The one idea
The deck is not the talk. You are the talk.
A slide's only job is to make the thing you are saying easier to follow — a number, an image, a structure, a name. When a slide contains everything you are about to say, your audience reads it faster than you can say it, and then you are the slow audiobook of a document they have already finished.
Slides should show the one thing you're talking about, not everything you're going to say.
Build the argument first. Then decide, per point, whether a slide helps. Many points do not need one.
People cannot read text and process speech simultaneously without interference. Dense text slides put your narration in direct competition with the reading your audience is already doing.
Build the argument before opening the software
The most common error happens before any design decision: opening slide software first. The tool immediately offers you a title and a bullet list, so you start filling in bullets, and you end up with a formatted outline rather than an argument.
Do this on paper, or in a plain document:
- What do I want them to do or decide? One sentence. If you cannot write it, no deck will rescue the talk.
- What must they accept for that to be reasonable? Usually three or four things. These are your sections.
- What is the evidence for each? One or two pieces. Your best evidence, not all of it.
- What will they object to? Address the strongest objection yourself, before they raise it.
Now you have the talk. Only then ask, for each point: does a slide help here? Sometimes the answer is no, and the strongest moment in a presentation is often a black screen while you say the thing that matters.
The argument, written before any slide exists:
Ask: approve two extra contractors for Q4.
They must accept: (1) we will miss the December launch as staffed, (2) the slip costs more than the contractors, (3) two people is the right number, (4) we can onboard them fast enough to matter.
Evidence: velocity against plan; cost of a six-week slip; the specific workstream that is short; the onboarding plan.
Objection: "contractors take longer to onboard than they save." Answer it on the slide before the ask.
That is a nine-slide deck, and it took twelve minutes to plan. Most people would have opened PowerPoint and produced twenty-five slides of status update with the ask buried on slide twenty-three.
One idea per slide
If a slide has two ideas, it has one too many. Split it — slides are free.
The title should be the point, not the topic. "Q4 revenue" is a topic. "Q4 revenue fell 12%, all of it in the North region" is a point. Someone who reads only your slide titles should get the whole argument. This is the same principle as headings in a document someone can use, and it does more for a deck than any design choice.
Text density. Rough guide: if a slide takes longer than about five seconds to read, it is competing with you. A short phrase, a single number, a chart, an image. Not sentences you will read out.
Never read your slides aloud. It is the fastest way to lose a room. Your audience finished reading before you got to the second line, and now they are waiting for you — which feels, from their side, like being talked down to. The slide and your voice should be doing different jobs.
Charts on slides
A chart in a report can be studied. A chart on a slide gets about eight seconds of divided attention, so it must be readable at a glance and from the back.
- One message per chart. If it supports two claims, use it twice, highlighted differently.
- Put the message in the title. "North region drove the shortfall", not "Regional revenue Q4".
- Highlight the thing you are talking about. Grey everything else. Colour is a pointer, not decoration.
- Cut the furniture. Gridlines, legends where direct labels work, decimal places nobody needs, 3D effects.
- Never paste a spreadsheet screenshot. Nobody can read a screenshot of a table of numbers on a projector, and pasting one signals that you did not decide what it showed.
If the chart genuinely needs studying, it belongs in a handout.
Handout or presentation? They are different documents
This is the distinction that resolves most slide arguments.
| Presentation deck | Handout / read-alone deck | |
|---|---|---|
| Read | While you speak | Without you |
| Text | Minimal | Complete enough to stand alone |
| Detail | In your voice | On the page |
| Charts | One message, glanceable | Can be studied |
The compromise — a dense deck that half works for both — fails at both. It is too heavy to present from and too fragmentary to read alone.
What to do instead: build the sparse deck for the room. Then, if people need something to take away, either write the detail into the speaker notes and send that version, or send a short document. Presenting from a deck and circulating something more complete afterwards is completely normal and expected.
Your manager writes the day before:
"Can you send me the deck for tomorrow so I can look through it tonight?"
If you send the presentation deck, she sees eleven slides with a number on each and no argument, and she will conclude you are underprepared.
Reply:
"Sending two things — the deck I'll present from, which is deliberately light, and a one-page summary of the argument and the ask so you can follow it tonight without me talking over it. The one-pager is the one to read."
You have explained the sparse deck rather than defending it, and you have given her the thing she actually asked for.
Building a deck
1 of 7Write the ask in one sentence. What do you want them to decide, approve or do? Everything else serves this.
