The Octalysis Brainstorming Engine: How to Generate 100+ Engagement Ideas in One Session

Octalysis brainstorming sessions don’t look like normal brainstorming sessions. I learned this the hard way when I sat in a room with senior executives at a Fortune 500 company who spent two hours generating four ideas. Points. Badges. Leaderboard. And a progress bar.

Someone called the progress bar “innovative.” Everyone nodded. The meeting ended.

Six months later, they called me because engagement hadn’t moved. Of course it hadn’t. Those four ideas had nothing to do with what their users actually wanted. They were disconnected from the psychology of the people using the product, disconnected from what phase of the experience those people were in, and disconnected from each other. Four isolated features floating in space.

Speed Run Notes

The structure: 8 Core Drives × 4 Experience Phases = 32 components. Generate 3-5 ideas per component. In practice, that often gets you past 100 ideas.

The phases: Discovery → Onboarding → Scaffolding → Endgame. Different motivations apply at each.

Color code as you go: Green (strong), Yellow (consider), Red (weak but keep for inspiration).

The selection chain: Player Types → Core Drives → Game Techniques. Never reverse it.

The Weaving Principle: Connect ideas across Core Drives into cohesive game loops. Interdependent mechanics beat isolated features.

Present wisely: Show only green favorites (15-25 ideas). Give explicit permission to reject. Score survivors by Power × Ease.

The mindset: Structure is the scaffold. Empathy is the craft. If you can’t picture a specific person having a specific emotion, the idea isn’t real yet.

In my consulting work, this structure often generates 100 or more ideas in a single session. Not because my clients are smarter. It’s because the structure forces you to look at the product from 32 different motivational angles, and when you do that, the ideas almost generate themselves.

This is Step 2 of the Octalysis Design Process, and I consider it the most creatively demanding step of the entire framework.

Why Most Brainstorming Produces Garbage

Here’s what nobody tells you about traditional brainstorming: it has three structural flaws built into the format itself, and no amount of enthusiasm or sticky notes can fix them.

Flaw 1: Ideas in a Vacuum. “What would be cool?” is the default brainstorming prompt, and it’s a terrible one. “Cool” is subjective, and more importantly, it’s disconnected from user psychology.

A leaderboard is cool. For achievers. For the socializer who just wants to get along with her coworkers and not make enemies, a leaderboard is a source of daily anxiety. Same feature, opposite emotional reaction, depending entirely on who your user is.

When you brainstorm in a vacuum, you generate ideas that sound exciting in the conference room but create the wrong emotions for the people who actually have to use them.

Flaw 2: Phase Blindness. What motivates someone to discover your product for the first time is completely different from what keeps a power user engaged on day 300.

A new user needs to feel “this is easy and I’m not going to embarrass myself” (Core Drive 2: Development & Accomplishment in its competence form). A veteran user needs to feel “there’s still something here I haven’t mastered” (Core Drive 7: Unpredictability & Curiosity, or Core Drive 3: Empowerment of Creativity & Feedback).

Traditional brainstorming treats these phases as identical. It generates one pile of ideas and assumes they’ll work everywhere. They won’t. An idea that’s brilliant for onboarding might be patronizing for power users. An endgame mechanic dropped into the first-time experience will confuse people and drive them away.

Flaw 3: The Loudest Voice Wins. The executive in the room loves competition because he’s an achiever. That’s how he became an executive. So the final list of “brainstormed” ideas always includes a leaderboard, always includes status badges, and always reflects the motivational profile of whoever had the most authority in the room.

I call this Designing for Yourself Instead of the Player, and it’s the single most common reason gamification projects fail in corporate environments. The people making the design decisions are almost never representative of the people using the product.

The 32-Component Matrix: 8 Core Drives × 4 Experience Phases

The Octalysis Brainstorming Engine solves all three flaws with one structural constraint.

8 Core Drives × 4 Experience Phases = 32 brainstorming components.

Each component gets 3 to 5 ideas. Do the math: that’s 96 to 160 ideas at the minimum, and once the room finds its rhythm, it’s common to push well past 100.

