BACKGROUND
Dynamic risk control and customized leverage in prediction markets.
The inception of "Flex Prediction" stemmed from a strategic initiative to offer our advanced users more leverage and flexibility within our Over/Under prediction markets. Traditionally, users were restricted to fixed target scores, limiting their strategic depth. The product vision was to break this limitation, allowing traders to dynamically adjust their risk-to-reward ratios. By empowering our pro-user base with this advanced market capability, the business aimed to drive higher trading volume, increase individual position sizes, and ultimately boost long-term retention among our most valuable users.
As the UX Design Specialist, my challenge was to translate this abstract market mechanic into a tangible, mobile-first experience. I focused heavily on refining the information architecture and visualizing the complex correlation between target scores, implied probabilities, and dynamic multipliers. By designing an intuitive slider interface and establishing clear visual hierarchies—while strategically accounting for real-time market volatility and data latency through adaptive UI states—I successfully crafted a feature that balances advanced market depth with seamless usability.
Industry
FinTech
Sports Tech
Prediction Market
Platform
Mobile
PC
Android
Main Tool
Figma
Claude
Lean UX
Deliverable
UX Research
Information architecture, user flow, edge case mapping, data visualization strategy, usability testing, competitor analysis, UX documentation.
UX Design
Interaction design, high-fidelity prototype, responsive design, typography & color hierarchy, UX component creation, design system integration, state management.
Tech Handoff
Design-to-engineering handoff, edge case handling, user acceptance testing (UAT).
STRATEGIC ALIGNMENT
Unlock user intuition with flexible market mechanics.
In spearheading the Flex Prediction feature, our primary focus revolved around breaking the constraints of static prediction baselines to achieve three overarching objectives.
To realize this vision, we aligned our efforts around three core pillars:
1
Empower users to intuitively customize their risk-to-reward ratio.
2
Drive trading volume, average position size, and pro-user retention.
3
Establish competitive differentiation by covering broader forecasting scenarios.
Research sets the stage.
MARKET INSIGHT
The "take it or leave it" dilemma.
Through market analysis and continuous product feedback loops, we identified a critical friction point in the traditional Over/Under mechanics. It was a strict "take it or leave it" system. If a user disagreed with the platform's fixed baseline, their only option was abandonment.
BEHAVIORAL STRATEGY
The risk-reward trade-off in forecasting.
By shifting away from a binary "right/wrong" mindset, we accommodate a broader spectrum of forecasting strategies. I framed the user psychology using the Jobs-to-be-Done framework to guide our design decisions.
COMPETITIVE ANALYSIS
Learning how others handle flexibility.
I analyzed both direct competitors and popular consumer trading apps to to learn how they simplify complex choices. This exploration revealed that the most successful platforms rely on highly visual, direct-manipulation interactions (like sliders).
UX builds the frame...
UX GOAL
Driving value, keeping it simple.
To turn market insights into reality, we needed to introduce this advanced prediction mechanic smoothly. The goal was to balance business growth with an intuitive experience for all users.
UX ROADMAP
Starting lean, scaling smart.
To manage technical costs and user learning curves, we avoided building a complex tool upfront. Instead, we launched a focused MVP to validate the core interaction model first.
USER FLOW
Mapping the journey, minimizing friction.
To ensure a seamless experience, we mapped an end-to-end journey that integrates naturally into existing user habits.
Whether initiated from the market list or the prediction slip, the flow guides users smoothly through adjusting parameters, reviewing updated returns, and confirming their transaction.
This closed-loop process empowers advanced users with strategic control without disrupting the core experience for casual participants.
INFORMATION ARCHITECTURE (IA)
Layering complexity, guiding the decision.
To manage the cognitive load of dynamic risk adjustment, I structured the interface into three progressive layers: Context, Control, and Execution.
The top section anchors the user with real-time data and market direction. The middle section translates abstract probabilities into a tangible visual chart, paired directly with an intuitive slider for immediate feedback. Finally, the bottom panel locks in the financial impact, allowing users to input their stake and execute the trade.
This top-down hierarchy ensures that users can quickly read the market, adjust their leverage, and confirm their positions with confidence, even in fast-paced live scenarios.
INTERACTIVE RISK CONTROL
Translating math into fluid exploration.
To make complex market mechanics instantly digestible, we paired a continuous slider with a dynamic line chart to create a tight, real-time feedback loop.
As users glide through the spectrum of target scores, the curve updates instantly, visually reinforcing the exponential nature of the risk-to-reward ratio.
By placing critical reference data like "Implied Probability" directly within this visual focal zone, I eliminate the need for mental calculation. This allows traders to fluidly explore different scenarios and pinpoint their ideal balance of risk and payout with zero friction.
... and prototype brings it to life.
INTERACTIVE PROTOTYPE
Hands-on with dynamic risk.
The best way to evaluate an interaction is to feel it. Try the prototype below to adjust the target score slider, and observe how the visual chart, implied probability, and estimated payout respond in real-time ↓↓↓
Test. Validate. Iterate.
QUANTITATIVE INSIGHTS (MIXPANEL)
Measuring market impact and feature adoption through event tracking.
To measure how the new feature performed against our initial business goals, I set up targeted event tracking in Mixpanel to monitor the core user journey.
I analyzed key indicators such as slider interaction rates, average position sizes, and funnel drop-offs. This quantitative overview provided a clear picture of user adoption, validating that advanced users were actively utilizing the customized leverage to adjust their market positions.
QUALITATIVE OBSERVATION (FULLSTORY)
Understanding the "why" behind the clicks through real user sessions.
To get a raw, unfiltered look at how users were actually using the new feature post-launch, I utilized FullStory for session replays.
I obeserved real user behaviors—tracking every click, every pause, and every moment of hesitation as they interacted with the new function. This qualitative deep dive allowed me to bypass assumptions and identify genuine friction points and cognitive overloads that numbers alone couldn't reveal.
Impact and reflection.
Beyond Binary
Transforming rigid predictions into dynamic, user-driven strategies.
USER IMPACT
From guessing to strategy
Standard Over/Under markets restrict users to rigid, binary choices. Flex Prediction disrupts this limitation by enabling users to freely adjust target scores and tailor their personal risk exposure.
This shift transforms passive guessing into an active, engaging trading strategy. It empowers users to leverage their specific insights for granular, high-confidence market positions.
BUSINESS VALUE
Competitive edge & aov
By introducing a dynamic interaction layer, we created a unique market differentiator. It stands out sharply against competitors who only offer traditional, static contracts.
This innovation serves as a powerful retention hook. Motivating users to pursue higher multipliers based on their own analysis directly drives engagement and boosts Average Order Value (AOV).
DESIGN REFLECTION
Visualizing risk vs. reward
The primary design challenge was abstracting complex probability mathematics into an intuitive, accessible interface, ensuring non-expert users wouldn't feel overwhelmed.
I utilized a direct-manipulation slider paired with real-time color transitions and immediate data feedback, allowing users to "feel" the risk-reward ratio and reducing cognitive load.