Customer Profit Analytics in the “Big Data” Era: What Has It Changed? What Remains the Same?

By on

by Jaime Fitzgerald @jaimefitzgerald

According to Peter Drucker, “There is only one valid definition of a business purpose: to create a customer.” (The Practice of Management, 1954). Put simply, companies exist to acquire and serve customers profitably. In today’s data-driven era of ‘Big Data” and related technology, the benefits of relentless “customer-centricity” are within our reach. However, to unlock the benefits of applying Big Data to optimize customer profitability, companies need to master the old-fashioned but high-impact basics….

To create and maintain profitable customer relationships, companies need to answer three key questions

  1. How do we define “customers”?  At a pharmaceutical company, does the word “customer” mean “patient” or does it mean “doctor” or even “insurer”?  In this case, all three would fit the generic definition of customer as intended by Drucker. Each group has needs the company exists to meet, impact on profits earned. They should be treated as distinct but related types of customer.
  2. What do we do for customers?  What is the value we bring to them? This is the reason they are willing to pay us…which leads to the ultimate question…
  3. How do we earn money from our customers?  To be sustainable, companies need to create and maintain profitable relationships with customers. This requires that customers will pay us more than our costs to enjoy our products.  We need to create customer relationships that yield positive net profits.

Customer Profit Analytics answers the last question at a granular level.  I say granular because it tells us “where the profits come from” on customer-by-customer basis. We are creating a “Profit & Loss Statement” at the customer-level. When integrated with other customer data, this allows us to answer questions such as:

  • Which customers yield the most profit?  Which customers are barely profitable after considering the cost of serving them?  Which customers are we losing money on?
  • Which segments of the market we serve yield the most profitable customers for us?
  • What buying habits and service behaviors impact customer profitability the most, and how?
  • How does customer profitability evolve over the lifetime of our relationship?

Most importantly, measuring profit at the customer level helps us understand why some customers are more profitable, opening to the door to a range of fact-based decisions which improve profitability.  For example:

  • Optimal design of strategies and tactics for customer loyalty and retention
  • Enhanced ROI on marketing initiatives
  • Product design, pricing, and positioning
  • Calibration of customer service policies and processes (for example, to prevent defection of high-value customers due poor service.  A single “moment of truth” can alter forever the predicted profit impact a customer will yield a firm…)

Does Big Data “Change the Game” for Customer Profit Analytics?  I would offer a “qualified yes”….

  • Big Data doesn’t change the basics. Success still depends on accurately measuring both revenues and costs at the customer level.  This is a mathematical process which applies longstanding best practices. (Little-known fact:  customer profit analytics have been around since the early 1980s, or earlier, when banks began building customer profile databases to perform segmentation analysis)
  • Big Data Technology (hardware, software, etc.) change the game via speed and granularity.  In the past, tools like SQL or even SAS, running on hardware with limited power, created constraints on how frequently customer profit metrics could be updated, and how granular these metrics could be.  Big Data technology adds firepower, expanding the frontier of what can be analyzed and optimized
  • Big Data improves ROI on customer profitability data itself.  What I mean here is that while customer profitability metrics are valuable, they are merely “descriptive.” They give you a customer-level P&L statement, a scorecard.  To make money, we need to make new and better decisions.  We need “prescriptive analysis,” fact-based recommendations, leading to profitable changes and decisions.

Put simply, integrating larger and more diverse datasets with customer profit metrics creates a powerful platform from which to understand, predict, and improve the most important metric in your business:  the profit you earn from your customers.  A few examples below illustrate the breadth of this potential:

  1. Predictive Analytics: using future customer profit as the “outcome variable,” we are now able to predict the future value of a current customer relationship, but also how our actions cause a change that future profit value.  For example, one of our clients discovered that for certain customers, if customer service agents waive a $20 fee upon request, projected customer profitability increases by more than $100.  By loading “decision flags” into their customer service systems, they are making better decisions in real-time.
  2. Customer Profiles:  The long-standing concept of “customer segments” grew up in an era of limited variables along limited dimensions (for example, demographics, geography).  With big data, the same concept exists, but we can explore far more dimensions of customer attributes, behaviors, motivations, beliefs, and sentiments.  Deciding which data is useful depends on business model and analytic objective, but the potential to “know your customers better,” is significant.  An example is “clickstream tracking” when customers visit your website.  With each click, customers are leaving us clues about their interests, their questions, and their needs.
  3. Customer Experience Measurement and Optimization:  In my opinion, measuring the experience of customers remains a critical gap for most corporations.  Customer behavior–and profitability–is significantly impacted by the experiences customers have with us.  Yet many firms struggle to measure customer experience at the operational level, and link these experiences to customer sentiment, loyalty, and profitability.  The good news is that big data makes this easier in many ways.  Examples include sensor-based data, text analytics, social listening, and more accurate sentiment analysis.

What do you think?  In your work experiences, what combinations of big data and related tools have unlocked the most insights about how to improve customer profitability?

Leave a Reply