The four Experience Phases come from the player journey I map in every consulting engagement:

Discovery: how users first encounter and decide to try your product. The psychology here is all first impressions: curiosity, social proof, urgency, and the perception that this thing is worth their time.

Onboarding is the first real experience, where users learn how things work. The psychology shifts to competence, small wins, guided exploration, and the feeling that “I can do this.”

Scaffolding is the regular ongoing use where habits form and users develop routines. This is where most engagement lives and where most products fail. The psychology needs variety, social dynamics, ownership, and creative expression to prevent monotony.

Endgame: the experience of veteran users who’ve seen everything your product offers. The hardest phase to design for. These users need new challenges, mentorship opportunities, exclusive access, and reasons to care when the novelty is gone.

The process works like this: I start with the Discovery Phase desired actions from the Strategy Dashboard (Step 1 of the Octalysis Design Process). Then I walk through all 8 Core Drives asking one question per drive: “How could this Core Drive motivate the desired action?”

How could Core Drive 1: Epic Meaning & Calling motivate a user to sign up? Maybe the signup process tells them they’re joining a community that’s changing their industry.

How could Core Drive 2: Development & Accomplishment? Maybe there’s a quick assessment that immediately shows them where they stand and what they could achieve.

How could Core Drive 7: Unpredictability & Curiosity? Maybe the signup page teases a result they’ll only see after completing the process.

I write each idea around the Octalysis octagon diagram, with the desired actions written at the top. After Discovery, I move to Onboarding and repeat the full sweep. Scaffolding follows. Endgame comes last.

As I go, I color-code in real time. Green for ideas I love. Yellow for ideas worth considering. Red for weak ideas that will probably get cut.

The color-coding matters more than you’d expect. It creates a visual heat map of where the strongest engagement opportunities cluster. After running through all 32 components, you can literally see which Core Drives and which Experience Phases have the most green. That tells you where the real design energy should go.

Something shifts around the 80th idea. You start noticing connections. The onboarding mechanism you wrote under Core Drive 4 (Ownership & Possession) would feed perfectly into the scaffolding mechanic you wrote under Core Drive 7 (Unpredictability & Curiosity). A game loop starts forming in your head, even though nobody asked for one yet.

That mental model becomes the foundation for Step 4 of the design process (the Battleplan), but it emerges organically from the brainstorming structure itself. The structure generates coherence. Freeform brainstorming never does.

The Five Deadly Brainstorming Sins

After years of running these sessions with clients across dozens of industries, I’ve cataloged five mistakes that consistently produce weak output. They all look perfectly reasonable in the moment, which is exactly what makes them dangerous.

Sin 1: Siloed Thinking

This is the most common failure mode, and it looks like this on the whiteboard:

“Core Drive 3 = scavenger hunt. Core Drive 4 = collecting stuff. Core Drive 5 = teamwork.”

Three separate features. Three separate development sprints. Three separate user experiences that share a product but don’t share a purpose.

Now compare that to this:

“Core Drive 3 = scavenger hunt. The scavenger hunt items become a collectible set (Core Drive 4). Completing the set unlocks a mystery box (Core Drive 7). The scavenger hunt is done in teams of four competing against other teams (Core Drive 5).”

That’s one cohesive experience built from four different motivational drivers. Each mechanic pre-loads the next. The scavenger hunt creates the items. The items create the collection. The collection creates the mystery. The mystery creates the team competition. You can trace a single thread of engagement through the entire loop.

The best brainstorm outputs aren’t lists of features. They’re the DNA of a complete experience.

The clearest commercial proof of this in recent years is Balatro. A small indie game that became a 2024 phenomenon precisely because every joker, every voucher, every planet card is woven into the others. Core Drive 7 (Unpredictability, what hand will I draw?) feeds Core Drive 4 (Collection, which jokers do I own this run?) feeds Core Drive 2 (Mastery, what synergies have I learned?) feeds Core Drive 3 (Creativity, how do I sequence them?). Pull any one mechanic out and the whole thing collapses. That interlocking is what made it a 5-million-copy success on a budget that would barely fund a single Coca-Cola TV spot. The weaving was the product.

