← Back to all experiences
Full syllabus · B09 · Write & communicate · Basic

Review edition · Complete text lesson. Platform screenshots are being prepared.

Lab / Work smarter

Suppose you could…

Review and improve a document

Identify factual, structural and tone problems, then produce a justified revision.

◷ 15 minutes✳ First-timerLearn by doing
BY THE END, YOU CAN…

Identify factual, structural and tone problems, then produce a justified revision.

Time is a practice estimate. Pause or return whenever you need.
01 / IMAGINE

A useful little possibility.

Your goal is a usable work artifact, not a perfect-looking AI answer. Identify factual, structural and tone problems, then produce a justified revision.

02 / TRY

Let’s make a start.

Use a fresh conversation in an AI tool you are permitted to use. Copy the fictional sample below, then paste the starter prompt with it. No file upload, web search or paid integration is required.

1. Mark unsupported claims before using AI.
2. Run the prompt.
3. Check the denominator in any percentage.
4. Ask whether the revision still implies measured productivity or an approved launch.

Your sample material

Fictional training material. No real customer information needed.

Verified facts: a pilot had 12 participants; 8 returned feedback; 6 of those 8 said the session was useful; no productivity measurement was collected; another pilot is proposed but not approved.
Draft: “Our revolutionary programme boosted productivity for everyone. All 12 participants loved it. We are launching weekly sessions next month. This proves the approach works.”
Download sample .txt ↓
YOUR STARTER PROMPT
Review the draft against the verified facts. First list issues under evidence, structure and tone. For each factual problem, identify the unsupported claim and its correction. Then rewrite in under 100 words. Explain three substantive changes; do not introduce new evidence.

Keep personal and confidential information out of external AI tools. Start with the sample material.

03 / CREATE

Turn a first try into your thing.

Compare the result with the checklist before opening the worked example. Identify one specific weakness and ask for a targeted revision, referring to the original sample. If all checks pass, tighten the wording without changing the facts. Save the reviewed result in your own document; do not send or publish it as part of this exercise.

04 / UNDERSTAND

Pause for the “aha.”

Revision is more than polishing. Separate factual fidelity from readability so a smoother sentence does not conceal an unsupported claim.

CHECK YOUR RESULT

Useful beats convincing.

Check these against your actual output. Your answers stay on this device; this is self-review, not automated grading.

Finished your first try? Reveal a worked example ↓

Revised: Twelve people joined the pilot. Eight returned feedback, and six of those respondents said the session was useful. This is encouraging feedback from a small, self-selected group; productivity was not measured. A further pilot is proposed but has not been approved.

Changes: replaced universal praise with 6 of 8 respondents; removed an unsupported productivity claim; changed a promised launch to an unapproved proposal. “75% of respondents” would be accurate; “75% of participants found it useful” would not be supported.

Download worked example ↓
NOW MAKE IT YOURS

A new case. Your own approach.

Apply the review principles to these new facts, not the previous sample’s specific numbers or answers.

Verified: 40 invitations, 10 replies, 7 interested, no purchases. Review “70% of our market will buy.” Write a corrected sentence and explain the sampling limitation.

05 / SHARE & REMIX

One possibility leads to another.

Repeat on a small, permitted task of your own. Record what you changed and which check caught a problem. Keep using fictional inputs if workplace material is not approved for your chosen tool.

Your reusable workflow card ↓
Task: Review and improve a document
Audience: [who will use this]
Approved source material: [paste]
Required output: Identify factual, structural and tone problems, then produce a justified revision.
Constraints and unknowns: [list]
Review before use:
- Six of eight respondents is not confused with all 12 participants.
- Productivity improvement is not asserted.
- The next pilot remains unapproved.
- Tone is measured rather than promotional.
- Three meaningful changes are explained.
Download workflow .txt ↓

You tried something. That counts.

Keep a note of what you made or what surprised you. It stays in this browser.