Do you even need a presentation?
Sumeet Gayathri Moghe (byline metadata reads Martin Fowler; published as first article of a guest series on martinfowler.com, editorially reviewed by Fowler) · rss · 2026-09-12 Reader 🖥️
A 'presentation' properly means the live, presenter-orchestrated act of storytelling, not the slide deck, so before building one you should work up a hierarchy of cheaper alternatives — document, infodeck, interactive web app, recorded audio/video — and reserve an actual (especially live) presentation for the rare cases that genuinely require synchronous human interaction.
practical · argumentative — The piece argues for action: choose the lowest-overhead communication medium sufficient for your goal, and reserve live presentations for situations that specifically require synchronous interaction. It builds a five-rung taxonomy (document, infodeck, web app, recording, live presentation) and argues, rung by rung, for defaulting downward. Illustrative narratives (the son's memoir, Ed Zitron, the author's own videos) support rather than replace the argument.
Structure
| Part | Role |
|---|---|
| Opening definition: a presentation is storytelling (narrative, timing, emotion orchestrated by a presenter), not the slide file. | thesis statement / redefinition that reframes the rest of the piece |
| "Most of the time, a document is enough" — case against using slides as a document format, case for linear writing plus writing-quality tools (Word, Hemingway, Grammarly). | evidence for the first, cheapest rung of the hierarchy |
| "Slideware as a companion tool" — slides are fine for making diagrams to embed in a document, just not as the primary artefact. | qualification / nuance on the document argument |
| "Infodecks are the novice designer's page layout solution" — defines infodeck/slidedoc, distinguishes from 'slideument', lists conditions for success (writing quality + design skill) and Thoughtworks/Fowler examples. | second rung of the hierarchy, with success conditions |
| "Use web interactions and applications to manage cognitive load" — coding-harness-built interactive content (word cloud, microcourse) as a third rung, with the commenting/shelf-life trade-off noted. | third rung of the hierarchy, with its own limitation |
| "Recorded audio and video are a substitute for live presentations" — argues recordings are usually the better default even when a presenter's voice is needed: audience controls pace, captions/chapters, forces brevity; discusses build/transition pacing via two of the author's own videos. | fourth rung, with the pivot argument that most presentations should be recordings rather than live |
| Cheat-sheet table mapping each artefact (document/infodeck/web app/recording/live) to the situation it fits. | synthesis of the whole hierarchy into a decision aid |
| "So, when do you go live?" — live presentation reserved for cases needing synchronous interaction (real-time influence, objection-handling, networking, community energising); demands heavy prep since the audience can't control pace. | conclusion: the top, most expensive rung, and the criterion for climbing to it |
| Closing pivot: having decided whether to present, the series will next address in-person/online/recorded delivery and, immediately after, how to craft the underlying story. | scope-setting for the series; explicitly defers presentation-craft itself |
Key terms
- presentation Presentation & Public Speaking Craft
An act of storytelling in which a presenter orchestrates narrative, timing, and emotion to create meaning that neither visuals nor documents can achieve alone — the human performance, not the file. - infodeck Presentation & Public Speaking Craft
A document built in slide-layout software, meant for standalone 'light reading' rather than as a presentation aid or as a hybrid presentation/document; succeeds only when both the writing and the information design are strong. - slideument / slidedoc
A hybrid artefact that tries to serve as both a live-presentation aid and a standalone document; the author explicitly excludes infodecks from this category ("Infodecks aren't slideuments"), implying slideuments fail at both jobs. 'Slidedoc' is credited to Duarte Design as another name for infodeck, used approvingly, while 'slideument' is used dismissively. - cognitive load (audience-managed) Cognitive Load
The mental burden placed on a reader/viewer when they, rather than the interactive medium, must do the work of organizing and pacing through information (e.g. scanning a dense document or spreadsheet unaided); interactive web apps are framed as a way to shift that burden from audience to artefact via chunking and guided exploration.
