Why Do Beautiful Slides Sabotage Your Best Ideas?

The first time my design team made me sit through a critique of my own slides, I almost wished I hadn’t asked. Slide after slide, they pointed at things I had been showing in workshops for years and asked one question: “What’s the main character on this slide?”

Half the time, my answer was the concept I was trying to teach. Their answer was the cool image I had picked to support it. We were both right, and that is exactly the problem. The cool image WAS the main character. My framework was the supporting character. Audiences had been remembering Gandalf and forgetting Octalysis for years, and I had no idea.

If you have ever finished a presentation and felt like the audience laughed at the right jokes but missed the actual point, this post is for you. Because there is a quiet, ruthless behavioral design problem hiding in your slide deck, and once you see it, you cannot unsee it.

⚡ Speed Run Notes

• Every slide has a main character: the ONE thing the audience walks away remembering. If it is not your concept, your slide is sabotaging you.
• Slides do not fail by being ugly. They fail by being beautiful in the wrong place. A high-resolution Gandalf will steal the show from a tiny Octalysis diagram every single time.
• There are five repeatable failure patterns: Main Character Theft, Story-Visual Mismatch, the Invisible Arrow, Cognitive Alignment Violation, and Foreign Language Screenshots. Every messy deck I have audited has at least three.
• For spectrum slides, the gradient IS the main character. If readers see two boxes instead of a continuum, the slide failed before you opened your mouth.
• The same content needs different slides for different goals. Lead-gen decks should leave gaps. Client decks should feel complete. Consulting decks should manage expectations.
• Slide design is behavioral design. You are not decorating screens. You are engineering what gets punched into a human brain in the next three seconds.

Table of Contents

About the Creator of the Octalysis Framework

Yu-kai Chou is an S-Tier Behavioral Designer and the creator of the Octalysis Framework, the gamification design system now applied to products and experiences reaching over 1.5 billion users. His book Actionable Gamification is one of the most-cited works in the field, and he has been ranked the #1 Gamification Guru in the World.

He has advised MrBeast, LEGO, Microsoft, Porsche, Tesla, Stanford, Harvard, and governments including Ukraine on turning behavioral psychology into product mechanics that actually change user behavior.

Verify: Wikipedia · Google Scholar · Wikidata · LinkedIn

Why I am the one writing this: I have given some version of the Octalysis workshop hundreds of times, at Stanford, at Yale, at Tesla, at IDEO, in front of MrBeast’s team, in front of executive groups at LEGO and Coca-Cola. Every one of those decks went through internal slide critiques where my designers would freeze on a slide and ask, “Yu-kai, what is the main character here?” The Main Character Principle in this article is not a theory I read in a book. It is the diagnostic my own team developed to stop my own slides from sabotaging my own ideas.

The Main Character Test (And Why Most Slides Fail It)

Here is the test. Pull up any slide you have used in the last six months. Look at it for three seconds. Then look away. What do you remember?

That thing you remembered. That is the main character of your slide. If it matches the concept you were trying to teach, your slide is doing its job. If it does not, your slide has been pulling against you the entire time you have been presenting it.

Most presenters never run this test, because most presenters design slides by asking the wrong question. They ask, “What do I want to say on this slide?” The right question is, “What will the audience’s brain see first?”

Those are not the same question. The first one is about the speaker’s intent. The second one is about how the human visual cortex actually allocates attention, and the visual cortex does not care what you wanted to say. It cares about what is biggest, brightest, most colorful, most familiar, most face-shaped. Whichever element on your slide wins those scoring categories wins the audience. The rest is decoration.

This is why behavioral design applies to slide decks the same way it applies to product onboarding flows or loyalty programs. You are not making art. You are engineering attention. And as in any engineering problem, the failure modes are predictable once you name them.

Here are five I see in nearly every deck I audit, including my own.

