Slide:ology

Duarte's starting position is that slides are a visual medium being used as a text medium, and that almost every failure of business presentation follows from that category error.

Her argument is that a slide is neither a document nor a teleprompter. A document can be dense because the reader controls the pace; a slide cannot, because the audience must absorb it while also listening to a person speak.

A slide is not a document. The reader controls a document's pace and cannot control a slide's.

Devmark’s reading — not a quotation from the book

Her opening instruction is to work away from the software: generate ideas on paper or a whiteboard, decide the structure, and only then open the application.

Her reasoning is that presentation software imposes a structure — a title and a bulleted list — before any thinking has happened, and that most presentations are shaped by that default rather than by their content.

Her chapters on charts: choose the form that matches the message, remove everything that does not carry information, and label directly rather than using a legend the audience must decode.

Her most useful instruction is that each data slide should make one point and that the point should be stated in the slide's title as a sentence rather than a label. A chart titled 'Q3 Revenue' asks the audience to find the message; one titled with the message does not.

Her core visual chapters: contrast, flow, hierarchy, unity, white space and the use of a consistent grid across a deck.

Her treatment of white space as an active element rather than waste is the one that most changes the appearance of ordinary business slides, because the reflex to fill the available area is what produces the crowded slide in the first place.

Her practical layout guidance: alignment, proximity grouping, visual weight, and the use of a small number of consistent positions across a deck so the audience learns where to look.

Her point that consistency reduces the cognitive cost of each new slide is the argument for a template that most teams apply as branding and few apply as function.

Her guidance on photographs and illustration: choose images that carry meaning rather than decorate, use them at a scale that lets them work, and avoid the generic stock vocabulary that signals nothing.

Her warning about images that merely illustrate a word in the title — a handshake for partnership, a mountain for challenge — names the single most common failure in corporate decks.

Her treatment of animation and transition as tools for directing attention and revealing structure, with the explicit condition that they must serve comprehension.

Her rule that a build should reveal a relationship the audience needs to understand in sequence — rather than simply delay text — is the distinction between animation that helps and animation that irritates.

Her chapters on the wider context: the deck that will be read without a speaker, the handout, the leave-behind, and the different design each requires.

Her insistence that the slide deck and the document are separate artefacts, and that combining them produces something that fails as both, is the argument against the dense slide that most organisations still have not absorbed.

Slides and the document are separate artefacts. Combining them produces something that fails at both jobs.

Devmark’s reading — not a quotation from the book

The software, the visual style and the examples belong to their period, and the book predates widescreen formats, remote presentation, and the now-dominant practice of decks circulated as documents rather than presented.

Its principles transfer, but a reader will be translating the examples constantly, and its silence on presenting to a camera is a real gap.

This book covers the visual craft and *Resonate* covers the narrative structure. They were written in that order and are best read in the reverse.

Structure first, slides second, is the sequence Duarte herself argues for, and a reader who reads only this one risks producing beautiful slides in service of a talk that was never designed.