Winning The Lotto A fresh look at customer segmentation - Accenture Interactive - Point of View Series
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Accenture Interactive – Point of View Series Winning The Lotto A fresh look at customer segmentation
Winning The Lotto A fresh look at customer segmentation No one doubts the value of good In today’s increasingly fragmented marketplace, being able to identify new customer segmentation. The market opportunities and capitalize on them, insights it provides can identify through personalization and relevance— market opportunities, improve both in terms of needs and timeliness—are the targeting and relevancy of paramount. Customer segmentation is communications, inform product recognized as an essential capability for a business that wants to deliver tailored and service development, and also offerings and communications that generate guide overall investment strategies and meet consumer demand. Why then, are across the customer portfolio. With so many companies struggling to achieve digital technologies fueling big the value that segmentation promises? Why are segmentation strategies rarely data, and tech-savvy consumers embraced enterprise-wide or able to drive opting out of mass-marketing more impactful initiatives? And how can campaigns, segmentation organizations gain deeper insights into the strategies need to be more robust marketplace and customers to ensure that their marketing initiatives will deliver the and compelling than ever. anticipated return on investment (ROI)?
The lotto effect First, consider a complex “wrinkle” in customer segmentation which often goes unnoticed. Take a typical customer segmentation effort today. A company wants to identify customer Customer attributes from segments with significant growth potential for a particular product. It has captured 50 customer attributes such as age, gender, to location, account tenure and other customer level information. From this relatively small set of attributes, the company wants to identify the “best” six attributes to base the segmentation upon—a surprisingly complex endeavor. Selecting the best six out of fifty is akin to playing a lottery—50 numbers… pick six… with more than 15 million ways to lose. With the advent of big data creating many Find the ‘best’ more sources and attributes to consider, the challenge only grows. For example, doubling the size of that same dataset doesn’t just double the possibilities; in fact, it causes the possible options to grow from 15 million to over a billion, a growth of over 6000 percent. Traditional analytic methods such as clustering require the person developing the segmentation, who is often a statistician, to determine the best set of attributes facing these odds. It’s no wonder that segmentation based on traditional approaches often lose From their relevance within six months. million In addition, there is also the issue of what does “best” mean? Paradoxically, the lotto to over billion effect, while seeming like an added obstacle, in fact provides the opportunity to search for the optimal attributes based on business objectives. Imagine a more advanced possible options approach to customer segmentation development— one that handles infinite numbers of attributes, and factors in the business objectives at the start of a project Up to 6000% to ensure that the resulting solution is aligned to those objectives. Leveraging artificial intelligence technology, next generation customer segmentation is now a reality. growth
A new way to play the game Too often, segmentation efforts result Figure 1: Segmentation: Traditional versus Next Generation Approach in a solution that is too focused on a single dimension of customer information, such as attitudes, value, or behaviors. Advanced Traditional Methods M segmentation, however, should have a AN UA balance across all the various customer L M Outcome dimensions, while also being actionable. AN Segmentation UA One dimension Most importantly, customer segmentation Analytics L Attributes selected Segmentation Focus Outcome needs to reflect business goals (see Figure 1). Segmentation NS by an analyst IO One dimension Analytics U AT One of the key values of leveraging artificial Attributes selected L Segmentation Focus EV A NS by an analyst IO intelligence technology for segmentation U AT development is the flexibility it provides. This AL EV is critical because the segmentation strategy needs to address three fundamental issues VS while simultaneously balancing customer dimensions and business objectives. AL INTEL L IGEN CE OP Next Generation IC I TI M IZ TIF 1. W ho to focus on? Approach INTEL L IGENCE OPT AR ER IC IAL IM The resulting segments must be highly IZ TIF differentiated across core value AR ER dimensions such as revenue, cost, Outcome profitability, tenure and other aspects Segmentation Optimized Segmentation Analytics to guide investment strategies. Objectives selected with 360° View Outcome by the Business Segmentation Optimized Segmentation 2. W hat to offer and what to say? Analytics Objectives selected with 360° View Each segment needs a rich, unique, and by the Business comprehensive profile to inform product, offer and messaging strategies. 3. W here to find them? In order for the solution to be actionable, the segmentation must be developed so that it can be applied to a customer database or target customers in a mass-marketing campaign. Even with the best intentions, objectives To understand the full potential of artificial often evolve as segmentations are developed, intelligence technologies, consider the Artificial intelligence technologies can be for example, when additional clarity is gained following example using Accenture’s adapted specifically to address these issues. during the review of a preliminary solution. proprietary solution, OptiCluster. Taking a By using advanced data mining techniques Instead of manually “tweaking” the highly-rated previous customer these technologies are able to find optimal segmentation strategy through traditional segmentation project carried out using solutions based on business objectives, methods, artificial intelligence technologies traditional methods for a large North insight quality, and actionability. This can be fine-tuned to ensure the resulting American retailer, OptiCluster was used to provides a unique approach that bridges solution is optimal for the new set of determine how the results would differ. The advanced analytics to business goals, while objectives. The end result is a solution that is solution was to provide customer insights in turn enabling strategies that drive both driven by the business objectives and and guide merchandising strategies. In all, significant market impact. empowered by analytics. there were 240 attributes, including
shopping behaviors, customer Figure 2: Impact of OptiCluster on Segmentation Results demographics, and profitability metrics. However, the goal was to find a measurably Original Results better solution by simply using two types of attributes: customer demographics and shopping preferences. This avoided any bias that profitability metrics can 120 impart on a solution. 110 As illustrated in Figure 2, OptiCluster helped 100 Average Cost in understanding “who to focus on” by 90 differentiating segments across the value dimensions of revenue and cost. In the 80 client’s original solution, only one fairly 70 small segment showed any significant difference from the other segments. By 60 Segment Size contrast, OptiCluster found a solution 50 with significant differentiation across the 125 150 175 200 225 250 275 300 segments, including much larger superior and poor performing segments. As a result, Average Revenue the client would be able to fine tune its marketing and merchandising strategies through better profiling, targeting, OptiCluster Results messaging, and offerings development. Equally important is the fact that the identified individual customer segments 120 have rich holistic, yet unique profiles. 110 Gaining this level of insight into customer segments is critical to developing offerings 100 Average Cost and messaging strategies. In the optimal 90 solution from OptiCluster, every profile dimension showed significant improvement 80 in clarity. In fact, when looking across all 70 profiling attributes, the indices from the 60 customer segmentation that OptiCluster Segment Size found had increased by more than 80 50 percent. The more rigorous a segmentation 125 150 175 200 225 250 275 300 approach, the stronger the foundation is for a more personalized brand experience, Average Revenue as it allows marketers to make their offerings and messaging highly relevant to the target audience.
