"Live 20 years in 20 minutes."
Money20 is an investment simulation game built at HumanX for a major FinTech education client. It doesn't just teach young adults about investing, it forces them to confront their psychological biases in a safe environment. The final product simulates 20 years of investing in a 20-minute session, currently live at humanx.io. As UI Design Intern, I owned research, information architecture, and the design decisions that made the behavioral framing land.
See Money20 Live →Context
It's too theoretical. It teaches the names of concepts without the feel of them. The target demographic, young adults in India, was paralyzed by two forces: the complexity of financial products and the fear of irreversible loss. The brief wasn't to teach what to buy. It was to tackle how people behave when money is on the line, before they had real money on the line.
The problem
Traditional financial education is too theoretical. Users were paralyzed by complexity of products and fear of irreversible loss.
The goal
Build a product that integrates behavioral finance concepts, loss aversion and herd mentality, directly into gameplay.
The business angle
Higher engagement and retention than static platforms. Position the client as an innovator in financial education.
↑ Real consequences felt
High consequence + high engagement. Empty quadrant.
Engagement →
Money20 runs on a tight player loop that compresses 20 years of investing decisions into a 20-minute session.
A year passes every minute. Markets move. Products unlock.
Medical bill. Festive bonus. Job loss. Real-life shocks force decisions.
Sell, hold, buy, follow. The system tracks the pattern.
End-of-session report shows the player their own bias profile.
Solution
Money20 integrates behavioral finance directly into the play loop. Players make 20 years of investing decisions in 20 minutes. The system surfaces their behavioral patterns back to them as they play, so the learning happens through experience rather than after it.
Each game mechanic was designed in service of a specific behavioral bias the research surfaced. Rather than teaching the concepts in lectures, the bias becomes the mechanic itself. Players feel the bias in action, then read about it in the end-of-session reflection.
Behavioral bias
Game mechanic
The result is a game that doesn't teach behavioral finance, it rehearses it. Players make the same mistakes their future selves would make, in a session that costs them nothing and teaches them everything.
Players can compete head-to-head or go solo against the computer. Solo removes performance anxiety. Competitive introduces the herd-mentality variable we wanted them to feel.
Investments unlock over time. Each unlock pairs with a behavioral nudge so players grasp their own risk tolerance, not abstract definitions. Sequenced unlocks let them build literacy without overwhelm.
If a player hasn't made a move for over two minutes, an in-game elephant prompts them with a helper tip. After copy and timing iteration: 20% lift in engagement with the guidance system.
Players see how top performers in their cohort structured their portfolios. Social proof is a real-world driver. Modeling it inside the game lets players feel the bias being formed.
Emergency expenses (medical bills, unemployment) and positive breakers (festive bonus, gifts) keep stakes real. Excluding real-world shocks sanitizes the simulation. The lesson lives in the liquidity scramble.
UI
The game throws a lot of information at every screen. Stock prices, portfolio allocations, real-life event prompts, time pressure. The UI's primary job wasn't to communicate the financial concepts. It was to get out of the way so players could focus on the decisions in front of them.
Players move between four core surfaces during a session: the dashboard, the investment screen, the event modal, and the leaderboard. I structured navigation so each surface owns one job and players never have to remember where to find a control. Persistent timer at the top, persistent portfolio at the bottom, decision content in the middle.
I anchored the palette around a calm neutral base, with a single accent for actionable moments (buy, sell, advance year). Green and red were reserved exclusively for portfolio gains and losses, never for UI states, so colors stayed truthful to the game's signal layer. Players learned that color meant something within seconds of session start.
IBM Plex Sans for narrative text, labels, and CTAs. Its open counters and clean grotesque feel kept the educational moments approachable. For every numeric surface (prices, percentages, time remaining, returns) I switched to IBM Plex Mono. The monospace alignment was functional, prices stayed visually stable as digits changed, and it gave the game its serious-finance signal without feeling like a Bloomberg terminal.
Every component, buttons, modals, cards, tooltips, was designed to feel calm under pressure. Generous padding, subtle elevation, no decorative gradients. The Elephant Hint and the leaderboard were the only moments that allowed playful detail, because those were the moments players were meant to lean in.
Impact
My responsibility didn't end with design. It extended to validating the design worked and influencing what shipped next. I ran unmoderated usability testing on a clickable prototype. The biggest finding was on the Elephant Hint, users didn't understand the motivation behind the prompt. We iterated copy and timing.
The system was right. The execution needed iteration.
I worked closely with the Content Expert to lock down the exact financial product unlock sequence so the UI could reflect it correctly. With the engineering team, I sat through QA on the real-life event modal to make sure the timer pause + resume logic was visually correct in the live build. Money20 shipped to production with HumanX's FinTech edu partner, currently live at humanx.io.
Reflection
Money20 was my first time designing a system where the goal was behavior change, not task completion. The framework that emerged from it has shown up in every project since.
The Elephant Hint felt like a small UX problem until usability testing showed it was the single biggest churn risk. In a behavioral product, the smallest interactions carry the most weight.
Feature priorityThe 20% engagement lift wasn't an insight. It was a copy and timing tweak that came out of testing. The system was right. The execution needed iteration.
ProcessPlayers learned more from a 30-second bad decision than from any tutorial. Designing for productive failure was a new craft I've been refining ever since.
CraftThe best behavior-change products don't teach. They let you teach yourself.
