Engineering customer obsession

Data mastery, predictive analytics, and advanced artificial intelligence are essential
to ensuring that customer obsession powers your business forward.

By Euan Davis

Download (PDF)

Published: December 2, 2024

Key takeaways:

Companies that master their data outsell, outmarket, and outperform the competition.
Generative AI enables companies across sectors to enhance customer experiences by synthesizing data to respond empathetically and effectively in real-time.
Many companies face difficulties managing and integrating data due to limitations in data infrastructure and organizational alignment. These challenges must be solved to become customer-obsessed.
As companies move toward real-time, hyper-personalized experiences, they will need to innovate continuously.
Virtusa will conduct a series of research studies focused on the data infrastructure and organizational changes required to enable customer-obsession.

Introduction

Nearly 20 years ago when it began segueing from direct mail to video streaming in 2007, Netflix Inc. realized unlocking customer intelligence would be integral to its success. What Netflix did next was to create the rulebook for the next 100 years in much the same way Ford did a century ago with its automated assembly line.

The then $1.1 billion revenue company deployed advanced machine learning and artificial intelligence (AI) to mine the vast array of customer data it captured on its streaming video platform. Of course, back in 2007, the term AI meant large and expensive rules-based systems that required “big data” to distill meaningful insights from digital signals. Netflix’s head start has delivered phenomenal results: With $33 billion in revenue in 2023, the company is now the king of streaming, the category killer in the $106 billion market.¹ Moreover, its humongous data storehouse and its technologies for sense-making give the company new advantages in understanding what kind of programs could attract big audiences in third-world countries such as India, now its second-largest market.²

Successful companies across industries can now anticipate what customers will need next and what to do about it – long before their competitors catch wind.

Outsell, outmarket, outperform

Yet Netflix is far from being alone in amassing a trove of customer data to outsell, out-market, out-service and out-innovate. And with generative AI, companies now have technology that can redefine what it means to be “customer-centric.” Some firms across industry are applying the technology to deliver business value at scale. Banks, for example, are supercharging human customer service agents with generative AI assistants that summarize vast repositories of customer data. The result: They immediately provide reps with answers that demonstrate customer empathy and deliver better experiences.

Check out Discover, a $20 billion-plus diversified financial services firm. Discover integrates generative AI into the customer experience: Contact center agents now have access to document summarization and real- time search assistance to answer questions promptly and accurately.³ Or consider the Allstate insurance claims app, “Tell My Story.” It lets a service rep listen to the customer explaining what happened in their own words rather than forcing them to use the firm’s tedious claims checklist.⁴ (The app uses natural language processing to extract the story and generative AI to assemble the summary of the claim detail.) Or imagine a hard-pressed medical professional saving hours each day (and months each year) by automating the laborious process of logging patient notes and parsing them for meaning. And once Apple perfects Apple Intelligence, it will deliver on the promise of contextual services based on the vast amounts of personal information held on our mobile phones. The possibilities will be endless.

Successful companies across industries can now anticipate what customers will need next and what to do about it – long before their competitors catch wind. They combine modern technology, sophisticated tech skills, and savvy organizational thinking on where to peel customer insights from the raw digital data that abounds inside and outside of their organizations. And now they are jumping on the next iteration of AI, generative artificial intelligence, because they see it as a rocket booster for catalyzing knowledge work.

However, success with generative AI takes more than a large language model, algorithmic acuity, and an intuitive application interface. It requires a stable foundation of clean and accessible data. It also demands water-tight security. When a company can do that, it can shift from being “customer-driven” to being “customer-obsessed.”

Our research on customer obsession

In the upcoming months, Virtusa will conduct in-depth research on the business and technology strategies of customer-obsessed companies.⁵ We’re staging our research in three parts, with each one focused on how leading U.S. banking and financial services, insurance, healthcare and life sciences, and telecommunications companies have changed the way they do business to thrive on customer obsession.

The first study will explore the data foundation and technology tools for customer obsession, as well as the policies and procedures to glean meaningful insights from customer data. We look at the important role that AI (especially, generative AI) is playing today, and is expected to play in the near future. In the following section, we describe in more depth what we’ll look at in the first phase of our research.