Failure Pattern #1: The Gandalf Problem

Years ago I had a slide explaining White Hat versus Black Hat motivation. I wanted to use a visual that signaled the moral spectrum, so I picked a beautiful, high-resolution image of Gandalf the Grey. He took up most of the slide. Off to one side, I had placed a small Octalysis octagon, the actual concept I was teaching.

My team watched me run the slide in a workshop one day, then cornered me afterward. They asked the question I now ask myself before every deck: “What’s the main character on this slide?”

“Obviously it looks like the main character is Gandalf, and Octalysis is some kind of little spell he’s casting.”

That is the Gandalf Problem. A high-quality image, perfectly chosen for atmosphere, becomes the thing the audience remembers. And the framework you spent five minutes explaining becomes a footnote in someone else’s story. It is not that the image is bad. The image is simply too good. It outcompetes the concept for attention, and the visual cortex picks the winner regardless of what your mouth was saying.

The fix is not to remove cool images. The fix is to demote them — visually and structurally — until they are clearly serving the concept rather than replacing it. We rebuilt those slides with simpler LEGO-style visuals: characters that supported the framework without dominating it. The information density of each slide went down. The retention rate of the actual concept went up.

The workshop diagnostic my team eventually built has three checks. Scale check: the supporting character has to be visually smaller than the framework element it is supporting. If Gandalf is 60% of the canvas and Octalysis is 15%, Gandalf wins. Flip those ratios. Frame check: the image should function as a frame around the concept, not the subject of the painting. Ask: if I crop the image out, is the slide still teaching the concept? If yes, the image is correctly a frame. If no, the image was loading the concept on its back, and the concept will fall off the slide the moment the image fades from memory. Core-Drive check: a striking image activates Core Drive 7: Unpredictability and Curiosity in the audience. Aim that curiosity at the framework, not at the image itself. The Gandalf slide failed this test because the curiosity activated by the image was about Gandalf (why is he glowing? what is he holding?) instead of about the framework. The cure was a visual that left the audience curious about why the Octalysis octagon was structured the way it was, with the image as supporting atmosphere rather than the puzzle.

Ask yourself: on every slide where you have placed a striking image, is the image serving as a supporting character to your concept, or has it quietly stolen the lead role?

Failure Pattern #2: The Roller Coaster Problem

I have a story I tell about theme parks. I love the question of why Disneyland can charge people money to wait in lines for hours and have them rate it as a great day. The answer is a behavioral design lesson about Core Drive 7: Unpredictability and Curiosity, plus Core Drive 5: Social Influence. My narrative beat in that story is the waiting, not the riding. That wait is where the magic lives.

For years, the slide that backed this story showed a roller coaster mid-ride. People were screaming with joy on the loop. Beautiful image. Wrong moment.

The story is about the line. The slide jumped to the payoff. By the time my mouth got to the line-waiting argument, the audience’s brain had already cashed in on the rollercoaster image and decided the slide was about thrills. The actual point, the patient and intentional design of the queue, never landed because the visual had already told them the topic was something else.

This is the Roller Coaster Problem: a story-visual mismatch where the image represents a different beat in the narrative than the one the speaker is on. It feels harmless because the image is technically on-topic. It is not. On-topic is too coarse a filter. The filter is whether the image matches the specific moment you are speaking to.

The diagnostic: name the exact emotional beat your story is on. Then look at the slide. Is the image showing that beat, or is it showing a different beat from the same story? If it is the wrong beat, you are training the audience’s brain to remember the wrong scene.

Failure Pattern #3: The Invisible Arrow

One of my favorite slides (and one of my biggest design failures) used to show two photographs side by side. On the left, a picture of someone doing soul-crushing data entry at a desk. On the right, a kid lit up with joy playing a video game. Between them, an arrow that read “work → game.”

The point of the slide was the arrow. The arrow IS the insight. The transformation from the left image to the right image is the entire reason gamification exists as a discipline.

I ran that slide in workshops for years before catching what was wrong with it. My designers caught it first.

