Find out what an examiner notices in your first 50 words
Tests your essay the way it is actually read — fast — separating conspicuous errors from invisible ones and naming the feature that tips the band.
An examiner spends a limited time on each script and forms an impression fast. Essay: [MY ESSAY] 1. Reading only my first 50 words, what impression forms, and what band range does it suggest? Be specific about what creates the impression. 2. Reading only my paragraph openings, does the essay's structure read clearly? 3. Now the whole script: what are the three most VISIBLE features — the things that stand out on a fast read, good or bad? Visibility is not the same as importance; be clear about which is which. 4. Which of my errors are invisible on a fast read and which are conspicuous? 5. If the examiner were deciding between two adjacent bands, which single feature would tip it, and which way?
Best on Claude · ChatGPT · Gemini.
What to fill in
| Token | What to paste |
|---|---|
| [MY ESSAY] | Your full essay. This prompt reads it in three passes, so it needs all of it. |
What you should get back
A first-impression read of your opening 50 words, a structural skim of your paragraph openings, your three most visible features, a conspicuous-versus-invisible error split, and the single feature that would tip a borderline band.
What it actually returned
Tested on Gemini 2.5 Flash, API, August 2026, 2026-08-06. Trimmed — this is an excerpt of a longer response, not the whole thing.
1. Reading only my first 50 words, what impression forms, and what band range does it suggest? First 50 words: "In today's modern world, many people believe that governments should spend money on public transport rather than building new roads. I completely agree with this opinion because of two reasons which I will discuss below." Impression: The initial impression is clear and direct. The essay immediately addresses the prompt, states a clear opinion, and signposts the structure. What creates the impression: — Directly addresses the prompt: No ambiguity about the topic. — Clear thesis statement: "I completely agree with this opinion." — Explicit signposting: "because of two reasons which I will discuss below" — Slightly generic opening: "In today's modern world" is a com…
Our verdict on this run
It did isolate the opening properly, quoting the first 50 words back before judging them, which is better than the contamination this prompt usually suffers from. The observations are sound and the generic-opener flag is fair. What it would not do is commit to the band range the question asked for — it described the impression at length and then moved on, so if you want that number you have to ask again. The most valuable section came later: the split between conspicuous and invisible errors, which is the part that tells you where to spend the last five minutes of the exam.
How this prompt goes wrong
It cannot actually restrict itself to the first 50 words. The whole essay is in its context, so the "first impression" is contaminated by everything that follows — you get a summary of the essay presented as an opening impression. The fix is mechanical: paste only the first 50 words for point 1 in a separate chat, then the full essay for the rest. Point 4 is also worth checking, because models tend to classify errors by severity rather than by how noticeable they are, which is the distinction that makes this prompt useful.
Tips
- Salience is real: an error in your first sentence costs more than the same error in the middle of paragraph three.
- Point 5 tells you where to spend your five proofreading minutes at the end of the exam.
- Run point 1 in its own chat with only the opening pasted, or the answer is contaminated.
Related prompts
- Mock examinerA 250-word verdict with no room for flatteryA tired, strict examiner persona capped at 250 words: the band first, three quoted justifications, one upgrade condition, nothing else.
- Mock examinerA 40-minute Task 2 test where the model says nothingOne realistic question, then silence for 40 minutes, then a compact evidence-based mark. No tips, no outline, no encouragement.
- Mock examinerRun a full Speaking test where the examiner never breaks roleA complete Parts 1–3 interview in examiner register — no teaching, no encouragement, no band — so it feels like the real thing.
- Writing Task 2Analyse only coherence and cohesion, nothing elseMaps your controlling ideas, referencing chains and linker repeats — the criterion most learners cannot self-assess — with grammar and vocabulary explicitly off-limits.
Common questions
Do examiners really decide in the first 50 words?
They do not decide, but they do form an expectation, and a script that opens badly has to work harder afterwards. Treating the opening as high-value real estate is sound preparation regardless.
Why does the first-impression answer describe my whole essay?
Because the model can see all of it. Restricting attention is something prompts cannot enforce — paste only your opening in a separate chat if you want a clean read.
What counts as a conspicuous error?
Anything visible without close reading: a wrong verb form in the first sentence, a paragraph with no clear opening idea, a linker repeated five times, an essay that is visibly short. Subtle article errors mid-paragraph are the invisible kind.