Moving from customer-driven
to customer- obsessed

We live in an age of data abundance. In fact, the data generated by core business processes alone long-ago surpassed human- level understanding. New systems, processes, and talent need to be orchestrated to make meaning from the deluge, which shows no signs of letting up.⁶ At large companies, much of this data is highly structured and captured in core business systems (typically enterprise resource planning systems) that can reach petabytes.⁷ However, a growing percentage of customer data is unstructured (generated by text, video and audio posts in social media and other platforms) and semi-structured (email interactions and chatbot conversations with customer service teams, etc.).

To make sense of this fast-growing repository of data, companies have built so-called data lakes and lake houses.⁸ With tools and dashboards to analyze this data, some companies have gained unparalleled insights into customers’ interests, attitudes and behaviors. As a result, they have become customer-driven. By customer- driven, we mean they have a good handle on such things as which customers are most likely to purchase their products and services, how to sell more effectively to them, and how improve support.

But in today’s always-on digital world, being customer-driven is not enough.

The customer data that turned around Uber

Consider Uber Technologies Inc. In 2018, the ridesharing pioneer realized it needed to refresh its ride pick-up experience for both riders and drivers. The company had just come off a $4 billion loss on revenue of $7.9 billion in 2017.⁹ With an initial public stock offering ahead, the company knew that 2018 had to be better to impress investors.

Turning losses into profits required Uber to dramatically reduce rider and driver dissatisfaction, which had produced unacceptable levels of customer cancellations and driver productivity.¹⁰ This was especially important given that Lyft and other rivals were eroding Uber’s first-mover advantage.

At the heart of Uber’s ride cancellation problem was poor customer data. If riders chose a poor pickup spot, Uber didn’t offer a better choice. Among the challenges: Flawed GPS signals. They can undermine driver navigation – for example, sending them the wrong way down one-way streets, or suggesting they pull over in spaces that prohibit parking. So Uber created a computational model for identifying the best locations for rider pickups based on historical pickup data. It then devised a metric for evaluating riders’ experience based on their complaints and other feedback viewed as “active data.” Uber also capitalized on “passive signals” such as GPS data and user activity to identify when a rider or driver was uncertain of the pickup location. As well, the company collected data from other organizations – on traffic patterns, congestion, parking restrictions and other vital information – all of which could make finding their Uber driver a chore and tempt them to find other transportation.

Although Uber recently ran afoul of Europe’s GDPR for alleged inappropriate use of driver data,¹¹ it continues to lead the ridesharing industry. The company’s revenue more than tripled from 2018 to 2023, from $11 billion to $37 billion.¹² Before its public offering in May 2019, Uber was able to point to a highly profitable 2018 – $997 million in net income. And while the company would lose a combined $25 billion from 2019 to 2022, it rebounded sharply in 2023 with a $1.8 billion profit. And Uber did that while holding off the competition in the U.S. market. The company’s market share as of March 2023 was 76%, up from 71% in July 2018.¹³

In short, Uber set out to collect and use all the
data it could collect, analyze and act upon to turn frustrated customers and angry drivers into loyal ones. It became customer-obsessed – zealous about delighting riders who have multiple choices for transportation, and pleasing drivers, who can take their skills elsewhere.

computational model for identifying the best locations for rider pickups

evaluating riders’ experience
based on their complaints and other feedback viewed as “active data.”

“passive signals” such as GPS data and user activity to
identify when a rider or driver was uncertain of the pickup location

The customer data that helped The New York Times generate a big increase in digital subscribers

The New York Times is another great example of how data-informed customer obsession can produce superior financial performance. Between 2011 and 2013, the number of web visitors to its homepage plummeted more than 60% (from 160 million to 60 million).¹⁴ Then-CEO Mark Thompson warned that if the newspaper publisher didn’t double online revenue by year- end 2020, he would drastically reduce newsroom investments. He pushed the company to embrace a subscription model to reduce its dependence on ad sales. To support this strategy, he invested heavily in a data and analytics infrastructure to better understand audience needs.