“I have been through this slide a dozen times and I realized I never even noticed there was an arrow here. The arrow is the transition. That arrow is the whole point of the slide.”

Both photos were huge, vibrant, emotionally charged. The arrow was small, gray, polite. The audience’s brain spent its attention budget on the photos and had nothing left for the arrow. They saw “boring work” and “fun game” as two unrelated images on a slide. They did not see a transformation, because nothing on the slide was visually screaming transformation.

This is the Invisible Arrow problem, and it generalizes. Whenever the main character of a slide is a relationship between things (a transformation, a spectrum, a comparison, a causal link), the visual element representing that relationship has to be the dominant element on the slide. If it is not, the relationship is invisible, and the slide reduces in the audience’s brain to a collage of the things it was supposed to relate.

Make the arrow huge. Make it the loudest element on the slide. If the arrow is the insight, the arrow gets the lead role.

Failure Pattern #4: Cognitive Alignment Violation

This one is subtle. Most presenters never even know they are doing it. It cost me years of fuzzy workshop comprehension before I learned to fix it.

The Octalysis Framework has a Left Brain side (extrinsic motivators, logic, ownership, accomplishment) and a Right Brain side (intrinsic motivators, creativity, social, meaning). When I am explaining the Left Brain side, I am explicitly telling the audience that this is the “left” half of human motivation.

So what should the slide do? Every visual element representing that concept (the framework callout, supporting images, sample text, highlighted Core Drives) should physically sit on the LEFT side of the slide. When I shift to talking about Right Brain, everything should slide over to the RIGHT.

I used to violate this constantly. The framework callout would say “LEFT BRAIN” but the supporting image would float over on the right side because that was where it looked compositionally balanced. The audience would feel something off, a vague unease they could not name, and the abstract concept would land less cleanly. They never said, “Yu-kai, your visuals are spatially misaligned with your verbal concept.” They just left the workshop with a fuzzier grasp of the framework than I had hoped for, and I blamed myself for not explaining it clearly enough.

The lesson: when your concept has a spatial metaphor (left/right, up/down, inside/outside, before/after), your slide elements have to honor that metaphor physically. Spatial alignment is a form of implicit behavioral design applied to your own audience’s perception. Get it right and the concept clicks subconsciously. Get it wrong and the audience fights an invisible drag they cannot articulate.

The before/after axis is where this fails most often in pitch decks. A team will tell a “before our product, customer churn was 30%; after, churn dropped to 8%” story while their slide places the 30% number on the right side of the screen and the 8% number on the left. The audience reads left-to-right (in left-to-right scripts), expects time to flow left-to-right, and gets a quiet wrongness signal they cannot name. Move the 30% to the left, the 8% to the right, add a visible arrow in the middle, and the same content suddenly feels like an obvious win. The data did not change. The spatial metaphor finally agreed with the narrative.

Failure Pattern #5: Foreign Language Screenshots

This one looks like a small thing. It is not.

I work globally. Many of the products I have advised on were originally launched in markets where the UI is in a language my workshop audience does not read: Mandarin, Korean, Japanese, or Russian. When I want to show a real product example, the most up-to-date screenshot is often the one from the local-language version.

For a long time, my instinct was: use the newest screenshot, even if the UI text is in another language. Newer = better, right?

Wrong. Every time the audience encounters a UI element they cannot read, their attention forks. Part of their brain stays with my voice and the concept. Part of it gets sucked into trying to parse unfamiliar characters, even though they know parsing is hopeless. That fork is cognitive friction, and cognitive friction is enemy of comprehension.

The fix is to use an older screenshot in a language the audience reads, even if the UI has since been redesigned. The visual fidelity matters less than the cognitive ease. An audience that is not silently fighting a foreign-language screenshot has more attention available for the actual concept I am teaching.

This is a small lesson with a big implication. Anytime your slide forces the audience to expend mental effort on something other than your main character, your slide is taxing the same attention budget you needed for the main character. Foreign-language UI is one example. Dense data tables, illegible captions, unexplained acronyms, and visually busy backgrounds are all variants of the same problem.