Sin 2: Checklist Accounting

Some people follow the 32-component structure robotically. They fill in every box, check every item, and at the end they have 32 boxes with ideas in them. Mission accomplished.

Except the ideas are lifeless. They were generated to fill boxes, not to move people.

The structure is a scaffold, not a factory. Every idea you write should make you viscerally imagine a real person reacting to it. Can you picture the moment of surprise when they open the mystery box? Can you feel the satisfaction when they complete the collection? Can you sense the group energy when a team is racing against another team?

If you can’t picture a specific human having a specific emotion, the idea isn’t real yet. It’s just words in a box.

Sin 3: Generic Game Technique Name-Dropping

Someone writes “Easter Egg” in the Core Drive 7 box and moves on.

That’s not an idea. That’s a category label. What kind of Easter Egg? Where does the user find it? What happens when they find it? How does it connect to the desired action? What makes this Easter Egg different from every other Easter Egg in every other product?

Every experience is different. The rule I teach my certification students: the game-technique name belongs at the end of the idea, not the front. “When users complete their daily check-in, they see a closed treasure chest on the dashboard that opens with an animation and reveals one of five collectibles toward a themed set. (Easter Egg / Mystery Box)” That works. “Easter Egg” alone is a vocabulary entry, not a design idea. The label is a tag for later reference. The experience description is the actual work.

Sin 4: Writing What Already Exists

This one sneaks in constantly. Someone writes, “Users are motivated by social proof because they can see how many people already signed up.”

That’s an observation about the current state of the product. It’s not a new engagement idea. The brainstorm is about generating net-new mechanics, not documenting why things currently work. If your idea describes something the product already does, cross it out and try again.

Sin 5: Designing for Yourself Instead of the Player

I saved this one for last because it’s the most insidious, especially in corporate settings.

Executives almost always push for leaderboards and competition. They’re achievers. Their entire career was built on outperforming peers. Competition is how they see the world.

But their employees might be socializers who want harmony, belonging, and cooperative success. Their customers might be explorers who want autonomy and self-directed discovery. Their students might be people who just don’t want to feel stupid.

Competition systems imposed on socializers create stress, increase turnover, and produce the opposite of engagement. I’ve seen it happen with multiple clients. Executives love the leaderboard. Employees hate it. And neither side can understand the other because they’re projecting their own motivational profile onto people who don’t share it.

The brainstorm must be grounded in the player type octagon from the Strategy Dashboard, not the preferences of whoever is sitting in the room.

Player-Type-Aware Brainstorming

Before generating a single idea, I pull up the Strategy Dashboard’s player type octagon. This octagon shows which Core Drives are strongest for the primary user persona we defined in Step 1.

If Player Type #1 is mostly motivated by Core Drive 2, Core Drive 5, and Core Drive 7, I weight my brainstorming heavily toward those three drives. Not evenly across all 8. The matrix has 8 columns, but that doesn’t mean every column deserves equal creative energy.

Here’s where a subtle distinction trips up even experienced Octalysis practitioners.

Some users want to feel capable. They don’t need to be exceptional or champion-level. They just want to feel like they’re not incompetent, not lost, not embarrassed in front of peers. I call these competence-oriented users, and they’re far more common than the achievement-oriented users most designers default to.

The competence-oriented user doesn’t care about being in 1st place. What matters is not being in last place. A trophy means nothing to her. What she wants is confirmation that she’s doing it right.

The Core Drive is technically the same, Core Drive 2: Development & Accomplishment. But the game techniques are opposite.

For achievers: rankings, public status, competitive milestones, “Top 10%” badges.

For competence-seekers: skill-based feedback (“You handled that correctly”), private progress indicators, mastery confirmations that only the user sees, gentle guidance that never implies they’re falling behind.

Player type awareness isn’t just about which drives to emphasize. It’s about how to express those drives for the specific humans you’re designing for. The brainstorm surface-level question is “what motivates them?” A better question is “what does that motivation feel like from the inside?”

The Game Technique Selection Chain

There’s a correct order for choosing game techniques, and almost everyone does it backwards.

