Sessions

Berlin, 5.- 6. Oktober 2022

Description: A newspaper subscription is defined by various critical events, ranging from the end of the trial subscription to receiving invoices. Based on predictive analyses that anticipate customer behavior during these events, we develop, test, and implement customized marketing interventions covering the whole subscription lifecycle. You will learn about modeling via a custom AutoML-pipeline and its close intertwining with marketing execution that aim to maximize subscription lifetime value at DIE ZEIT.

For a publisher it is key to understand the customer journey of a submission process i.e., which online touchpoints and marketing channels drive submission. An attribution model has been evaluated using different machine learning approaches and is introduced into the marketing organisation of Springer Nature. The combination of describing elements with a predictive component and simulation approach helped us the understand the journey, prioritize and quantify the value of marketing activities.

This talk will introduce the foundational business skills you’ll need to deliver business value and grow your career as an analyst. Drawing on best practices, published research, case studies and personal anecdotes from two decades of industry experience, we give an overview of foundational skills related to Company, Colleagues, Storytelling, Expectations, Results and Careers–emphasizing how each topic relates to your unique position as an analytics professional within a larger corporation.

Making eye contact is one of the most powerful ways to build relationships—whether it’s a new date or a potential business partner. But is it also true for robo-advisors which give consumers investment advice? Today many consumers do not trust robo-advisors, which are mostly text-based interfaces. In this talk, we present a new robo-advisor prototype in the form of a virtual social robot that makes eye contact and show how it impacts consumer trust and investment decisions in online experiments.

Adversarial archetypes exploit workings of any system to disrupt robustness and decision-making of the underlying algorithms. This deep dive presents AuthSHAP, a model agnostic and robust implementation of SHAP to uncover the extent to which key features are not appropriated by any model in the decision making. This ‘knowledge’ is significant information to a fraudster to design intelligent or adversarial attacks. The presentation shows that even in black-box settings, it is possible to understand the vulnerability.

Since 2017, Hauke is a Data Scientist at the innovation department of EWE AG in Oldenburg. Before that, he studied mathematics and worked as a researcher in the field of applied statistics. At EWE, Hauke is responsible for turning data into business value through data science projects in diverse application areas of the energy sector. Additionally, he focusses on spreading data literacy and data culture to enable colleagues to work with data and generate new ideas for data-driven solutions.

Proper utilization of Machine Learning and Predictive Modeling allow companies to increase profit, gain competitive advantages, grow, and win market completion. This is illustrated by results achieved across Major Telecom and Banking Companies. Efficient Proactive Retention, Revenue Stimulation, and Second-best Offer approaches require powerful Churn and Propensity models. There is a big number of techniques that can be applied to any modelling task, but it is almost impossible to know at the outset which will be most effective. Also, there are many ways to interconnect IT and Business. One way to address these topics is through advanced analytical base tables that proved to be highly efficient in all the illustrated cases.

Markdowns allow retailers to get rid of dead inventory and ensure turnover. Traditionally, even AI tools use a rule-based approach that constrains optimization and hurts revenues. Fashion retailer Benetton partnered with Evo to explore a new prescriptive approach to optimize markdown performance through complex process mapping. The resulting algorithm relied on a formal AI forecasting model feeding a price optimization model—a clear improvement that ultimately increased revenues by more than 5%.

During this session, we will discuss the different risk-based categories of AI laid out by the EU’s Artificial Intelligence Act and find out how to become more admissible as per the Act. Thereafter, we will walk through the concrete steps, tools, and practices such as monitoring, explainability, model fairness, and compliance that are instrumental in achieving Responsible AI and building more risk-proof and market-friendly solutions.

Evaluation and optimization of the effect of personalized treatments plays an important role in marketing campaigns and CRM strategies at PAYBACK. However, optimization goals are highly variable due to the wide range of applications and constraints imposed by the industry. Here we compare the use of optimization tools to overcome problems encountered in various business settings, such as cost/profit optimization with restrictions enforced by the CRM objectives.

Transforming data into value is a strategic focus for companies across industries. With the gap between leading data driven businesses and late movers growing rapidly, decision makers are wondering what secret ingredient helps companies to successfully leapfrog roadblocks on their value journey. Since this is evidently not a one-dimensional challenge, we’ll de-mystify some unavoidable buzzwords and take a hands-on perspective on what really helped companies to sustainably turn data into value.

Machine learning models are becoming obsolete and must be retrained – this is the current widespread tenor. Is this actually true? And if yes, which components does a CI/CD pipeline for machine learning really need – and which are optional? How can the whole thing be implemented without building a complete Machine Learning Platform team? And which challenges are still difficult to solve at present? A field report including (mis)decisions, which will help to choose the right path for your own challenges.

Geo-experiments – advertising experiments where the treatment and control groups are chosen based on users‘ locations – provide a privacy-friendly alternative to cookie-based online experiments that can also be used to measure offline effects of online advertising. This session discusses the algorithms we use at Google to design the experiment regions based on geographical user behavior, and the rigorous statistical methods to analyze randomized experiments based on these regions.

The changes in the ads ecosystem have led marketers to lean on existing aggregate experimentation tools that assume a predetermined treatment effect. Choosing the treatment group to ensure you have high chances of detecting an effect is non-trivial. Built by Meta Open Source, GeoLift solves this problem by building well powered geographical experiments. Join us to go over why geographical experiments are necessary and their implications in the marketing industry, along with a demo of GeoLift.