The Spectrum Rule: When the Gradient IS the Star

One of the most dangerous slide patterns in my deck used to be the spectrum slide. I would explain Implicit versus Explicit Gamification, a continuum and not a binary, and the slide would show two boxes labeled “Implicit” and “Explicit” with a list of examples under each.

The audience would walk away believing in two categories. They would ask questions like, “Is this product implicit or explicit?” The premise of the question proved the slide had failed. The framework was supposed to teach them that products live somewhere on a continuum, with most real-world systems blending both. The slide had taught them the opposite by visually separating the two ends into discrete containers.

This is the Spectrum Rule. Whenever your concept is a gradient, the gradient itself has to be the loudest, most dominant element on the slide. Not labels. Not example lists. The arrow with smooth color transition from one end to the other. A bar with intermediate gradations. Some continuous shape that visually screams “this is a range, not two categories.”

If a viewer who saw your slide for two seconds and then looked away would describe what they saw as “two boxes,” the slide failed. If they would describe it as “a spectrum from X to Y,” the slide is doing its job.

The Spectrum Rule is the Main Character Test applied to a specific concept type. In every case, the relationship itself is the main character, and that relationship needs the visual real estate.

Workshop Goal Changes Everything

Here is the part that breaks most presenters: the same content needs different slide design depending on what you want the audience to do after the workshop ends.

I run three distinct modes of workshop, each with a different design objective for the slides:

Public exposure and lead-gen. If I am speaking at a public conference and I want people to feel they need to learn more (to subscribe, buy a book, sign up for Octalysis Prime), the slides should leave strategic gaps. I am intentionally activating Core Drive 7: Unpredictability and Curiosity. The slide hints at depth without delivering it. The audience leaves slightly hungry, with the felt sense that there is more to know. That hunger is the conversion mechanism.

High-ticket client workshop. If a company has paid serious money for a private workshop, the slide design objective inverts. They want to feel they got every dollar’s worth. Every framework comes with the full diagram. Every concept is made comprehensive on the slide itself. The activation is Core Drive 2: Development and Accomplishment. The audience walks away feeling they mastered the framework, that the workshop was worth the spend, and that they have something concrete to bring back to their team.

Workshop attached to a consulting project. When the workshop is the kickoff of a paid engagement, the slide design has a third objective: managing executive expectations. The slides need to telegraph rigor, methodology, and prior client success. The activation is Core Drive 5: Social Influence and Relatedness, through visible client logos and named applications, plus Core Drive 4: Ownership and Possession, by getting the audience to mentally claim the framework as theirs by the end of the day.

Same content. Three different slide design briefs. The main character is the audience action you want, and the visual elements adjust accordingly.

Most presenters use the same deck for all three modes. Then they wonder why their lead conversion is low at conferences and their executive clients feel underserved in private workshops. The answer is upstream of the slides. They never asked which audience action the deck was supposed to be optimizing for.

The Universal Principle: Beyond Slides

The Main Character Principle is not really about slides. It is about attention engineering, and attention engineering is the underlying problem behind every interface humans interact with.

Think about it this way. Every screen, every webpage, every mobile app onboarding flow, every email, every dashboard, every checkout page, every notification: each has a main character problem. There is exactly one thing that screen exists to make the user understand or do. There are usually a hundred things competing for their attention.

The diagnostic translates one-to-one:

1. What do I want them to see? (Define the main character.)
2. What will they actually see? (Audit visual dominance: what is the biggest, brightest, most face-shaped element on the screen?)
3. Do those two answers match? (If not, redesign until they do.)

I have used this exact diagnostic on product onboarding flows for clients with hundreds of millions of users. It works the same way for a dashboard at a SaaS company that nobody uses, an email campaign with a 0.4 percent click-through rate, a landing page that converts at 1 percent when it should convert at 5 percent. The pattern is always the same. The team has built something where the visual main character is not the action they wanted the user to take.

