What are the responsibilities and job description for the Data Scientist Experimental Design position at Mediabistro?
Our client, a leading media company, is seeking a Data Scientist to support in-market control tests and build machine learning models to increase subscriptions and revenue. Will also help to improve app engagement and customer experience. In this role, you will develop models designed to predict churn / LTV, enhance segmentation and customized messaging, and measure the effectiveness of advertising spend.This role will also support test design and evaluation of marketing initiatives across multiple channels spanning the customer lifecycle. A keen interest in using analytics and data science to optimize digital media and conversion is critical.What you’ll do : Support testing marketing initiatives to maximize media spend performance, drive app engagement and increase subscriber retention across the customer lifecycle by applying proper statistical rigor. Responsibilities include experimental design (sample sizing, timing), quality checking implementation, and test measurement and evaluation.Deliver presentations describing experimental results and recommendations to business partners. Prepare dashboards providing daily monitoring of test results.Build machine learning models for predicting key outcomes used to inform marketing tactics across digital acquisition and user journey orchestrationWork closely with engineering to productionize and monitor model performanceRepresent Commercial Analytics capabilities and data-driven insights within cross-functional teams to help achieve business objectivesDevelop complex SQL queries combining data from a wide variety of sources in preparation of feature engineering and test analysisCreate custom Python libraries available for team use to automate key methods. Examples include statistical methods for timing / sizing experimental design, verifying test populations are unbiased and matched-market test design.Types of projects you’ll be working on : Test onboarding / engagement / retention campaign attributes across email, push and in-app messaging. Attributes include segmentation, offer, messaging, creative, frequency and cadence.Develop machine learning models to predict retention at various stages in the user lifecycle. Incorporate predictions into test results to optimize marketing tactics by segment and present feature importance to business stakeholders.Measure incrementality of paid media campaigns using matched market testingDevelop forecasting tools for business metrics using machine learningWhat you’ll need : Master’s degree in Data Science, Statistics, Engineering, Mathematics, or Computer Science5 or more years of experience as a Data ScientistExcellent programming skills in PythonUnderstanding or experience working within an enterprise data warehouse environment (including SQL, procedural SQL, and ETL)Capability to build, tune and explain machine learning models for supervised and unsupervised learning. Methods include, but are not limited to, regression, clustering and ensemble methods.Understanding of statistical concepts, lift analysis and A / B testing fundamentalsAbility to produce data visualizations for explaining technical conceptsExperience with cloud technologiesExperience working within an advertising agency or creative environment preferredFamiliar with marketing communications strategy and processes, including online media, application development and web site developmentLocation : New York City or Washington, DC2 Days a Week Onsite
