did it happen that way — what sense can you make of it? — with a space after it, ready for your sentence. Three ways to use them. Answer one directly and delete the question. Keep two or three as paragraph openers while you draft, and strip them in the edit. Or ignore them entirely and write freely, using the prompts afterwards as a checklist of what the stage usually needs.
None of this writes your content — a question is scaffolding, not material. If an inserted prompt survives into your final draft, a grader will see the question sitting in your reflection, so the edit pass is not optional. The same applies to the emotion-word bank: inserting "overwhelmed" adds a word, not an experience. If the word does not match what happened, delete it — a Feelings stage with two honest emotions beats one with six borrowed ones.
Word budgets that add up to your limit
The length select distributes your total across the stages automatically: 25/10/20/25/10/10 for Gibbs (Description, Feelings, Evaluation, Analysis, Conclusion, Action Plan) and 30/40/30 for What? So What? Now What?. At 1,000 words that lands at 250/100/200/250/100/100 — numbers you can quote in your plan. Change the total and every budget recalculates immediately; the rounding is handled so the parts always add up to the whole. The meters are guides, not rules: a stage at eighty or ninety per cent of budget is healthy, and the amber tint just marks over-budget text so you can see where trimming is cheapest.
Two honest notes. First, briefs override builders — if your assignment sheet assigns its own proportions, follow it and treat these numbers as defaults for when it does not. Second, budgets describe finished drafts; mid-draft, an under-filled Analysis is a to-do, not a failure. The whole-draft meter above the stages sums every stage against your target, and the stages-filled count shows whether the shape is emerging. Four stages strong and two empty is a normal evening's progress; six stages at ten per cent each is a sign the incident you picked is too thin — pick a bigger incident rather than inflating six small ones.
Stage-strength checks: gentle flags, not scores
Four checks run as you type, and they are flags, not scores. There is no total, no percentage, no pass state — a reflection that clears every flag can still be dull, and one that trips a nudge can still be excellent. What each check looks for:
| Check (stage) | What it reads for | Example nudge |
|---|---|---|
| Coverage (Description, What?) | Who is named, when is pinned, where is set, and enough substance (about twelve words) for the other flags to read. | "When: pin the moment — a shift, a day, a time." |
| Emotion words (Feelings, So What?) | Two or more words from the twenty-four-word bank, counted as distinct feelings. | "Name at least two feelings — the word bank below inserts them." |
| Reasoning link (Analysis, So What?) | A connective that carries reasoning: because, therefore, however, which suggests. | "Add a reasoning link — because, therefore, however, which suggests." |
| Action plus timeframe (Action Plan, Now What?) | Action verbs and a deadline — the difference between "I will try to communicate better" and "I will ask my preceptor to observe one handover before Thursday." | "Give the plan a deadline — a shift, a day, a week." |
The checks run on both models — What? inherits the coverage check, So What? takes the feelings and connective checks, Now What? takes the action check. Each flag is a heuristic with visible seams: the word lists are ordinary ones, so "afternoon" counts as a time marker whether or not it pinpoints the moment, and a scene told entirely through dialogue can hide its people from the who-check. A nudge can be wrong about your sentence; your knowledge of the event outranks it. Treat every flag as a question to answer — either fix the stage or decide the flag does not apply, and move on. For prose-level habits — stock phrases, uniform rhythm, hedge stacking — the AI writing tells checker is the complementary lens: this page checks the skeleton, that one checks the skin.
Two failure modes this builder is built to catch
Summary-only description
The draft retells the event in chronological order until the word count runs out. Every sentence is "then": then the handover started, then I spoke, then the staff nurse checked the chart. Description reads fine, but nothing after it exists — Feelings is one adjective, Analysis is one "it was a learning experience", and the marker's comment writes itself: descriptive rather than reflective. The tell-tale shape is a Description stage near full budget with every other stage starving, which is exactly what the per-stage meters expose. The fix is not writing less description but reallocating attention: people, time, place and sequence in a quarter of the draft, and the meaning-making in the half that follows. The default budgets encode that shape — if your draft fights them, and the brief has not set its own split, the draft is probably describing too much.