The technology gave Times executives unprecedented insights on services that online customers would gladly pay for through separate subscriptions. It launched offerings spanning games (e.g., Wordle and Spelling Bee), cookbooks, product evaluations (Wirecutter), and sports feature stories (The Athletic, an early 2022 acquisition whose subscription service has replaced the NYT’s vaunted sports section).

By January 2020, the Times’ digital initiatives had delivered more than $800 million in revenue, meeting Thompson’s goal ahead of schedule. Last year, the company added 300,000 digital subscribers, bringing its total to 10.3 million – 94% of whom were digital only. They didn’t need the print version of the paper.¹⁵ This pushed annual digital subscription revenue to over $1 billion for the first time.

Meanwhile, CEO Thompson, hired by cable channel CNN in 2023 as CEO and chairman, now faces a similar task in making its TV and streaming news viewers more loyal.

The customer-obsessed enterprise

To be sure, by no means are Discover, The Times, Uber and Netflix the only companies that have vaulted ahead based on savvy uses of customer data. Take Coca-Cola, which uses AI and digital customer data to inform sourcing, distribution, sales and production decisions.¹⁶ Witness Starbucks, which distills mobile and customer rewards data to drive new product offers and discounts.¹⁷ Or look at Amazon, which pioneered the mining of customer data (reviews, rankings, views and click- throughs) to provide highly relevant and hyper-personalized purchase recommendations.¹⁸

What’s the difference between being customer-driven and being customer- obsessed? The way we see it, customer- driven companies excel at responding to customer requests – for instance, providing easier ways to order their products and service, or better service when something goes wrong. Customer- obsessed companies do that, too.

However, they are much more proactive, near autonomous in how they operate and deliver value. They track every interaction with customers, and many more aspects of those interactions (as Uber did last decade), including the customer experience they provide.

(In Netflix’s case, it was its customers’ experience in viewing its content.) Customer-obsessed companies use that data to suggest products that fit customers’ needs or build new ones from scratch. On the supply end of their business, they find or invest in products that customers are likely to buy, which increases the ROI and reduces risks in product development.

Customer-obsessed companies extract meaning and context across millions of customer interactions and transactions, at scale. They place the customer at the center of their strategy -- and their operations. Moving forward, they will deliver localized, curated experiences laden with real- time emotion-intensity analysis.

Doing this is not easy, of course. Companies that deliver such customer experiences do so by infusing their marketing, sales and service processes with greater contextual meaning. In return, they get supercharged customer loyalty, business model innovation and operational efficiency. Get it right, and any customer no matter how large or small, will receive hyper-personalized messages and buyer journeys orchestrated in real-time. Enterprises will operate in new ways based on customer-obsessed insights, fueled by ever-increasing cross-company access to data.

It’s not easy to move from being customer-driven to customer-obsessed. In fact, despite their high aspirations, many companies struggle to even be customer-driven. Less than half (48%) of Fortune 1000 executives polled earlier this year said their organization was data driven.¹⁹ Yet this was more than double the 24% who said the same in 2023.

What holds back data-driven companies from being customer-obsessed?

A key reason is what we call their data foundation. The first of our three studies will explore this. We’ll look at three primary issues:

The role of a central data organization to control the data, set data strategy and data standards, and set and monitor privacy guidelines. How do such groups create “data products” for marketing, sales, service, and product development to drive customer delight? We will showcase best practices that reveal how this centralized team should be structured and how they measure the success of their initiatives.
Where a central data organization should report. Should it be to the CIO or CTO? Or should different data groups report to the corporate function they service – marketing, sales, service, etc.? What’s clear is that data and AI teams need to quickly embrace new technologies and anticipate rapid changes in business needs, informed by instinct and vision.
The insights that such data organizations must provide. Should the focus be on what functional managers ask for – or on customer behaviors that are hidden to those managers but need to be revealed? Or both? (Note: Generative AI has proven to be a cost-effective way of building and continuously improving the data foundation.) In our experience, customer-obsessed companies see generative AI’s supercharged machine learning algorithms as a way to more quickly spot patterns in customer behavior and attitudes and proactively fill market needs before they are even acknowledged by customers. We’ll highlight the best ways organizations can reorient around newly invigorated customer-obsessed processes – processes that require mastering AI-powered customer data.