Try this
Here is slide 4 of a deck:
Q4 Performance Update
- Revenue for Q4 was 8.2 crore against a target of 9.3 crore, representing a shortfall of approximately 12%
- The North region underperformed significantly, contributing 78% of the total shortfall, with South, East and West broadly on plan
- Primary causes appear to be the delayed product launch in October and a competitor price reduction in early November
- Recommendation is to revise the Q4 plan for North and reassess pricing
- Further detail is available in the appendix
Rebuild it.
Your challenge
Level 3 · IndependentTake a deck you have presented, or one someone sent you.
Write out the argument on one page first: the ask, what must be accepted, the evidence, the main objection. Then rebuild the deck from that page, one idea per slide, with every title stating a point.
You have succeeded when someone can read only your slide titles, in order, and correctly describe your argument. If the titles read as topics, you have built a contents page, not an argument.
What people usually get wrong
- Opening the software first. You will produce a formatted outline instead of an argument.
- Reading your slides aloud. The audience finished before you started, and it reads as condescension.
- Titles that name topics. "Q4 Revenue" instead of the finding wastes the most-read text on the slide.
- One deck for presenting and reading. It fails at both. Make the sparse one and send something else.
- Screenshots of spreadsheets. Illegible on a projector, and an admission you did not decide what the data showed.
- Charts with everything in a different colour. Colour is a pointer. If everything is highlighted, nothing is.
- No appendix. The first hard question exposes it.
- Rehearsing silently. You skip the sentences you cannot actually say aloud.
- Burying the ask at the end. Meetings run over and the ask is what gets cut.
How someone experienced does it
Experienced presenters know the deck is a fallback, not the plan. They can give the whole talk with no slides at all, which is what makes them calm when the projector fails — and it happens often enough that the calm is worth having.
They also front-load. Most decks build to a conclusion, which assumes the audience will still be there. In reality a senior audience will interrupt at slide three and the meeting will end early. So they put the ask and the answer first, and use the remaining slides as the evidence for anyone who wants it. If the meeting ends after six minutes, the point has already landed.
The most useful habit is designing for the interruption. They know which three slides they would keep if given four minutes instead of twenty, and they mark them. Someone senior will say "just get to the point" — and having already decided what the point is, they get there without visibly abandoning a plan.
When not to use this
Not every meeting needs slides, and reaching for them by default is a real failure of judgement.
A discussion where you want people talking is actively hurt by a deck — projecting slides makes people watch a screen rather than each other. A decision that fits in a one-page document is better as a one-page document read in the first three minutes, which many organisations do deliberately. A status update usually needs a written summary, not fifteen slides.
The test: if you want them to listen and respond, a deck helps. If you want them to think, argue and decide together, a document read in the room is usually better, because everyone has the same information at the same time instead of one line at a time under your control.
Prove it
Rebuild one deck from the Challenge, then read only the titles aloud in order.
If that reads as a coherent argument, the deck will work. If it reads as a list of subjects, you have not decided what you are claiming yet — and that is a thinking problem the slide software was hiding from you.
Keep learning this
Paste this into any AI assistant. It turns the assistant into a tutor that tests you instead of just answering you.
Act as an experienced practitioner who is good at teaching. I have just learned building presentation slides that support a spoken argument. Assume I am intelligent but relatively new to this — treat me as beginner level. Work through this in order, and wait for my reply at each step: 1. Ask me 5 questions that test whether I actually understood building presentation slides that support a spoken argument. Do not reveal the answers yet. 2. After I answer, tell me which parts I got right, which I got wrong, and which I only half-understand. Explain only what I misunderstood — do not re-teach what I already know. 3. Give me one practical challenge based on something I could genuinely encounter at work or in daily life. Do not solve it for me. 4. Evaluate my solution the way an experienced person would judge it, including what a professional would have done differently. 5. Tell me what to learn next, and why that comes next. 6. Give me trustworthy sources for deeper study — prefer official documentation, primary research or standards bodies over blogs and videos. Rules for you: no buzzwords. No motivational filler. Say "I'm not certain" when you are not certain, and tell me which parts of your answer I should verify myself. Clearly separate facts from your recommendations and your opinions.
Become independent at this
Use this when you want a path from where you are to actually good, with checkpoints you can test yourself against.
I want to become independently capable at building slides that support you instead of replacing you — not permanently dependent on AI, tutorials or step-by-step guides. Design a progression for me with five stages: Beginner, Guided practice, Independent practice, Real-world application, Professional level. For each stage tell me: - what I must know - what I must be able to do without help - the mistakes people make at this stage - one practical challenge - one real project that would prove I reached this stage - one way I can test myself honestly Then tell me the signals that I am ready to move to the next stage, and the signals that I have skipped ahead too early. Keep the theory to the minimum I actually need. Focus on ability I can transfer to situations you and I have not discussed.