The correct chain: Player Types → Core Drives → Game Techniques.

The mistake I see constantly: someone browses my list of 200+ game techniques, picks the ones that sound cool, and tries to bolt them onto their product.

This is like browsing a pharmacy, picking medications with the best packaging, and taking them all at once. Medication works when it’s prescribed for a specific diagnosis. Game techniques work when they’re prescribed for a specific motivational problem.

What the actual process looks like:

First: who are the players? What’s their motivational profile?

Second: what’s the motivational problem? Is the experience monotonous? Is there no sense of progress? No social connection? No feeling of ownership?

Third: which Core Drive solves that problem? Monotony maps to Core Drive 7: Unpredictability & Curiosity. Lack of belonging maps to Core Drive 5: Social Influence & Relatedness. No sense of progress maps to Core Drive 2: Development & Accomplishment.

Fourth: what specific technique fulfills that Core Drive in this context, for these users?

You don’t need to memorize all 200+ game techniques. I don’t expect my students to do that. What I expect is that you understand the Core Drives well enough to invent new techniques on the spot when the situation calls for it.

Most of my client work involves creating new hybrid techniques, not pulling established ones off a list. The techniques are examples of how Core Drives manifest in design. They’re not the menu. The Core Drives are the menu. The techniques are just dishes other chefs have already cooked.

The Weaving Principle: Where Ideas Become Game Loops

This is the part most people miss, and it’s the part that separates a brainstorm that produces a feature list from a brainstorm that produces a living system.

The genius of the 32-component matrix isn’t that it generates more ideas. Any brainstorming method can generate volume if you lower your standards enough. The genius is that it generates connected ideas.

When you walk through the same desired action through the lens of all 8 Core Drives, you naturally discover how mechanics can interlock. The scavenger hunt (Core Drive 3) could feed items into a collection (Core Drive 4). Completing the collection could trigger a mystery reward (Core Drive 7). The hunt could be organized as team-based competition (Core Drive 5), and top-performing teams could get early access to the next hunt (Core Drive 6: Scarcity & Impatience).

Five Core Drives. One cohesive experience. And it wasn’t designed top-down. It emerged from the structure of the brainstorm itself.

This is what I mean when I tell my students that a game loop should be “forming in your head” by the 80th idea. The structure creates coherence that freeform brainstorming can never achieve because freeform brainstorming doesn’t force you to examine the same user action from 8 different motivational perspectives.

The Weaving Principle is the difference between a product that has gamification features bolted onto it and a product where the engagement is structural. In the bolted-on version, you can remove any feature and the rest keeps working. In the woven version, removing one mechanic weakens all the others because they feed each other. That interdependence is what makes engagement durable instead of gimmicky.

The Alfred Effect: Contextual Intelligence in Design

Let me show you how the brainstorm can transcend mechanical ideation with an example I improvised during a teaching session.

The product was a weather app. Boring, right? Everyone thinks weather apps are commodities. There’s nothing to gamify.

Here’s what happened when I ran it through the Octalysis lens.

The app already knows you check the weather every morning. It knows your location. It probably knows your commute schedule if it has calendar permissions. What if it knew you usually go to the gym at 8 AM?

Today the weather is unexpectedly gorgeous. Instead of just showing you 72°F and sunny, the app suggests: “Skip the gym. Hike Mission Peak instead.”

You live in San Francisco. There are probably hundreds of interesting places within an easy drive that you’ve never visited. The weather app becomes a location-based quest system: “Complete the hidden gems in your region.” Each one tagged with the optimal weather conditions for visiting.

I named this the Alfred Effect, after Batman’s butler. Alfred knows Bruce Wayne well enough to anticipate needs Bruce hasn’t articulated yet. A system with genuine contextual intelligence uses data it already has (your location, your habits, your preferences) to create engagement opportunities you didn’t ask for but immediately want.

Pokémon GO is the closest mass-market analog. Niantic’s app uses location, time of day, weather, and your prior catches to surface the next thing it thinks will pull you outside. When it works, the suggestion feels uncanny. When it misfires, it’s just a notification you ignore. The Alfred Effect lives or dies on that hit rate.

