Jeevisha Anandani

• Identified domain authentication as a critical growth and deliverability gap (<2.5% adoption). Quantified its causal impact via PSM (+$3.97 incremental revenue; +5.4pp retention) and translated insights into redesigning Mailchimp’s onboarding checklist (increased prominence of authentication, persistent setup tasks, improved feedback states). Drove a 16%+ increase in early-life domain authentication post-launch. • Built and led the causal measurement strategy for Mailchimp’s new email editor rollout, showing that Nuni’s apparent C2S declines (−381 bps in cross-editor analysis; −1320 bps post-migration) were driven by increased creation behavior (+7.7% creates; +27–40% post-migration) rather than performance degradation. Core metrics remained stable (sends flat; revenue flat; churn +40 bps), enabling leadership to proceed confidently and focus investments on editor parity, send-completion UX. • Led causal measurement for Mailchimp’s Brand URL onboarding step using PSM and mediation analysis, showing it directly drives +18% 31-day revenue (+$1.08/user; up to +$13.63 for ITP) and +2 pp 45-day retention (+7 pp for trials). Insights validated emphasizing the step in onboarding and guided creative-tool nudges to amplify impact. • Led an A/B experiment (~39K web FTUs/week), partnering with product and engineering teams, to streamline Mailchimp’s Account Setup flow (11→7 steps). Analysis showed early mandatory fields increased drop-off while core activation (Payoff Rate 20%) and data capture (Website 49%, Vertical 69%) remained flat/slightly negative. Findings informed progressive disclosure and deferred data collection strategy to optimize FTU experience. • Built a segmentation view of Mailchimp C1 customers to identify high-value personas based on integration usage, engagement, and account attributes, uncovering segments particularly with low engagement within email and integrations to help understand early retention drivers.

• Developed a framework for offers to forecast the timeline for auto cut off based on when the budget will burn. • Lead the efforts to revamp the way customer support help is being offered to Fetch users. Designed a 3-stage framework to route tickets to agents by Tier1 categories (Fetch Play, Referrals, Offers, etc), automate the agent’s first responses at Tier3 level (400+ categories) and set up alarms on ticket volume to monitor any spike in the tiers. • Built a robust classification model to tag and automate Support tickets (80%+ precision for each category) first responses using Natural Language. The pipeline is set up using AWS workflows, the monitoring dashboards are hosted on Grafana and Tableau and the orchestration is managed by Airflow. Presented the estimated annual $ saved to the CTO($1.2M). • Designed a topic modeling framework to proactively detect new trends/issues in support tickets to alert the team in case of a high influx and improved response time by 50%.

• Built an end-to-end workflow redesigning the retention strategy employing uplift churn modeling technique using their scan, purchase and offers redemption behavior. The framework improved retention by ~ 20%. • Forecasted hourly signups for the acquisition team to track funnel growth and optimize impact of bottlenecks. • Built a classification model to measure the propensity of 360+ days lapsed users returning to the app. Led an experiment targeting those users through Paid Media Ads which saw a 50% improvement in cost when compared with CAC. • Devised an experiment to see the impact of our top partner brand offers on returned users within their first month. 66% of users redeemed their offer within the first week (highest being with Frito Lays) resulting in min $10K revenue with each one. • Measured user retention and engagement of the CEO’s followers using propensity matching logistic regression model via Python. Saw an increase in scanning behavior and offers redeemed by 6.2% and 10.9% respectively and presented to the CEO. • Designed and led an A/B/n test to check the feasibility of ‘Nudge’ feature in the app. All point segments saw positive incrementality in scans and compared practical vs statistical significance to arrive at the most optimal one.

• Generated insights for the client on various sides of facilities management using Tableau and SQL. • Created vendor performance tracking dashboard across different maintenance trades to direct focus on the Top vendors • Enabled process automation for daily tracking of work order performance for their On-Demand facility management application

• Analyzing transactional level data of clients to understand the key factors of customer attrition • Built alternate classification models to predict customer churn and ultimately went ahead with Random Forest

• Implemented an uplift model redesigning the customer target strategy by focusing on users who will buy ONLY because of the marketing action and increased the ROI by 6% • Built an XGBoost Classification Model to predict the propensity of a client to register for the mobile app. The model results were used to target the right population and increased the registration rate by 134%. • Measured the impact of advisor coaching by developing a propensity matching model and identifying pseudo control using Greedy Match. Accomplished mean growth in the commission metrics by 12%, plans by 5%, and client acquisition by 8% • Assessed the productivity of advisors who were certified in financial planning using propensity matching in SAS and delivered a 4% CAGR in concession, 10% in financial planning, and 14% in client acquisition

• Strategized initiatives to transfer low profitable clients from Franchise to Call Centre groups using SAS delivering a 26% rise in concession, 9% in High valued Client Acquisition and a 20% in Financial Planning • Built an automated dashboard to provide a comprehensive view of Advisor performance which was used by the leadership team to track goals and reassess budget allocation. It helped in reducing implementation time by 70% while ensuring quality and consistency. • Conduct complex queries and utilize data extraction, reporting and BI tools to develop and validate analytical solutions

I worked there as an intern in the Turbine Block. My job was to study the processes employed in the making of hydrodynamic bearings in the turbine and have an overall brief overview.

I worked in the Material Handling department as an intern. My work mainly comprised of managing the inventory and the available space by using optimization techniques and innovative designs for different kinds of storage containers.