Propositions
- explicitA presentation, properly defined, is the presenter's live act of storytelling — not the slide deck or slideshow file.conf: high
- explicitSlideware is a poor format for well-structured written communication because it forces readers to consume information in fragmented, disconnected bursts.conf: high
- explicitLinear documents outperform slide decks for most corporate information transfer because they let readers build understanding progressively, integrate with writing-quality tooling (Word/Hemingway/Grammarly), and reduce the risk of LLM-summarization hallucination through explicit, bridged prose.conf: high
- explicitInfodecks succeed only when both the underlying writing is disciplined/concise and the visual/information design is skilled; the scarcity of design sense among knowledge workers is the main reason most infodecks fail.conf: high
- explicitInteractive web applications (built via coding harnesses) can reduce an audience's cognitive load better than static documents by chunking content and adding guided exploration, but they typically lack native commenting/feedback affordances and have an uncertain maintenance shelf life.conf: high
- explicitRecorded audio or video is usually a better default than a live presentation, because the audience controls pace (speed up/slow down/replay), gets captions and chapter markers, and the act of recording forces the presenter to cut fluff.conf: highextendsHow to Speak · p5 — Winston frames slides as suited to 'exposing ideas' (job talks/conferences) versus boards for teaching; this piece generalizes the same slides-are-a-tool-not-the-message point into a full medium-choice hierarchy and adds recorded video as a distinct, often-superior rung Winston's live-only lecture doesn't consider.
- explicitThere is no fixed formula for how many slide builds/transitions a recorded talk needs — the right number is whatever serves the story, and can vary by roughly 5x between formats (129 over 8 minutes vs. 141 over 24 minutes in the author's own examples).conf: high
- explicitA live presentation should be reserved for situations that specifically require synchronous human interaction — real-time influence, fielding objections, networking, energising a community — because it is the costliest option: it demands scheduling overhead and heavy preparation, and (unlike a recording) the audience cannot control its pace.conf: highextendsHow to Speak · p1 — Winston argues speaking/writing/idea-quality determine success and teaches how to speak well, but never asks whether the situation calls for live speech at all; this piece supplies exactly that prior, missing decision criterion (reserve live presentation for cases needing synchronous interaction).
- implicitCommunication media form an implicit cost/overhead hierarchy — document, infodeck, interactive web app, recording, live presentation, in ascending order of production and audience-attendance cost — and the rational choice is the cheapest rung that still achieves the communicative goal.conf: medium
- implicitOrganizations and individuals default to over-using slide decks and live presentations for information that would be better served by writing — otherwise there would be no need to write an article arguing readers should ask 'do you even need a presentation?'conf: high
Arguments
For most corporate communication, a linear written document should be the default medium, not a slide deck. strength: medium
Explicit premises:
- Slides used as a document format fragment information into disconnected bursts.
- Documents let readers build understanding progressively through linear structure.
- Writing-quality tools (Word, Hemingway, Grammarly) exist to improve document clarity and can't easily be applied to slide text.
- Explicit, linear writing is easier for both humans and LLM summarizers to parse accurately, reducing miscommunication/hallucination risk.
- Editing and reorganizing text is easier in documents than in slides.
Implicit premises:
- Reader comprehension is better served by completeness and linearity than by visual chunking alone.
- Most corporate communication is consumed asynchronously, so it doesn't need a live presenter's orchestration to land.
- The audience is time-poor and willing/able to read prose rather than needing everything pre-visualized.
Inferences:
- Teams that habitually write internal updates, proposals, or status reports as slide decks are producing systematically worse communication artefacts than if they wrote documents.
- Because LLM summarization is now common, the 'reduces hallucination risk' argument gives document-first communication a new, AI-era justification beyond the traditional readability case.
Live presentation should be the medium of last resort, reserved for situations that specifically require synchronous interaction. strength: medium
Explicit premises:
- Live presentations require significant scheduling overhead (bringing people together in the same physical or virtual space).
- Live presentations demand heavy preparation because, unlike recordings, the audience cannot slow down, speed up, or replay the presenter.
- The distinctive value of 'live' is real-time interaction: influencing people and fielding objections as they arise.
- Recordings capture most of the benefits of a presenter's voice (tone, pacing, drama) without the scheduling and pacing-control costs.
Implicit premises:
- Real-time influence and objection-handling genuinely cannot be replicated asynchronously — i.e., synchronous presence has an irreplaceable function, not just a habitual one.
- The overhead of a live event (everyone's time, scheduling) is a real cost that organizations under-price when defaulting to meetings.
Inferences:
- A large share of existing recurring live meetings/presentations in organizations (status updates, FYI briefings) are probably miscategorized and should be converted to documents or recordings.
- If synchronous interaction is truly the differentiator, then a live presentation with no Q&A or discussion segment has lost its own justification and should have been a recording.
Narratives
- The author laid out his son's trip memoir in PDF using slideware, producing a decent-looking ebook without specialized design skill.