The weather data is the hook because people check it daily. But the value is the personalized recommendation that connects weather conditions to lifestyle decisions nobody else is helping you make.

This is where brainstorming becomes art. You’re connecting disparate data points (weather, location, daily habits) into a new kind of experience that none of the individual data points could create alone.

One important caveat, and the bar moved in 2024. ChatGPT’s persistent memory, Gemini’s session-aware context, Spotify’s daylist, and Notion AI’s project-aware suggestions actually do real Alfred-level work now. The technology is no longer the bottleneck. The bottleneck is design judgment: knowing which contextual data to surface, how often to interrupt, and when to stay silent. Plenty of products still ship what I used to call artificial stupidity, guessing constantly, getting it wrong, eroding trust faster than they build engagement. The difference now is that you can no longer blame the model. If your personalized recommendation is wrong more than it’s right in 2026, the issue is that nobody on your team designed for genuine contextual relevance. Alfred would have known when to keep his mouth shut.

How to Present a Wall of Ideas Without Overwhelming Stakeholders

You never show the full wall of ideas to a client. That triggers analysis paralysis and turns a productive meeting into a debate about the weakest ideas on the board.

I show only the green-highlighted favorites. Usually 15 to 25 ideas.

Then I frame the presentation with a specific sentence: “Some of these are meant to be thrown away. If anything goes against your core beliefs, just tell us.”

This framing is deliberate. It gives clients explicit permission to reject ideas without feeling like they’re insulting the work. Paradoxically, this permission makes them far more receptive to the bold ideas: the ones that actually move the needle.

For some clients, I convert the top ideas into a one-idea-per-slide deck. Each slide shows the idea, which Core Drive it activates, which Experience Phase it serves, and which desired action it targets. This format lets non-gamification stakeholders evaluate ideas on merit rather than getting lost in framework terminology they don’t know.

The ideas that survive the presentation feed into Step 3 of the Octalysis Design Process: the PE Feature List. Each surviving idea gets scored by Power (how much engagement impact) multiplied by Ease (how simple to implement). The top 12 to 24 by PE score become the features you actually design and build.

Bad ideas stay on the board, by the way. I don’t erase the red ones during the brainstorm. Sometimes a terrible idea sparks a brilliant one 20 minutes later. The PE Feature List handles the killing. The brainstorm’s job is creation, not curation.

Lightbulbs vs. Red Flags: Scoring Ideas Beyond Color-Coding

Color-coding sorts the wall. It doesn’t score the ideas inside the green pile. For that, I use a method I’ve taught for years but that almost nobody outside Octalysis Prime applies: counting lightbulbs and red flags for each surviving idea.

A lightbulb is what happens on someone’s face when they hear an idea and instantly recognize it as right. The eyebrows lift. There’s a small involuntary “huh.” You can see the recognition before they speak. That reaction is intuition firing, and intuition is the accumulated weight of every product they’ve ever used, every game they’ve ever played, every campaign that worked or didn’t. The logical brain can’t articulate why the idea is good in the moment. The face says it anyway.

A red flag is the inverse. “What if users abuse it?” “What if engineering can’t ship that in time?” “What if it conflicts with the loyalty program?” Each red flag has an answer. That’s the trap. You can resolve any single red flag verbally. What matters is the count.

For each idea in the green pile, I tally:

• Lightbulb count. How many people in the room or in test audiences had the involuntary recognition reaction? Two or three lightbulbs per idea is a strong signal. Four or more is a winner.
• Red flag count. How many distinct concerns came up, regardless of whether someone immediately answered them? An idea with seven red flags that all have plausible answers is still an idea with seven red flags. The volume itself is the warning.

The selection rule: take the ideas with the most lightbulbs and the fewest red flags. Not the ideas with the loudest advocates, not the ideas that scored well on Power × Ease alone, and not the ideas that the senior person liked. Lightbulbs are intuitive resonance. Red flags are accumulated risk. Both are signal the conversational scoring usually buries.

