DEVENDRA PATIDAR

New Paid User Journey Team

• Worked on promotional constructs, ran AB experiments with these constructs to decide the winning construct with the aim of retaining the new users along with guardrail of ROI.
• Identified key levers (e.g., win2cea, active day streaks, contraction/expansion of users) to control spends during high supply, and built tradeoff framework for decisioning on spends saving vs retention hit
• Collaborated with product and design to launch user-facing features like Stories, Contextual FPS, AI Coach, Performance Review by identifying pain points in the user journey & proposing friction-reducing solutions
• Led feature prioritization hypothesizing the impact of the solution, and success tracking using engagement and conversion metrics, and iterated based on user feedback and data analysis.

User Quality Score & Personas

• Built a machine learning model to predict user quality using behavioral features like gameplay frequency, app engagement, communication interaction, demographics, skill level, transactions pattern & modes.
• Segmented the user base with clustering exercise based on non-game play features like playing pattern, diversification, tenure, feature usage, risk appetite, friction to add money to carve out 7 behavioral personas (e.g., Loyal, Seasonal, Bouncers etc).
• Pointed out the potential factors which motivates each persona to play (Triggers - rewards, high prize pool) and what may be the reason for their churn (Barriers - losing streak, lack of education). Based on this, created persona-specific engagement journeys including feature nudges, content, and bonus types.

Picks (New Game Mode)

• Built a simulation framework on Databricks to test the game logic, scoring system, and contest dynamics before launch. Used synthetic data to evaluate edge cases and user win distribution.
• Designed the point system for Picks to align with user psychology — ensuring balance between perceived fairness, skill differentiation, and excitement.
• Worked closely with PMs and data engineers to evaluate user feedback and feature performance post-launch, using funnel analysis and cohort behavior.

• Raised threshold rating from 4.5 to 4.65 for providers falling in retraining bucket; enhanced the net promoting score of providers by 6%
• Developed different data-driven strategies using key levers like retention rate of the new users & lead acceptance rate of service providers
• Increased selection to approval ratio of service providers by 5%, formulated plan of action for onboarding process based on extracted data