→ Supports the proposition that slide-based tools are 'the novice designer's page layout solution' — a low-skill-barrier way to produce infodeck-style artefacts. - Ed Zitron delivers regular voice-only monologue videos on YouTube with no visuals and has built a cult following.
→ Supports the claim that a well-written monologue alone (audio or talking-head) is sufficient to convey ideas compellingly, without needing slides at all. - The author's own two videos — an 8-minute anti-brainstorming case with 129 builds/transitions, versus a 24-minute on-screen talk about cultural identity with 141 builds/transitions — are contrasted for pacing.
→ Supports the proposition that there is no fixed formula for slide-build density; the right pacing depends on format (voiceover-only vs. on-screen) and story needs, not a rule of thumb. - A microcourse rebuilding a long-form article on leadership and cultural identity by chunking it, adding interactions, and inserting reflection questions.
→ Supports the claim that interactive web-app formats can manage audience cognitive load better than a static document, by actively pacing and testing comprehension rather than leaving that work to the reader.
Questions addressed
- ✅ What is a presentation, as distinct from a slide deck?
Defined explicitly as the presenter's live storytelling act (narrative + timing + emotion), with the slide file recast as, at most, a supporting tool. - ✅ When is presenting (in any form) actually necessary, versus a document, infodeck, or interactive web app?
Gives a five-rung hierarchy plus a cheat-sheet table matching each artefact to the situation it fits, and states the presentation should be reserved for when 'the story needs your voice.' - ⭕ Given that a presentation (voice) is warranted, should it be live or recorded?
Argues recordings are usually the better default and gives reasons, but explicitly defers the full live-vs-recorded-vs-online treatment to 'the remaining articles in the series' — so the question is opened here, not closed. - ✅ How many slide builds/transitions should a recorded talk use?
Explicitly answered as 'there's no formula' — the author declines to give a rule and instead offers two contrasting personal data points as anchors, effectively resolving the question by dissolving it. - ⭕ How does a team or individual decide, in a given case, whether the interactive-commenting feature they lose by choosing a web-app format actually matters?
Gaps
- incompleteThe piece asserts several strong causal/behavioral claims (documents build understanding better than slides; recordings reduce hallucination in LLM summarization; interactive apps reduce cognitive load) without citing any study, user test, or even a specific before/after example from the author's consulting practice. All of it reads as seasoned-practitioner intuition rather than evidence.
Read next: Cognitive-load and multimedia-learning research (e.g. Sweller's cognitive load theory, Mayer's multimedia learning principles) would let a reader assess how much of this holds up empirically versus being a persuasive-but-untested house style. - incompleteThe article never addresses the organizational/political reasons teams keep defaulting to live meetings and slide decks despite the overhead the author identifies (status signaling, visibility, habit, hierarchy expectations) — it treats the choice as a purely rational medium-fit decision, ignoring why the 'irrational' default persists.
Read next: Read into organizational-change or meeting-culture literature (e.g. work on meeting overload, status theatre, or Cutler-style institutional-overload framing) for why the rational choice the author recommends is hard to adopt in practice. - incompleteProduction-cost realism is thin: the author (self-described non-software-engineer) asserts you can build a web app or produce cinematic video 'without high-end skills' via coding harnesses and consumer tools, but doesn't address the actual skill floor, tooling access, or organizational approval overhead in a regulated or non-creative-adjacent workplace (e.g. a telco) where such tools may not be readily available or sanctioned.
Read next: Would benefit from a practitioner account of building infodecks/microcourses inside a large, non-media, non-consultancy organization to test whether the low-barrier claim survives outside a design-forward consultancy like Thoughtworks. - incompleteThe article is explicitly scoped as the first of a series and defers the actual craft of storytelling/narrative construction — the piece that would tell you HOW to write the story once you've decided a presentation is warranted — to 'the remaining articles.' As captured, this document only answers whether/which medium, not how to execute the chosen one well.
Read next: Read the follow-up articles in Moghe's series (next: 'how to craft your story') once captured in Reader, and also Winston's 'How to Speak' (extract-winston-how-to-speak) for live-delivery craft this piece deliberately does not cover.
Closes gaps elsewhere
- How to Speak · g1 — Winston's lecture was flagged as covering only in-person live speaking, with no treatment of video, remote, or mediated formats. This piece directly supplies that missing treatment: recorded audio/video as a first-class, often-preferable alternative to live presentation, plus infodecks and interactive web apps as further mediated alternatives, with explicit criteria for when each is appropriate.