Navya Tatikonda

Consumer Financial services

• Designed and implemented causal experiments (incentive and non-incentive) to assess the impact on user behavior, leading to an 80% increase in engagement among inactive Add Cash users, and a 40% increase in the onboarding of new direct deposit users.
• Applied unsupervised learning (e.g., K-Means, DBSCAN) on multi-product usage behavior to uncover customer segments, improving targeting precision, and campaign ROI.
• Automated instrumentation tracking and validation checks using Python and SQL, reducing manual QA time by 80%, and enabling faster experimentation cycles.
• Used statistical models to determine sample sizes, optimize audience stratification, and analyze treatment effects.
• Built scalable data pipelines and event instrumentation to enable full-funnel user behavior tracking for consumer financial services products.
• Led stakeholder alignment and translated complex experiment results into actionable insights for the product, engineering, and marketing teams.
• Owned analytics for the entire Funds-In product suite, leveraging data science to identify opportunities for increasing customer fund inflows, and optimizing product adoption.
• Partnered with engineering teams to define and

• Constructed a sophisticated NLP model that accurately predicts the underlying reasons for customer dissatisfaction with the support team by analyzing the discourse between customers and support agents.
• This involved developing a multi-class, multi-label model that significantly reduced the manual effort required by a team of 300 analysts, resulting in a 35% increase in identifying multiple drivers of dissatisfaction and minimizing human errors and bias.
• Utilized ML classification techniques to interpret NPS scores and provided recommendations to improve the product based on customer feedback, resulting in improved customer satisfaction, and increased revenue for the company.
• Analyzed survey data to identify user segments for both members and trailers, providing insights and recommendations to improve the product offering, and increase customer satisfaction.
• Conducted in-depth analysis, product performance analysis, NPS survey analytics, text mining, and sentiment analysis for a popular streaming network and a famous search engine, resulting in improved insights and decision-making capabilities for the team and stakeholders.
• Employed EDA and feature engineering techniques to improve model performance, resulting in increased accuracy and better predictions of customer dissatisfaction with the support team.

• Developed a model to forecast SLA breaches, enabling proactive resolution, and improving service quality.
• Built NLP techniques to extract root causes from unstructured support ticket text, significantly reducing issue triage and resolution time.
• Conducted a detailed analysis on O365 ticket patterns and designed a scalable self-serve digital community solution, reducing manual engineering effort by 90%. With all the detailed analysis, proposed and implemented a robust way of handling O365 tickets through a digital community, resulting in a 90% savings in the time engineers worked on tickets.