Hitting the jackpot While traditional analytic approaches generally focus on the question of what to offer and the key messages to use, the limited dimensional segmentations of demographics, attitudes, or behaviors, end up sacrificing the question of who to focus on. The absence of a value dimension such as profitability, revenue, and Net Promoter Score (NPS), makes it very difficult to develop an investment strategy for each customer segment. It also makes it nearly impossible for marketing to win support for the segmentation strategy across the enterprise. While other business units Case Study 1 Case Study 2 may find segment profiles of interest (for instance, what are Gen Y’s attitudes One electronics retailer was able to In another case, the optimized towards our brand?), this isn’t sufficient identity a half billion dollar loss it had segmentation for a financial service to change behaviors. However, if the incurred in a single year from a segment provider revealed strategic information business sees a substantial financial impact that was not identified using traditional about two key segments: while a unique for addressing the needs of a customer segmentation methods. In the absence segment—only five percent—of its segment, then the segmentation can of a value dimension within its previous customer base accounted for 50 percent provide the necessary leverage to drive solution, the company had decided to of the company’s profits, another alignment and mobilization to address their invest equally across the customer base. segment was eating into 10 percent of needs (such as set up a web chat customer This significantly eroded its higher value company profits. With the additional service capability). The results are customers. Once an advanced customer insights provided by a segmentation potentially huge. segmentation analysis was completed, that balanced value, demographics, the company was able to develop an preferences, and behaviors, the company effective strategy to address the issue. was able to refine its marketing As a result, the company was able to messages to the target audiences. In not only take action to recover from the addition, the company was also able loss, but also better manage customers to fine-tune its customer acquisition going forward. By making necessary strategies to improve its longer term operational changes to call center customer portfolio. The result? A massive dashboards, interactive voice response 600 percent ROI by the second year. services (IVR) and performance metrics, This approach to segmentation proved the company provided personalized so useful that the company executives services aligned to customer segments made it an integral part of their strategic and their associated value. planning in subsequent years.
Beating the odds Segmentation strategies will vary for each company depending on their brand, market, and priorities. By leveraging artificial intelligence technologies, analytics can now be aligned with unique business objectives to determine the optimal segmentation. Through the more precise narrowing of the “best” attributes, segment profiles will be both richer and more comprehensive. And marketing strategies based on next generation customer segmentation will be more effective in personalizing and targeting offerings and messaging to relevant customers. With the advent of big data and more advanced analytics, customer segmentation techniques based on traditional approaches will fast become obsolete in today’s multi- dimensional world. Not only do companies risk wasting their investment on the segmentation effort, but more importantly they risk being misled on where value exists in the market and among their customers. To learn more about next generation customer segmentation strategies, contact: Jeriad Zoghby jeriad.zoghby@accenture.com
About Accenture Interactive Accenture Interactive helps the world’s leading brands drive superior marketing performance across the full multichannel customer experience. Working with over 4,000 Accenture professionals dedicated to serving the marketing function, Accenture Interactive offers integrated, industrialized and industry-driven marketing solutions and services across consulting, technology and outsourcing powered by analytics. Follow @AccentureSocial or visit accenture.com/interactive. About Accenture Accenture is a global management consulting, technology services and outsourcing company, with approximately 261,000 people serving clients in more than 120 countries. Combining unparalleled experience, comprehensive capabilities across all industries and business functions, and extensive research on the world’s most successful companies, Accenture collaborates with clients to help them become high-performance businesses and governments. The company generated net revenues of US$27.9 billion for the fiscal year ended Aug. 31, 2012. Its home page is www.accenture.com. Copyright © 2013 Accenture All rights reserved. Accenture, its logo, and High Performance Delivered are trademarks of Accenture. This document makes descriptive reference to trademarks that may be owned by others. The use of such trademarks herein is not an assertion of ownership of such trademarks by Accenture and is not intended to represent or imply the existence of an association between Accenture and the lawful owners of such trademarks. Information regarding third-party products, services and organizations was obtained from publicly available sources, and Accenture cannot confirm the accuracy or reliability of such sources or information. Its inclusion does not imply an endorsement by or of any third party. The views and opinions in this article should not be viewed as professional advice with respect to your business.
You can also read