# Public Speaking and Presentations > Answer four questions before touching a slide tool: Parent: [Writing and Documentation](https://llms-explorer.com/tree/writing-and-documentation/) · 17 facets · 73 facts · page: https://llms-explorer.com/tree/public-speaking-and-presentations/ ## Audience Analysis Before You Open PowerPoint - Answer four questions before touching a slide tool: — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#audience-analysis-before-you-open-powerpoint) - Who is in the room? Job function, seniority, and domain expertise. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#audience-analysis-before-you-open-powerpoint) - What do they already believe? Start from their current position. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#audience-analysis-before-you-open-powerpoint) - What one action do you want them to take? Name the action. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#audience-analysis-before-you-open-powerpoint) - What is the one thing they must remember? Write that sentence before slide 1. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#audience-analysis-before-you-open-powerpoint) ## Slide-Doc vs Presentation Slides - Presentation slides (Reynolds / "Presentation Zen" model) exist to support a live speaker. The speaker carries the argument; the slide carries visual reinforcement. One idea per slide. Minimal text. Large images or diagrams. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#slide-doc-vs-presentation-slides) - Slide-docs (Tufte-influenced) are standalone documents formatted as slide-shaped pages. They contain complete sentences, dense supporting data, footnotes. They can be emailed and read without a presenter. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#slide-doc-vs-presentation-slides) - The core failure mode: the slidument - bullet-heavy slides that are too dense to look good on screen and too sparse to be useful to a reader. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#slide-doc-vs-presentation-slides) ## Kawasaki 10/20/30 Rule - 10 slides - humans can't absorb more than ten concepts in a single sitting — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#kawasaki-102030-rule) - 20 minutes - even if you have a full hour, present in 20 minutes — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#kawasaki-102030-rule) - 30-point minimum font - if you cannot fit your message in 30pt type, you have too many words — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#kawasaki-102030-rule) ## Nancy Duarte's Resonate Framework - The central insight: every great speech moves back and forth between "what is" and "what could be." — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#nancy-duartes-resonate-framework) - The contrast pattern: — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#nancy-duartes-resonate-framework) - Establish the current reality (what is) — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#nancy-duartes-resonate-framework) - Describe the future possibility (what could be) — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#nancy-duartes-resonate-framework) - Repeat this oscillation; each cycle raises the stakes — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#nancy-duartes-resonate-framework) - End with the "new bliss" - the world after the audience acts — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#nancy-duartes-resonate-framework) - The STAR moment (Something They'll Always Remember): one designed moment of high contrast, surprise, or visceral demonstration. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#nancy-duartes-resonate-framework) ## Three-Act Talk Structure - Act 1 - Setup (10–15% of total time): Establish stakes; define the problem; signal what the audience will be able to do differently by the end. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#three-act-talk-structure) - Act 2 - Exploration (70–75%): Develop the argument. Front-load complex material; place a re-energizing moment two-thirds of the way through. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#three-act-talk-structure) - Act 3 - Resolution (10–15%): Do not introduce new material. Synthesize, do not summarize. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#three-act-talk-structure) ## The Headline-Deck Approach (McKinsey / BCG) - Every slide title is a complete, standalone sentence stating the takeaway, not a topic label. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#the-headline-deck-approach-mckinsey-bcg) - Bad title: "Market Analysis" — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#the-headline-deck-approach-mckinsey-bcg) - Good title: "The North American market is growing 12% YoY and outpacing EMEA in every segment." — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#the-headline-deck-approach-mckinsey-bcg) - A reader who reads only slide titles should follow the full argument. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#the-headline-deck-approach-mckinsey-bcg) ## Talk Openings — Five Patterns - Startling fact or statistic: drop a number the audience does not expect — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#talk-openings-five-patterns) - Story: open with a specific scene - one person, one moment, sensory detail — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#talk-openings-five-patterns) - Contrarian claim: state something the audience believes that you are about to challenge — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#talk-openings-five-patterns) - Direct question: ask a question the audience cannot immediately answer — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#talk-openings-five-patterns) - Scene-setting: describe the world as it will look after your idea is adopted — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#talk-openings-five-patterns) - Do not open with: your name and title, an apology, "great to be here," agenda slides. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#talk-openings-five-patterns) ## Talk Closings — Five Patterns - Callback to opening: return to the story or question you opened with — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#talk-closings-five-patterns) - Call to action: one specific, concrete thing the audience can do in the next 48 hours — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#talk-closings-five-patterns) - Single image or quote: full-bleed slide with one sentence — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#talk-closings-five-patterns) - Summary triple: three short phrases that encapsulate the talk's three main points — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#talk-closings-five-patterns) - The future vision: describe the world three years from now if the audience acts — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#talk-closings-five-patterns) ## Q&A Handling - The bridging technique: acknowledge the question, then bridge to your main message. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#qa-handling) - "I don't know" as a complete answer: say it cleanly, then add "I'll find out and get back to you by [date]." — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#qa-handling) - Hostile questions: do not match the energy. Lower your voice slightly. Answer the question beneath the question. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#qa-handling) ## Rehearsal Discipline - Minimum effective rehearsal for a 30-minute talk: — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#rehearsal-discipline) - Three times out loud, alone - full volume, full sentences, full transitions — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#rehearsal-discipline) - Once to a real human - their confusion is signal — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#rehearsal-discipline) - Video review of yourself - watch on mute first (body language), then with sound (pace and filler) — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#rehearsal-discipline) - Pacing target: 125–150 words per minute. Time your talk at 90% of the slot. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#rehearsal-discipline) ## Virtual vs In-Person Calibration - Virtual: look at the camera lens, not at faces on screen. Place the camera at eye level. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#virtual-vs-in-person-calibration) - Audio quality (virtual): audio quality matters more than video quality. A USB condenser microphone, a quiet room, and a soft surface behind you. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#virtual-vs-in-person-calibration) - Energy calibration (virtual): increase vocal variety noticeably above what feels natural in a room - the camera flattens affect. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#virtual-vs-in-person-calibration) ## Anti-Patterns to Eliminate - Reading slides verbatim. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#anti-patterns-to-eliminate) - Walls of bullets. Each bullet resets the audience's attention to near-zero. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#anti-patterns-to-eliminate) - "As you can see here..." - If it is visible, do not narrate its existence. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#anti-patterns-to-eliminate) - Apologizing for slides. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#anti-patterns-to-eliminate) - Running long. Going over time disrespects every person in the room. — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#anti-patterns-to-eliminate) - The vague closing. "So, in conclusion, we looked at X, Y, and Z..." — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#anti-patterns-to-eliminate) ## Sources - Reynolds, Garr. "Presentation Zen" (2nd ed., New Riders, 2011) — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#sources) - Duarte, Nancy. "Resonate" (Wiley, 2010) — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#sources) - Kawasaki, Guy. "The Art of the Start 2.0" (Portfolio/Penguin, 2015) — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#sources) - Anderson, Chris. "TED Talks: The Official TED Guide to Public Speaking" (Houghton Mifflin Harcourt, 2016) — [source](https://llms-explorer.com/sources/mdb-context-hub/public-speaking-and-presentations/#sources) ## Where this helps - Preparing a conference keynote or TED-style talk where the goal is a memorable single takeaway rather than exhaustive coverage of a topic. — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Deciding whether a deck should be built as presentation slides (visual support for a live speaker) or as a slide-doc (a standalone document that survives being emailed and read without you). — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Structuring a technical readout for executives using the headline-deck approach, where a reader skimming only titles should still follow the argument. — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Diagnosing why a rehearsed talk keeps running long, and calibrating rehearsal and pacing before the actual delivery. — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* ## How to apply this - Answer the four audience-analysis questions — who's in the room, what they believe, what action you want, what they must remember — before opening any slide tool. — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Choose openings and closings from the documented patterns (startling fact, story, or contrarian claim for openings; callback, call to action, or summary triple for closings) rather than defaulting to a title slide and a "questions?" close. — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Write every slide title as a standalone, complete-sentence takeaway using the headline-deck approach, so a reader skimming only titles can follow the full argument. — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Rehearse the talk out loud at least three times alone at full volume, once in front of a real person, and once reviewed on video muted for body language before delivering it live. — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* ## Common mistakes - Reading slides verbatim instead of letting the slide carry visual reinforcement while the speaker carries the argument. — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Building walls of bullet points, where each new bullet resets the audience's attention back to near-zero. — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Narrating what's already visible on the slide ("as you can see here...") instead of trusting the audience to read it. — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Closing with a vague "in conclusion, we looked at X, Y, and Z" instead of one of the documented closing patterns — callback, call to action, single image, summary triple, or future vision. — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* ## Known issues - The slidument failure mode is easy to fall into by trying to serve both live and asynchronous readers at once — a deck dense enough to work as a standalone slide-doc is usually too text-heavy to work well projected during a live talk, and vice versa. — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - The Kawasaki 10/20/30 rule and the three-act time allocation are calibrated for pitch- and keynote-length talks; a longer technical session or workshop needs a different time-budget structure even if the same narrative principles still apply. — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Virtual delivery requires actively increasing vocal variety above what feels natural in person, since the camera flattens affect — a talk rehearsed and calibrated for an in-person room can read as flat when delivered virtually without adjustment. — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - The Q&A "I don't know, I'll get back to you" technique only works if the speaker actually follows up by the stated date — used without follow-through, it damages credibility more than a direct "I don't know" alone would. — [source](https://llms-explorer.com/tree/public-speaking-and-presentations/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* ## Context files - [Public Speaking and Presentations](https://llms-explorer.com/downloads/sources/mdb-context-hub/public-speaking-and-presentations.md)