Welcome to Synpulse’s digital reading experience – Please choose your region of interest

The Magazine
Management. Expertise. Inspiration.

Date: 20/02/2017

Title: Customer Analytics: Turning Data into Value

Teaser: Utilizing advances in technology and analytical methods to pursue a customer-centric strategy for competitive differentiation and value creation.

Button: Read more


Customer Analytics: Are You Turning Data into Shareholder Value?

Utilizing advances in technology and analytical methods to pursue a customer-centric strategy for competitive differentiation and value creation.

Author: Vladimir Dimitroff

Customer focus is nothing new in business: from the ancient model of the small retailer knowing intimately his neighbourhood customers, through relationship marketing concepts of the late 20th century and CRM as a strategic business discipline (before it became technology), to modern-day customer experience and omnichannel engagement management. What keeps it current and amplifies its business importance is the exhausted sustainability and diminishing returns from other competitive and growth approaches. Globalization is reducing opportunities in new markets and price competition in commoditized sectors is a dead-end street. The Customer dimension (as in Treacy & Wiersema’s seminal «Discipline of Market Leaders» book) remains the only sustainable model for true differentiation and value creation.

In the drive to customer intimacy, technology has played a critical role and is behind the current wave of customer-centric business strategies. What the proverbial «corner shopkeeper » once remembered in his head (about a handful of customers), became possible with millions using computer systems. As storage and processing power grew (and their cost shrank), our appetite for data and respective insight also grew exponentially. Today it would not suffice to look just at transaction data, companies combine a multitude of sources and data types and this growing data universe became known as Big Data. It not only grows in Volume, it also grows with accelerating Velocity and its Variety rapidly increases. These «3V-s» of the Big Data definition are more recently extended to 5 and even 7: adding Value, Veracity, Variability and Visualisation as intrinsic attributes of Big Data. The old question of «Do we have enough data?» has long given way to the challenge «Can we access, organize and utilize all existing data?» Emerging technologies like distributed storage and non-relational databases are offering help with this, as are new analytical tools and methods.

The intersection of customer centricity and data-driven business is currently one of the areas of greatest growth potential in all industries, and particularly in financial services. We believe this fusion area is possibly the key to success for many financial institutions, and will therefore explore some of the associated challenges and opportunities.

«R» is for relevance: Segmentation and beyond

One of the fundamental principles of customer centricity is differentiation: treating different customers differently, based on certain business logic. In practical terms most companies have defined differentiated strategies and policies for groups of customers, popularly called segments — but often without clearly defined rationale for such differentiation. The first and most important purpose of segmentation is resource optimisation: in the reality of limited resources it is of utmost critical importance to deploy them without waste and for maximum outcomes. This makes value segmentation the logical starting point, as allocating more resources to more valuable customers is easily justified, savings from less valuable segments also help to optimise returns.

Relevant allocation of company resources according to customer value is a proven way to ensure optimized returns.

Value, however, is not generated out of nothing: value obtained from customers is only the return from value delivered to them. Customers get value when their (real or perceived) needs are satisfied, hence understanding the needs is critical for providing value to customers and getting value in return. And since needs are a fuzzy qualitative entity very hard to analyze, a number of proxies have been used in segmentation: from good old demographics (a poor predictor) to increasingly preferred behaviours (observable, documented and dynamically trackable). Being relevant to customers’ needs and expectations is one of the imperatives of customer-centric companies and a powerful competitive differentiator.

Finding the right proxies to respond relevantly to customer needs guarantees increased customer value, strong loyalty, and new business.

Segmentations vary not only by the chosen dimension (value, needs or any proxy thereof), but also by the hierarchical level of management objectives and the granularity of segments. Macro-segmentation (also called strategic segmentation) is done at the highest level, with very large groups of customers and is only useful in long-term strategic planning and organisation structure optimization. Operational segmentation is the most widely practised type, where segments are sufficiently differentiated to merit different treatment, but large enough for economical resource allocation. They are also sufficiently stable (few migrations over time) to be useful for operational plans (quarterly, annual) and their execution. Finally, high-velocity data and newer technology capabilities allow dynamic micro-segmentation. Tactical in nature, it is suitable for short-term and ad-hoc campaigns, response to market events and localised actions. While all three hierarchical and granularity levels need to be addressed, customer analytics tend to yield maximum benefits at the operational level, where insight can shape propositions and processes, the design of enabling systems, align budgets and performance metrics. This type is most widely addressed and often referred to as just «segmentation». What can a financial institution do about its customer segmentation in the age of Big Data? Quite a lot: sophistication and accuracy of segmentation algorithms and models can be significantly improved. Models no longer need to be descriptive (only reflecting history), but must become predictive (supporting current decisions with future expectations) and even prescriptive (automating decisions with deep learning techniques that factor multiple scenarios and sequential causality). This, of course, requires harnessing the volumes, velocity and variety of data in the increasingly complex and large ecosystem of concurrent channels and solutions. Not least, it requires adequate expertise and capacity of available skilled resources.