I covered the deeper version of this attention-engineering thinking in Actionable Gamification, but you do not need the book to apply the test. You can run it on the slide you are presenting tomorrow morning. Run it on the email you are about to send. Try it on the homepage of any product you have shipped.

If the main character of the screen is not the thing you needed the human to do, your design is making a worse version of itself for free. The audience does not know your intent. They only know what their visual cortex tells them is important, and the visual cortex is voting with what is biggest and brightest, not with what your brief said the slide was about.

I have a phrase I have used with my design team for years about this: you are punching ideas into people’s brains, and you have to be careful what gets punched in. That is what slide design is. Not decoration. Not aesthetics. Punching the right thing into a brain that is going to forget 90 percent of what you said before lunch.

The Main Character Principle is the diagnostic for figuring out what is actually getting punched in.

Frequently Asked Questions

How do I run the Main Character Test on my own slides without my own bias getting in the way?

You cannot. That is the trap. You designed the slide knowing what you meant it to say, so when you look at it your brain fills in the meaning that is not visually present. The fix is to ask someone who has not seen the slide before (a colleague, a junior designer, anyone) to look at it for three seconds and tell you what they remember. Their answer is the truth. Yours is wishful thinking.

Is the Main Character Principle the same as “less is more” minimalism?

No. Minimalism is a style. The Main Character Principle is a diagnostic. A maximalist slide can be perfectly aligned to its main character if every busy element supports the same lead role. A minimalist slide can fail the test if the one element on it is the wrong element. Style is not the question. Visual dominance is.

What if my concept actually needs multiple main characters on one slide?

Then it needs multiple slides. The brain only allocates lead-character status to one element per scene. If you cram two concepts onto one slide, the audience will pick one and forget the other, and you do not get to choose which one. Splitting into two slides is almost always the right answer.

Does this principle apply to data visualizations and charts?

Yes, even more so. The whole point of a chart is to make one specific takeaway visually obvious. If the audience looks at your chart and asks “so what?” you have left them to do work that the chart itself was supposed to do. The main character of a chart is the conclusion the data supports. If the conclusion is not the visually dominant element of the chart, the chart is failing its job.

How does this connect to the Octalysis Framework?

Slide design is one of many surfaces where you are doing behavioral design on humans. Every Core Drive can be activated through visual emphasis on slides: Core Drive 1 (Epic Meaning) through hero imagery, Core Drive 6 (Scarcity) through prominent countdown elements, Core Drive 8 (Loss Avoidance) through visually weighted warnings. The Main Character Principle is how you choose which Core Drive your slide is engineering for, instead of accidentally activating five at once and getting none of them to land.

Apply the Main Character Principle to Your Own Slides

If this diagnostic resonated, here are three ways to go deeper, each matching the workshop mode you are in:

• You are presenting tomorrow morning. Open your deck. Run the three-second test on every slide. The slides that fail the test are the ones to fix tonight. No tools needed beyond what is on this page.
• You want the full Octalysis lens behind the diagnostic. Actionable Gamification is the book where I lay out the 8 Core Drives the Main Character Principle is engineering for. Slide design is one surface; the same diagnostic generalises to onboarding, email, dashboards, and entire products.
• You want to apply this to a real product or workshop. The Octalysis Prime community is where my deeper teachings live — including the slide-design walkthroughs I do not put on the public blog. For high-stakes workshops or pitch decks where the conversion math actually matters, the Octalysis Group is the consultancy I built specifically for this work.

The Main Character Principle is one diagnostic. Octalysis is the framework underneath all of them. Pick the door that matches where you are.

Related Reading

• The Octalysis Framework: A Complete Gamification System
• Core Drive 3: The Golden Core Drive of Creative Expression
• Core Drive 7: Unpredictability and Curiosity
• Actionable Gamification: The Book

WOULD YOU LIKE YU-KAI CHOU TO WORK WITH YOUR ORGANIZATION?