Marketing Analytics using Microsoft Azure

This intensive course gives you hands-on experience using R with Microsoft Azure covering the key techniques for analyzing customer data for Sales and Marketing. The focus is on getting to the business results and you will return to your organization with the skills you need to deliver and demonstrate commercial impact from your work.

About this course

This course will give you an overview of the key techniques used in Marketing for analyzing customer data. It will provide hands-on training using R and Microsoft Azure applying these techniques to real business scenarios and using real data.

The focus is on getting to the business results with just enough technical and mathematical detail to allow us to get there reliably. We will demonstrate how you deliver and measure commercial impact and how you can leverage Microsoft R and the Azure platform to create value for your organization.

We cover how to manage the customers through the full life cycle from acquisition through segmentation and cross-sell to retention and churn. Additional techniques include survey and sentiment analysis, price elasticity analysis, and customer lifetime value models.

We have been running this course since 2013 and have 100% positive recommender score from our students.

Feedback from students

This course has consistently achieved 100% positive promoter scores both on recommending the course and recommending the instructor.

Actual student quotes:

Great course, a nice overarching framework, liked the way everything was tied back to campaigns.

Thanks Allan! Glad to take the course with you and hearing from your experience! I liked the course because it was not only R (technical) but also Marketing.


You are expected to be comfortable using R at a basic level. If you are familiar with loading data using read.csv or similar; if you can manipulate data.frames using merge and the *apply family of functions; and if you know how to fit at least a linear model using lm and a model formula (of the type y ~ a + b*c) then you should be fine.

You will need access to a working copy of at least Microsoft R Open with the ability to install packages. For the advanced exercises and homework, you need an Azure subscription and a copy of R. Your company probably have an Azure subscription or you can sign up for a trial. We will be creating HDInsight Spark clusters on Azure for distributed computing and use Microsoft SQL Server for deployment of models, but don’t worry: we will show you how to set it all up using your Azure credentials.

Target audience

Business owners (marketing managers, product managers, customer base managers) and their teams; customer insights professionals.

Course contents

The course follows the customer life cycle from acquiring new customers, managing the existing customers for profitability, retaining good customers, and finally understanding which customers are leaving us and why. Additionally, we will cover sentiment analysis, price elasticity analysis, and customer lifetime value models.

We will be working with real (if anonymous) data from a variety of industries including telecommunications, insurance, media, and high tech.

Part 1: Inflow - acquiring new customers

Our focus is direct marketing so we will not look at advertising campaigns but instead focus on understanding marketing campaigns (e.g. direct mail). This is the foundation for almost everything else in the course.

We look at measuring and improving campaign effectiveness. including:

  1. The importance of test and control groups. Universal control group.
  2. Techniques: Lift curves, AUC
  3. Return on investment. Optimizing marketing spend.

Part 2: Base Management: managing existing customers

Considering the cost of acquiring new customers for many businesses there are probably few assets more valuable than their existing customer base, though few think of it in this way. Topics include:

  1. Cross-selling and up-selling: Offering the right product or service to the customer at the right time.
    1. Techniques: RFM models. Multinomial regression.
    2. Value of lifetime purchases.
  2. Customer segmentation: Understanding the types of customers that you have.
    1. Classification models using first simple decision trees, and then
    2. Random forests and other, newer techniques.

Part 3: Retention: Keeping your good customers

Understanding which customers are likely to leave and what you can do about it is key to profitability in many industries, especially where there are repeat purchases or subscriptions. We look at propensity to churn models, including

  1. Logistic regression: glm, rxGLM, rxLogit, and newer techniques (especially gbm as a general tool)
  2. Tuning models (caret) and introduction to ensemble models.
  3. Using Azure for model creation and optimization.
    1. Distributed data manipulation: using dplyr.spark or sparklyr to manipulate your data across HDInsight Spark cluster on Azure.
    2. Distributed computing: the foreach framework in the context of caret and package doAzureParallel to distribute the processing using HDInsight on Azure.
    3. Comparison with other cloud-based Advanced Analytics tools: including Microsoft Cognitive Services, Azure Stream Analytics, and Azure Machine Learning, to understand when to use them.

Part 4: Outflow: Understanding who are leaving and why

Customers will leave you – that is a fact of life. What is important is to understand who are leaving and why. Is it low value customers who are leaving or is it your best customers? Are they leaving to competitors or because they no longer need your products and services? Topics include:

  1. Customer lifetime value models: Combining value of purchases with propensity to churn and the cost of servicing and retaining the customer.
  2. Analysing survey data. (Generally useful, but we will do a brief introduction here in the context of exit surveys.)
  3. Price elasticity analysis: discovering the optimal price for your products and services.
  4. Deploying models: options for deploying the model in the business, especially using SQL Server R Services for in-database analytics processing.

Course format

Instructor-led training consisting of five half-day sessions with in-class exercises as well as homework. We have successfully given the course in both physical and online (virtual) classrooms.

About the instructor

Allan Engelhardt has 30 years of experience in customer insights and customer relationship management after originally graduating in experimental particle physics and working at the CERN laboratory. He has worked with large corporations across Europe and North America to transform the way they look at their customers and derive value from data, and has held interim management roles at leading Dutch and Irish mobile phone operators building their Insights and Customer Base Management teams.

Allan is one of the founders of CYBAEA, which provides analytics-as-a-service across the globe with a strong focus on commercial results.

His teaching style focuses on practical example and emphasizes results over theoretical sophistication: his courses are for practitioners who need to deliver value to their organizations and while he covers just enough theory to make sure his students are on a firm footing his teaching is not geared to more theoretical students. Expect much hands-on work and very few formulae.

Feedback from students

[I liked that the course] was coming from a person with industry experience.

A lot of hands-on information that is hard to find elsewhere.

Dates and more information

Contact us to register your interest and receive more information on this course, and we will let you know the next time we run a public class. If you have several colleagues who are interested then we can also run a course just for you.