• Improving the quality of predictive models and introducing a probabilistic model for attributing lost users. Thanks to this, it was possible to increase the accuracy of forecasts to 87%.[Python] • Creation of informative reports to understand changes in forecasts, metrics and features. All this made it possible to minimize losses in the purchase by 10%. • Implementation of the "Prompter" recommendation system, which provides a list of the best settings for campaigns. This allowed to increase profits by 5%. [Python, SQL]. • Creating an auto-updating report on new growing games. [R]

• Used predictive analytics to calculate the optimal rates for a purchase of motivated traffic. As a result, 10% reduction in user engagement costs with similar performance indicators. [R,Python]. • Implemented algorithms for automated financial forecasting, to assist management in developing longer-term business plans. [R, Shiny]. • Used machine learning methods and data mining to predict LTV, likely payers, advertising revenue. Thanks to this, it was possible to increase the accuracy of forecasts to 80%, which allowed to improve the efficiency of monetization and to reduce the cost acquisition by more than 30%. [R, Python, SQL]. • Creating a Genre Map from Audience Intersection Data from AppAnnie, resulting a 30% decrease in time needed to identify similar games and to search for new features.

• Analyzed and evaluated the main monetization and engagement metrics, resulting in costs reductions and a 20% decrease in time needed to highlight problem cohorts. [Python, R]. • A retrospective research of trend forecasting methods accuracy. As a result, finding of the optimal method, which allowed to reduce the forecast error by 20%. [R, Python]. • Report with automatic signals on significant deviations from KPI. [R, Python].

● Gathered, processed and analyzed data to understand of customer behavior, to highlight the most suitable target groups and market segment. This led to 10% audience increase and 15% customer loyalty improve. [TNS Gallup, SQL]. ● Created dashboards and reports. Interpreted and analyzed data to identify key metrics and transform raw data into meaningful information. As a result, a 20% decrease in time needed to understand customers and competitors data. ● Supported key company customer Veropharm (analysis and evaluation of the effectiveness of the advertising campaign, investment and sales). As a result, it became the largest Russian manufacturer of pharmaceuticals.