Pasted theory with no link to the incident
The opposite failure: a properly cited paragraph of theory — a cycle, a communication framework, a model of reflection — sitting next to the incident with no connective tissue. "According to Gibbs (1988), reflection proceeds in six stages. This incident showed the importance of communication." The citation is real; the thinking is missing, because nothing in the paragraph could not be swapped into another student's essay. The Analysis check nudges here: a stage with no because, therefore or however usually has description in an analysis costume. The fix is mechanical before it is inspirational — take one element of the theory and one specific moment of the event, and write the sentence that joins them: "the framework puts medication changes at a fixed point in the sequence, which is exactly the step I skipped." One honest sentence of linkage is what the rubric's top band usually names.
What this builder cannot do
The honest limits, because a scaffolding tool that hides its edges is worse than none:
- It cannot supply the experience. Every sentence in the export comes from your keyboard. The tool has no event to describe and no feelings to name, and the demo is invented.
- It cannot judge truthfulness. The flags read text, not events. A fabricated reflection can clear every check — that is a limit of all scaffolding, and the reason the honesty box sits above the tool.
- The checks are heuristics. Word lists misfire. They nudge, they do not grade, and your judgement outranks every flag.
- Budgets are defaults. Your brief, your module handbook and your supervisor's comments override every number on this page.
- It is not a ghostwriter. If your course restricts AI assistance, the compliant use is prompts plus your own sentences. A reflection generated from a summary of your week is not a reflection, and it usually reads like one.
- It is one lens. It checks the skeleton — stages, coverage, budgets. The style scanner checks the skin, the free writing tools hub holds the rest of the kit, and the study guides hub holds the craft guides this page assumes.
Reflection blocked because the event still feels messy?
If the deadline is close and the incident will not organise itself, a quoted, deadline-bound writer can help you structure your own account around your placement notes while the experience and the voice stay yours. Get a free price quote and see the terms first.
Quiz: five questions on reflective structure
Five questions on the Gibbs cycle, the failure modes and the honest limits of scaffolding. Every wrong answer explains itself, and four options keep the guessing honest.
1. What does this builder actually do?
2. In the Gibbs cycle, which stage comes immediately after Description?
3. Your Description stage reads as a tidy retelling: events in order, no people named, no times, no place. What is the strongest fix?
4. Why does the Analysis stage lean on connectives like "because", "therefore" and "however"?
5. The emotion-word bank inserts words like "anxious" or "relieved" into your Feelings stage. What can it not do?
Frequently asked questions
Can I use the word I in a reflective essay?
Yes, and you are expected to. Reflection is written in the first person because the assignment asks what you did, felt and learned — a detached third-person voice usually costs marks. Two exceptions to check in your rubric: some modules still want academic support in the Analysis stage, and professional programmes often want patient details anonymised, so the ward stays the ward and names become roles.
How long should each Gibbs stage be?
Roughly in proportion to the work each stage does. For a 1,000-word reflection, a common split is about 250 words of Description, 100 of Feelings, 200 of Evaluation, 250 of Analysis, and 100 each for Conclusion and Action Plan — the builder's default 25/10/20/25/10/10 percentages. The two thinking stages earn the depth; Description is context, not the main course. If your brief assigns its own split, the brief wins.
Do reflective essays need references?
Often, yes — usually in the Analysis stage, where you connect what happened to a module concept, a guideline or a study. Description and Feelings normally need none. A reflection with zero references can read as if the literature does not exist; one where every sentence is a citation can read as if you were not there. The balance most rubrics reward is your experience first, the literature called in as a lens.
Does this tool write the reflection for me?
No, and that is a design decision rather than a missing feature. It inserts prompts and optional words, tracks your budgets and flags thin stages — but every sentence in the export comes from your keyboard. A reflection written from invented feelings collapses under a viva question or a seminar follow-up, and the specific, honest details are exactly what the rubric's top band describes. The demo text on this page is invented and labelled; do not submit it.
Which model should I choose, Gibbs or What? So What? Now What?
Follow the brief first: if it names Gibbs, use Gibbs, and if it names Driscoll or Borton, use the three-stage version. With a free choice, Gibbs suits a significant incident with several layers — feelings, evidence, causes, a plan — because six stages force each to get its own space. What? So What? Now What? suits a shorter word count or a simpler event; it compresses the same logic into three moves, and its So What? stage does the heaviest lifting.
Did the builder spot something your draft was missing?