The one idea
Decide the argument before you open the slide software.
A presentation is not a document with pictures. It is one point, defended, in front of people who can ask questions. Slides are a visual aid to that. Opening PowerPoint first is like starting a journey by choosing a font.
The test: if the projector failed, could you still give this talk? If the answer is no, you do not have a presentation. You have a document you intend to read aloud, and the audience can read faster than you can speak.
Work out what you want to say and what you want them to do, then make slides that help you say it.
One message, three supporting points, one ask. Slides carry evidence and structure; you carry the argument.
The audience has one channel for language. When a slide is dense text, they read it instead of listening to you, and you become redundant to your own presentation.
Structure before slides
Write this on paper, in five minutes, before anything else.
| Element | Question it answers |
|---|---|
| The one message | If they forget everything else, what remains? |
| The ask | What do you want them to do or decide? |
| Three supports | Why should they believe the message? |
| The objection | What will the most sceptical person say? |
| The opening line | The first sentence, written out word for word |
Three supports, because an audience holds about three things. Four is a list they will not remember; one feels thin.
On paper, before any slides:
- Message: our support costs are rising because we are answering the same five questions over and over.
- Ask: approve two months of one engineer's time to build self-serve help.
- Supports: (1) 41% of tickets are those five questions; (2) each costs about 20 minutes of an agent's time; (3) three competitors solved this with self-serve and their ticket volume fell.
- Objection: "we tried a help centre in 2023 and nobody used it" — address it in the talk, not in Q&A.
- Opening line: "We are paying people to answer the same five questions four thousand times a year."
That is the whole talk. It fits on an index card. The slides — maybe seven of them — are now obvious, because each one has exactly one job.
Note where the objection sits. Raising the strongest counter-argument yourself, before someone else does, is the single most effective persuasion move there is. It costs you ninety seconds and it converts your biggest sceptic from an opponent into someone who feels heard.
One idea per slide
The rule is simple and almost universally broken: a slide should make one point, and its title should be that point.
Compare:
- Title: "Q3 Support Metrics" — a label. The audience must read the chart and work out what you think about it.
- Title: "41% of tickets are the same five questions" — a claim. The chart below is now evidence for something, and they got the point in one second.
Everything else follows:
- If you can read it out, it is too much text. Slides carry evidence — a chart, a number, an image, three words. Sentences belong in your mouth.
- Never read your slides aloud. The audience finishes reading before you finish speaking, and then waits, mildly irritated.
- Text you would need to squint at is text nobody reads. The same contrast and size thresholds that accessibility standards set for the web apply to a projected slide — pale grey on white looks elegant on your laptop and vanishes on a projector in a lit room.
- One chart per slide. Two charts is two points.
- If a slide needs an apology — "sorry, this is a bit busy" — it needs splitting, not an apology.
Presenting data honestly
You will be tempted to make your numbers look better. Some of it is deliberate; most of it is accidental, from copying charts you have seen.
Start bar charts at zero. A y-axis starting at 90 turns a 2% change into a cliff. On a line chart tracking a narrow range, a truncated axis can be legitimate — but label it clearly, and say so out loud.
Say the denominator. "Satisfaction rose 30%" means nothing without the base. From 10 to 13 responses is noise. Give the actual numbers alongside the percentage.
Show the whole period. If you are showing six months because the twelve-month view is unflattering, someone will eventually ask for twelve months, and the question after that will be why you did not show it.
Say what the data cannot tell you. "Tickets fell 20% after the release. I can't separate that from the seasonal drop we see every August — I'd want another month before claiming it." This sounds like weakness. It is the opposite: it is the sentence that makes everything else you said trustworthy, and senior people notice it immediately.
Label the axes, and state the units. Rupees or lakhs. Per month or cumulative. The number of times a room has argued for ten minutes over a mislabelled axis is not small.
Handling questions
Questions are not an attack on your presentation. They are evidence that people were paying attention, and they are usually the most valuable part of the session.
Let them finish. Do not answer the question you think is coming. You will frequently be wrong, and interrupting reads as defensiveness.
Repeat or rephrase it. "So the question is whether the help centre would just repeat 2023 — is that right?" This buys you three seconds, confirms you understood, and lets the rest of the room hear it.
Answer the question asked. Not the adjacent one you prepared for.
Then stop. Over-answering is the most common failure. You answer, feel the silence, and keep talking until you have introduced a doubt nobody had.
"I don't know"
Say it. Then say what you will do.
"I don't know. I'd guess it's around two weeks but I don't want you deciding on a guess — I'll check and send you the number by tomorrow."
That answer costs you nothing. Everyone in the room knows the difference between a confident answer and a correct one, and inventing a number that turns out wrong costs you far more than the moment of not knowing.
Variants worth having ready:
- You partly know: "I know the licensing cost; I don't know the implementation cost. Let me give you the part I'm sure of."
- It is out of your scope: "That's really a question for finance — I'll ask Nitin and come back."
- It challenges your conclusion: "That's a fair challenge. If the seasonal effect explains it, my case is much weaker. Here's why I don't think it does — and here's what would change my mind."
That last one is what confidence actually looks like. Not certainty — knowing what evidence would change your view.
