Technology, media, and telecom

Understand customer behavior and optimize Interactions

Our analytics tools help leading technology, media, and telecom companies to deeply analyze customer churn and improve customer engagement using data-driven methods.

Use cases

insight

Analyze textual, audio, and video data

We build customer analytics solutions that are able to extract useful signals from virtually any source — call transcripts, customer reviews, camera footage, and many others. These signals can be automatically converted into meaningful attributes and tags and then used in downstream models and analytical processes.

acquisition

Improve Conversion Rates for New Visitors

We build models that personalize the experience of unregistered visitors based on their in-session behavior and device and location data.

These models also identify customers with a high propensity for cart abandonment and prescribe the best mitigating action.

predictive analytics

Analyze the dynamics of churn risk

We develop models that not only identify at-risk customers but also estimate how the risk level is likely to evolve in the long run. This helps to determine optimal treatment and intervention time.

prescriptive model

Automatically optimize next best action

We extensively use prescriptive models to determine optimal marketing actions for each customer. These models incorporate a wide range of signals and optimize both customer engagement and value.

Our clients

Retail

Hi-tech

Manufacturing

Finance & Insurance

How our customer analytics platform works

Reference odels

We provide a comprehensive set of reference models that help to analyze, quantify, and predict various aspects of customer behavior. These reference implementations include models for predicting and mitigating customer churn, complaints, and other customer relationship issues based on behavioral data, user-generated text, and other data sources.

Efficient productization

Our solutions are designed for efficient productization and integration with marketing and customer support processes. We develop models and services that produce precise and actionable prescriptions rather than complex scores and fragmented pieces of analytics.

How to get started

We provide flexible engagement options to help you build customer intelligence solutions faster. Contact us today to start with a workshop, discovery, or proof of concept.

Workshop

We offer free half-day workshops with our top experts in personalization and data science to discuss your marketing technology landscape, customer experience strategy, and opportunities for optimization.

Proof of concept

If you have already identified a specific use case for personalization or customer analytics, we usually can start with a 4–8 week proof-of-concept project to deliver improvements and tangible results.

Discovery

If you are in the stage of requirements analysis and strategy development, we can start with a 2–3 week discovery phase to identify the right use cases for personalization, design your solution or product using industry best practices, and build a roadmap.

Learn more

Customer churn prevention: a prescriptive solution using deep learning

In this article, we discuss how to build a solution that helps to quantify, investigate, and fight customer churn, complaints, and other issues related to customer dissatisfaction. The described approach has been successfully implemented for several clients and has proven to be a powerful generic framework.

Customer2Vec: representation learning for customer analytics and personalization

In this article, we focus on the learning of useful semantic representations (embeddings) for products and customers using neural networks. These embedding vectors then can be used as features in the downstream propensity and recommendation models as well as help to improve customer analytics and segmentation.

Building a next best action model using reinforcement learning

In this tutorial, we discuss how traditional targeting and personalization models such as look-alike and collaborative filtering can be combined with reinforcement learning to optimize multi-step marketing action policies (aka Next Best Action policies). We use methods that were originally developed and tested by Adobe and Alibaba, and they proved to be effective in a number of practical use cases.

Download pdf

Read more on customer intelligence and personalization

Would you like to learn more about algorithmic foundations of personalization and customer analytics? We published a 500-page book on enterprise data science that is available for free download, and there are several chapters on personalization in it.

Download pdf

Read more on advanced customer intelligence

This report provides an overview of recent advances in customer intelligence by examining 10 industrial case studies. These case studies were selected from the consulting practice of Grid Dynamics and public reports to cover the most important, common, and innovative trends in data science and machine learning methods used in modern customer intelligence and marketing analytics. The report covers the following 4 major areas of active research and industrial adoption:

Deep learning models that incorporate a wider range of signals and data, including textual and visual data.
Deep learning models that process sequences of events, including User2Vec models.
Reinforcement learning models for the dynamic and strategic optimization of marketing actions.
Econometric and deep learning models that quantify financial and operational risks.