Your deck lives two lives. On stage you talk over it, so it can be sparse and let you fill the gaps out loud. In an inbox it travels alone, with nobody there to narrate, so it has to explain itself. Most founders build for one life and then use it for the other: they email a lean talking deck that says nothing without them, or stand on stage reading a dense one aloud while the room reads ahead. This chapter shows you how to build a single deck that does both jobs, and how to steer the investor's eye through it so the point you care about is the first thing they see.
One deck, two jobs: spoken over and read alone
Strip the question to its bones and there are two kinds of deck. A talking deck is sparse. Three words and one image on a slide, because you are standing right there to fill in everything the slide leaves out. A reading deck is the opposite. It arrives in an inbox with no one beside it, sits among tens of other decks, and carries the whole argument on its own, in full sentences, because nobody is there to narrate.
Most founders build one and use it for both. They design a dense reading deck, then stand on stage reading it out loud while the room reads ahead of them. Or they build a beautiful talking deck, three words a slide, then email it, and the investor scrolls through forty near-empty slides that say nothing without the person who made them.
The best deck does both at once. Nothing you could remove without losing content, nothing you could add without making it fat. That is a high art, and it is the whole job of this chapter.
Build the story first, the slides second, the polish last
Here is the order we build in, every time, and the order matters more than any single slide.
Story first. Before you open a design tool, write the argument as words. One sentence per slide, the single thing that slide has to say, nothing else. No visuals, no charts, no layout. Read those sentences top to bottom and fix the story at that level, because a broken story is cheap to fix in a text file, and expensive to fix once it is forty designed slides. This is the same move as writing action titles: your slide titles are the story, and if reading them aloud is not your pitch, the story is not finished yet.
Slide order second. Move the sentences around until the argument climbs in the right sequence: the hook, the gap, the game, the proof, the ask. You are ordering claims, not decorating slides. It is still just text, so it is still cheap to change.
Simplify and make it visual last. Only now do you cut every sentence to its shortest honest form and hand each slide one image that carries it: a screenshot, a looping bit of gameplay, one chart, one number. Polish comes at the end because polish on a broken story is wasted work, and you feel the waste the moment you have to tear it up.
People scan your deck, they never read it
Nobody reads a slide the way they read a book, left to right, top to bottom, word by word. They scan. The eye lands on the biggest, boldest, brightest thing first, then the next, then the next, and it gives up long before it reaches the small print. You do not get to choose whether the room scans. You only get to choose what they land on first.
So you build a hierarchy, on purpose, out of four levels. Size, color, contrast, bold against not-bold, an icon against plain text: those are the tools that decide the order the eye travels. Level one is whatever shouts loudest. Level four is the quiet gray line nobody reads until they study the PDF at their desk. Every element on the slide sits somewhere in that order, whether you planned it or not. Planning it is the skill.
The four levels of the eye
| Level | What the eye does | What belongs there |
|---|---|---|
| Level 1 | Lands here first, before you speak | The one claim of the slide, written as your action title. |
| Level 2 | Reads this next, still scanning | The proof that makes level one land: the screenshot, the number, the gameplay loop. |
| Level 3 | Reaches this only if interested | Supporting detail and second-tier evidence, short and calm. |
| Level 4 | Reads this alone at a desk, later | The fine print for the reading deck: sources, caveats, the honest footnote, small and gray. |
Get this right and one deck serves both readers. The person in the back row catches levels one and two without trying. The investor at their desk finds levels three and four when they go looking. Same slide, two reading distances, no second file to maintain.
The main message lives in levels one and two
Two rules follow from the levels, and they fix more ugly slides than any font choice.
First, your main message lives in levels one and two, always. On stage you should be able to say nothing that is not already visible up top, and the room still gets the point. That is what makes it a talking deck. Everything you push down to levels three and four is there for the reader who studies the deck later, not for the listener in the room. If the point of the slide is hiding in a level-three bullet, the back row never sees it and the whole slide misses.
Second, the least important text goes small and gray on white. Not deleted. A reading deck needs its sources, its caveats, its second-tier detail. But third in the eye's path, quiet, out of the way. Gray on white is not a mistake to fix. It is a decision about where something sits in the order.
Two smaller habits carry a lot of weight. Left-align your body text, always: the eye finds the start of every line in the same place, and centered or right-aligned paragraphs make it hunt. And never let a generic label sit in level one. 'How it works', 'Gameplay', 'Features', 'The Problem': those are furniture labels, and they waste the one spot the eye is guaranteed to hit. Level one is for a claim, not a category.
Your first slide is paid ad space
Your first slide does its work before you open your mouth.
On stage it sits projected during the changeover, ten to forty seconds while the host reads your name and you walk to your mark. In an inbox it is the thumbnail, the first thing an investor sees among tens of decks, the image that decides whether they read on or archive you. That is attention you are paying for. Treat it as paid ad space.
So put your brand or logo on it, a short line that sparks curiosity, and sometimes the ask. That is the list. What does not belong is the phrase 'Pitch Deck'. Nobody ever leaned in because a slide announced it was a pitch deck, and it sits on half the decks we see. Think of the first slide the way you think of a YouTube thumbnail: its only job is to set the mode the investor reads everything else in, to make them curious enough to press play.
One more thing that does not belong on slide one: a QR code. Nobody scans a code that flashes past in a ten-second window, and a phone raised to snap it is a listener you just lost. The QR earns its place on the last slide, the one that stays up through the entire Q and A, where people finally have the minutes to point a camera at it.
A looping GIF beats video, and beats a live demo
If you show one moving thing in a short pitch, make it a GIF. A short loop of real gameplay, three to seven seconds, running silently on the slide while you talk over it. I have never seen a GIF fail. I have seen plenty of everything else fail.
A video eats the one resource you cannot get back on stage: time. Thirty seconds of trailer is thirty seconds you are not talking, and it drags in sound levels, playback, and the black pause while it buffers, every one of them a way to lose the room. A live demo is worse. It fails far more often than a video, and it fails in the one place you cannot afford it, live, in front of the person deciding whether to fund you. Builds crash, controllers unpair, the wifi dies. Always at the worst moment.
The GIF sidesteps all of it. Cut the best three seconds out of your trailer, loop it, and talk over the top. The loop shows the feel of the game while your voice carries the point, and nothing about it can break on stage. Save the full trailer and the live demo for the sit-down later, where you have the time and a second screen.
Put it together: one deck that does both jobs
Here is the whole chapter as one pass over your deck.
The two-deck pass
- Story first: one sentence per slide, in order, before you touch a design tool.
- Read the slide titles alone. If that is not the pitch, the story is not fixed yet.
- Every slide has one level-one claim, never a category label like 'Gameplay' or 'How it works'.
- The message lives in levels one and two, so a listener needs nothing else read aloud.
- Detail, sources and caveats go small and gray, left-aligned, for the reader at their desk.
- First slide is an ad: brand, a curious line, no 'Pitch Deck', no QR code.
- Last slide holds the QR and stays up through the whole Q and A.
- Any motion is a three-to-seven-second gameplay GIF, with a flat screenshot as backup.
Run a deck through that pass and it does both jobs at once: it carries you on stage, and it carries itself in an inbox. That is the high art the whole chapter is chasing, and it is the difference between a deck the investor sits through and a deck the investor forwards to a partner. Build the story, steer the eye, and let the first slide and the loop do the talking before you say a word.