One nuance worth noting: an idea with zero red flags is suspicious. Either you’re flattering yourself, or the idea is so obvious that competitors already shipped it. Real ideas have real concerns. The job of scoring isn’t to find an idea with no concerns. It’s to find the idea where the concern-to-conviction ratio is most lopsided in conviction’s favor.

This isn’t a replacement for the PE Feature List in Step 3. PE handles formal scoring against business metrics. Lightbulbs vs. red flags is the intuitive cut you make before PE — the way you decide which 25 ideas out of 150 deserve the time investment of formal scoring at all.

Frequently Asked Questions

Do I need to memorize all 200+ game techniques before running a brainstorming session?

No, and trying to do so would actually work against you. If you understand the 8 Core Drives at a deep level, not just their names but how they feel from the user’s perspective, you can generate effective techniques on the spot. I’ve run sessions where the best ideas were hybrid techniques that don’t exist in any textbook. The Core Drives are the foundation. The techniques are just examples of how those drives have been expressed before. Know enough to pattern-match, then trust your own creative instincts.

What if my team only generates 30-40 ideas instead of 100+?

Two common causes. First, you’re skipping Experience Phases. Most teams brainstorm heavily for Onboarding, decently for Scaffolding, barely for Discovery, and not at all for Endgame. Go back and check your coverage. Second, you’re committing Sin 4: writing observations about what already exists instead of generating new mechanics. If your idea describes the current product, cross it out. The less obvious combinations (Core Drive 6 for Onboarding, Core Drive 8 for Discovery) often produce the most original work precisely because they force you out of comfortable defaults.

How long does a full 32-component session take?

Three to four hours for a trained facilitator working with an experienced team. A full day for teams new to the method. The first few Core Drives feel slow as people learn the rhythm. By Core Drive 5 or 6, the pace picks up because the thinking pattern has internalized. Don’t rush it. Weak ideas from a rushed brainstorm create real costs downstream: features that get built, launched, ignored, and eventually removed. The extra hours in the brainstorm room are the cheapest part of the entire design process.

Should we brainstorm alone first, then come together, or start as a group?

Wrong question. The right question is which bottleneck you actually have. If you’re worried about not generating enough ideas, run solo first. Five people working independently against the 32-component grid will reliably produce 100 to 160 ideas before anyone says a word out loud — solo work removes the anchoring effect where the first idea spoken sets the frame for everyone else, and it gives quiet team members the airtime they’d never grab in a group. If you’re worried about generating too many ideas to triage (which is the more common problem with experienced Octalysis teams running the higher-detail Octalysis Prime canvas), start as a group. Discussion creates real-time pressure-testing — someone proposes a vague mechanic and another person immediately asks what the user actually sees. Weak ideas die on contact, and the surviving ideas already carry the team’s shared understanding. The hybrid I default to with new clients: 15 minutes solo, then group reconciliation. You get the raw quantity from the solo phase and the quality filter from the group phase. Just don’t let the solo phase run so long that the group session devolves into list-reading.

Can this method work for projects that aren’t about gamification?

The 8 Core Drives describe universal human motivation, not game-specific motivation. I’ve applied this brainstorming method to healthcare onboarding, government policy design, marketing campaigns, educational curricula, and internal employee programs. If your project involves motivating human behavior in any context (and nearly every project does), the 32-component matrix applies. Swap “game techniques” for “engagement mechanics” and the method transfers directly.

How does this brainstorming step connect to the rest of the Octalysis Design Process?

This is Step 2 of 5. Step 1 (Strategy Dashboard) defines your business metrics, player types, desired actions, and feedback mechanics. Step 2 (this method) generates a large pool of ideas, often 100+. Step 3 (PE Feature List) scores ideas by Power × Ease to select the top 12-24. Step 4 (Battleplan) maps those selected features into a cohesive experience flow showing how users move through the game loop. Step 5 (Live Iteration) tests the design with real users and refines based on behavioral data. Each step feeds the next. And critically, the brainstorm output is only as strong as the Strategy Dashboard that precedes it: garbage strategy in, garbage ideas out.

Where to Take This Next

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