Nerves are a preparation problem
Almost everyone is nervous. What separates people is not calm; it is that practised people are nervous about smaller things.
Untangle what you are actually afraid of, because the fixes are different:
| The fear | The fix |
|---|---|
| Forgetting what to say | Learn the structure, not the words. Three supports on an index card |
| Being asked something you cannot answer | List the five hardest questions and prepare answers. One will come up |
| Freezing at the start | Write and rehearse the first sentence exactly. Only that one |
| The technology failing | Arrive early, test it, and have the deck in a second place |
| Being judged | Rehearse aloud, standing, to one real person |
Rehearsing out loud is the part people skip, and it is the part that works. Reading through it in your head takes four minutes and teaches you nothing. Saying it out loud takes fifteen and reveals every sentence you cannot actually say, every transition that does not exist, and the fact that you are four minutes over.
Two more things that genuinely help. Nerves peak in the first thirty seconds and then fall — knowing this means you can ride out the worst part rather than interpreting it as evidence you are failing. And the audience cannot see your heart rate. What feels like visible panic reads, from the fourth row, as a person talking slightly fast.
Building a fifteen-minute talk
1 of 7Write the index card. Message, ask, three supports, the objection, the opening line. Five minutes, on paper. Do not open any software.
Try this
Your slide is titled "Q3 Support Metrics" and shows a bar chart of ticket volume by category. You are asking for two months of engineering time.
Rewrite the title, and write the sentence you say while it is on screen.
Your challenge
Level 3 · IndependentTake something you would genuinely have to present — a project, a proposal, a recommendation — and build a fifteen-minute version.
Rules:
- The index card exists before any slide does.
- No more than ten slides.
- Every slide title is a claim, not a label.
- Every chart has labelled axes, a stated denominator, and a zero baseline unless you say otherwise out loud.
- You have written and answered your five hardest questions.
- You rehearse it out loud, standing, timed, to one real person.
You have succeeded when your listener can tell you, without prompting, what you were asking for and the three reasons why. If they can only remember that it was "about support", the structure failed — and it failed on the index card, not in the deck.
What people usually get wrong
- Opening the slide software first. You end up with a deck that looks finished and argues nothing.
- Slide titles that are labels. "Overview", "Background", "Results". The audience has to work out your point from the contents.
- Reading your slides. They finish before you do and then wait.
- Truncated y-axes. Sometimes honest, usually not, and always worth flagging out loud.
- Percentages without denominators. A 50% rise in a sample of four is not a finding.
- Over-answering a question. You answer, then keep talking, then create a doubt that was not there.
- Bluffing an answer. Everyone can tell, and being wrong later costs far more than "I'll find out".
- Rehearsing silently. Reading it in your head is not rehearsal. It teaches you nothing about timing or the sentences you cannot say.
- Apologising at the start. "Sorry, I'm not great at presentations" tells people to expect something bad. Skip it — nobody was thinking it.
How someone experienced does it
Experienced presenters prepare two versions of every talk: the full one, and the two-minute one for when the CEO walks in late, the previous session overruns, or someone says "can you give us the short version?" The short version is the message, the ask, and the strongest support. Being able to produce it on demand is a reputation-maker, and it also proves to you that you know your own argument.
They also present to the decision-maker, not to the room. In most rooms one person decides and the rest advise. That person's objection is the one the talk must answer, and their vocabulary is the one to use. This is not sycophancy — it is recognising that a talk which persuades six advisors and not the decider has achieved nothing.
And they treat the questions as the real event. The presentation exists to earn the discussion. So they leave time for it deliberately — a fifteen-minute slot gets a ten-minute talk — and they end not on "any questions?" but on something that opens the discussion they want: "The part I'm least sure about is the seasonal effect. What do you think?" That hands the room a productive question instead of leaving them to find one.
When not to use this
Do not present when:
- A one-page document would do. If there is no decision and no discussion, you are making thirty people sit through something they could read in four minutes.
- You do not have the numbers yet. Presenting a half-formed case burns the audience's willingness to hear the finished one. Say "I'll bring this in two weeks when the data's in."
- The decision has already been made. A presentation framed as consultation when the outcome is fixed is transparent and expensive. Announce it instead, say why, and take questions.
Why dense slides make you redundant
Reading and listening compete for the same processing. When a slide contains sentences, the audience reads them — reading is faster and more automatic than listening — and while they read, you are effectively noise.
This is why the deck that feels safest is the worst one. Dense slides feel protective because they are a script you cannot forget. But every word you add to protect yourself is a word that competes with you for attention, and the result is a room that has read your slides and heard almost nothing you said.
The practical consequence: put on the slide the things that do not compete with speech — a chart, a photograph, a single large number, three words. Keep the sentences in your mouth, where they have your tone, your emphasis and your ability to respond to a face that has gone blank.
Prove it
The index card, the deck of ten or fewer slides, and your five hardest questions with answers.
Then deliver it to one person and ask them a single question afterwards: "What was I asking for?" If they cannot answer in one sentence, the ask was never clear — and that is a five-minute fix on the card, not a rebuild of the deck.
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 structuring a presentation and handling questions. Assume I am intelligent but relatively new to this — treat me as intermediate 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 structuring a presentation and handling questions. 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 presenting an argument to a room — 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.