Relevance of customer experience design and delivery

Customer experience (often abbreviated as CX) is the subject of ever growing interest. It is important to every business, as the experience (perceived interactions with any and every part and aspect of the business: communication, product, service, and more) shapes customer attitudes and motivates behaviours, which can be beneficial or detrimental to the company and its shareholders. Historically the focus has been on the rational and objectively measurable part of the experience: length of queues, clicks required to achieve a web task, speed of problem solving by support staff, mobile app responsiveness, etc. These are all important but by no means sufficient; they have become a baseline «hygiene» without which you can go out of business, but which cannot take you to the next level of competitiveness. First scientists, and then business leaders became aware that a vast part of the customer experience is beyond the rational. More than half (some studies suggest over 85%) of human decisions are emotionally motivated and/or driven by subconscious processes. This is where the original volumetric approach to CX assessment and management gave way to newer, less rational and emotion-aware methods. Today CX leaders solicit the help of behavioural psychologists and neuroscientists to introduce structure and discipline into harnessing emotions for predictable business outcomes.

There are still CX experts who preach to always «delight» or «wow» your customers but there is little evidence-based rationale in this. It does sound nice and attractive, but applying identical treatments to all customers misses the point of relevance, not to mention that it is hardly practical or viable. Contemporary methods combining advanced psychology science with sophisticated mathematical models have allowed to uncover causality between stimulus (company behaviour), response (customer emotions invoked), and effect (emotion-motivated customer decisions and behaviours). As the latter drive or destroy value, this is a powerful method to design and manage experiences that cause value-driving customer behaviours. In some available tools «value» can be as specific as targeted company KPI-s, e.g. NPS scores, churn or revenues.

The data-driven, Customer-centric organization: A journey

To harness the combination of customer analytics and Big Data, a company must recognise the various dependencies and synergies between these two areas. Customer intelligence is dependent on data and can be practised even with «small data» (transaction records, interaction history and basic demographic profiles). However, it is immensely amplified by the richness and detail in Big Data and leads to new levels of competitive advantage through utilization of practically endless sources and data types. Big Data, on the other hand, is useless if observed for its own sake, without a pragmatic focus. It finds in customer intelligence one of its most sophisticated and pragmatically rewarding utilizations.

Customer analytics is possibly the ultimate application of Big Data for concrete, pragmatic business outcomes. To benefit from this, a company needs to adopt a structured approach and build adequately evolving capabilities.

An organization can only behave in certain ways in the market if it possesses the right enabling assets and structures. Simply called «capabilities», these features do not belong to a single functional silo like IT or Marketing. Like all transformational strategic disciplines, full benefits are achieved if the capabilities penetrate the whole enterprise across functions, operations and organizational boundaries. This cross-functional view leads to viewing capabilities in a range of critical strategic domains (fig. 1)

graphic graphic
Fig. 1: The CMF (Capability Maturity Framework) analyses a business in a cross-section of 6 'domains' (key business areas) and 4 evolution (maturity) stages. The maturity in different domains and sub-domains usually varies, therfore alignment is as important as overall progress.

Another characteristic of strategic capabilities is their constant evolution; therefore we will look at some major maturity stages on a journey to perfection and market domination. At every level of this journey, maturity is characterized by qualitative differences from the previous stage, only achieved after building specific new capabilities in all domain areas.

Viewed as a matrix of maturity stages across domain areas, this represents a powerful framework for understanding, and tool for planning and managing the company progress along the journey. An example of such a matrix is illustrated below:

  • The Strategy domain examines the extent and the way «data-driven» and «customer-centric» have been adopted by the leadership and communicated inside and outside the organization.
  • Governance and Operations is concerned with the organization and stewardship of data.
  • People and Culture looks at available skills and capacity, plans to grow them, talent management, and how overall corporate culture supports data use and customer focus.
  • Production addresses the actual data analyses happening within the business, types of analytical tasks performed, techniques used, and the nature of analyzed entities and their attributes.
  • The Technology domain describes the available and planned information architecture and IT infrastructure and their key components (platforms and systems).
  • Deployment of analytics deals with the utilization of the output in everyday decision-making and fact-based business management.

In all the 5 domains, capabilities are described in 4 key stages of maturity: Basic, Aware, Competent and Vanguard. The methodology provides detailed criteria for moving to a higher maturity level in each individual capability and each domain as a whole. The illustration shows how a company with an overall average score of ~2 (Aware) can have more advanced capabilities (Competent) in some areas and at the same time lag behind in other areas with just basic capabilities. These internal discrepancies create stress and impede the utilization of partial good capabilities because weak areas neutralise them. Some maturity «building blocks» may be work-in-progress, with partial capabilities already above the previous stage, but still incomplete. And the ultimate (Vanguard) capability maturity is an idealised perfection, more of a distant strategic target, rather than practical benchmark already achieved by anyone.

Where next?

What can a company do to improve its analytical, data utilization and customer-centric capabilities? A good starting point is to select a professionally designed framework and assess the current maturity of capabilities across all domains. This can be done as an express «health check» to trigger strategy formulation or identify pain points needing immediate fixing — or as a full, detailed assessment to analyze gaps and develop a transformation programme. In either case a structured way of looking at cross-functional capability levels is highly recommended.

There are instances where known gaps demand urgent filling and then standalone projects or ad-hoc initiatives addressing the specific area are appropriate. They, however, can be run in parallel with framework use (they will naturally fall into the right spots on the framework and resulting roadmap), and do not preclude the organisation from pursuing comprehensive long-term change. Synpulse has the expertise to assist in either of these scenarios, and the outcome should always lead to data-driven and customer-centric success.



Marouane Bakhtar

United Kingdom

Cookies help us deliver our services. By using our services, you agree to our use of cookies. Find out more.