Reduce on-slide reading load while preserving claims, evidence, qualifiers, and speaker context
4 min setupTested with:ChatGPTReviewed: 2026-08-28
1
Add your context
Your text stays in this browser. AILesson Prompts does not send it to a model or server.
2
Your prompt
Unfilled fields remain visible as placeholders, so you can still copy and edit the prompt
Rewrite the slide text for live projection.
Slides with numbers, titles, and body text:
[slides]
Audience, room, screen, remote/hybrid use, language, and viewing time:
[setting]
Fixed facts, citations, terms, density, speaker-note, and accessibility rules:
[limits]
Give each slide one assertion-style title and one visual message. Shorten syntax, remove prose that competes with the speaker, and move necessary detail to speaker notes rather than deleting it. Preserve all material numbers, periods, denominators, conditions, uncertainty, attribution, and citations. Do not create slogan-like certainty, fragment a sentence into ambiguity, use undefined acronyms, or suggest tiny footnotes as a solution. Return slide-by-slide projected copy, transferred speaker notes, a suggested visual hierarchy, removed text with rationale, and a meaning/density audit. Flag slides that need structural splitting rather than tighter wording.
Private by defaultPrompt assembly happens locally in your browser. Avoid placing confidential information into any AI service unless your organization allows it.
From input to outcome
A worked example
See how concrete context turns this recipe into a usable result
Actual input
Current slide text
Slide 4 title: Results of the trial. Body: During the 12-week period, 96 out of the 143 employees in the pilot teams responded to the survey, and of those respondents, 71% indicated that they would prefer the flexible-shift arrangement to continue. It is important to note that only 18 of the 39 night-shift employees responded and 50% of those respondents preferred continuation. These results may not represent employees who did not respond. Source: Staff survey, June, Table 4.
Audience and display setting
Executive hybrid meeting, large screen plus laptop viewers. Audience gets about 45 seconds on this slide. English. Presenter explains subgroup context aloud.
Content and style limits
Keep both denominators, both percentages, non-response limitation, and source. Maximum title plus two short callouts on screen. Minimum body text equivalent to 24 pt. Details may move to speaker notes; no tiny footnote.
Example output
Slide 4 projected copy
Title: Survey support was positive overall, but weaker among night-shift respondents
Callout 1: All pilot teams — 71% preferred continuation (96 of 143 responded)
Callout 2: Night shift — 50% preferred continuation (18 of 39 responded)
Source block at readable size: Staff survey, June, Table 4. Results may not represent non-respondents.
Speaker notes: The pilot lasted 12 weeks. Emphasize that these percentages describe respondents, not all employees. The night-shift subgroup has a particularly small response count, so the difference should prompt follow-up rather than a firm conclusion about all night-shift employees.
Visual hierarchy: title; two equal-size comparison cards; source/limitation band, not a footnote. Removed “it is important to note” and procedural prose because hierarchy now supplies emphasis. Meaning audit: both response denominators, percentages, subgroup, source, 12-week context in notes, and representativeness limit remain. No causal or population-wide claim added.
Why this works
1
Moving detail to notes lowers visual load without discarding evidence the speaker still needs.
2
Assertion titles tell the audience why the displayed evidence matters before they parse it.
Check the result
Can each slide's main point be understood at a glance without changing its certainty?
Were evidence, qualifiers, citations, and necessary detail retained on slide or in notes?
Are any slides still carrying more than one distinct message?
ChatGPT