Looking ahead

While our first study explores the data foundation and technology tools for customer obsession, our second study will probe the challenges that U.S. financial services, insurance, healthcare and life sciences, and telecommunications companies face in using customer data to lift marketing, sales, service and product development. We will also examine how best-practice companies are using all forms of AI (including generative AI) to shift from being customer-driven to customer-obsessed.

Our third study will reveal how continually improving customer data and customer-facing processes can transform customer data lifecycle management. We’ll learn how best- practice companies are using generative AI in sourcing and extracting, cleansing, testing and usage. Importantly, we will reveal how these companies reduce errors from generative AI. We will also examine how they create and monitor effective customer data policies to prevent data hacking and privacy and regulatory violations.

Together, this report series will offer fact-based insights and practical recommendations that will help your company compete on the basis of customer obsession.

Download (PDF)

Acknowledgments

The author would like to thank Krishna Thiagarajan, Senior Vice President of Analytics, Insights & Data; Saurabh Agrawal, Senior Vice President of Data Offerings; and Gwyllen Daandels, Senior Vice President, Head of Offerings in Europe, for their valuable contributions to this essay.

References

Grand View Research estimate of global video streaming market in 2023; https://www.grandviewresearch.com/industry-analysis/video-streaming-market
https://bwmarketingworld.com/article/india-stands-as-second-largest-market-for-netflix-526779#:~:text=According%20to%20Netflix's%20stock%20exchange,which%20garnered%208.3%20million%20views.
https://builtin.com/artificial-intelligence/ai-in-banking
https://www.youtube.com/watch?v=ArjJ5K10mZ8
Customer obsession has emerged as a way to transform businesses through profound customer understanding, engagement, and empowerment. It is a strategic imperative for businesses to thrive and differentiate themselves in today’s fiercely competitive and digitally driven landscape—what’s not to like? https://www.forrester.com/bold/customer-obsession/
https://www.statista.com/statistics/871513/worldwide-data-created/
One petabyte holds 1000 terabytes (TB) or 1,000,000,000,000,000 bytes.
https://en.wikipedia.org/wiki/Data_lake
https://www.macrotrends.net/stocks/charts/UBER/uber-technologies/revenue
Northwestern Kellogg School of Management report, “Uber: Applying Machine Learning to Improve the Customer Pick-up Experience,” by Mohnanbit Sawhney, Biru Shah, Ryan Yu, Evegeny Rubtsov, and Pallavi Goodman, January 14, 2020
https://www.nytimes.com/2024/08/26/business/uber-netherlands-fine-driver-data.html
https://www.statista.com/statistics/1173911/uber-global-net-revenue-region/
https://backlinko.com/uber-users#ubers-us-rideshare-market-share
https://www.cjr.org/analysis/cnn-thompson-zaslav-turnaround-time.php
https://www.nytimes.com/2024/02/07/business/media/new-york-times-q4-earnings.html
https://www.artificialintelligence-news.com/news/how-coca-cola-is-using-ai-to-stay-at-the-top-of-the-soft-drinks-market/
https://d3.harvard.edu/platform-digit/submission/starbucks-winning-on-rewards-loyalty-and-data/
https://www.rejoiner.com/resources/amazon-recommendations-secret-selling-online
Wavestone. https://wwa.wavestone.com/app/uploads/2023/12/DataAI-ExecutiveLeade rshipSurveyFinalAsset.pdf

Virtusa Research Center

The Virtusa research center is the ideas engine of Virtusa. We are committed to delivering data-driven insights, in-depth analyses, and groundbreaking research reports that spark innovation and guide business transformation. Our global team of industry experts explore how leading companies are using digital technology to out-market, out-sell, out-service, out-innovate and otherwise out-compete in the marketplace.

Learn more

Engineering customer obsession

Introduction

Successful companies across industries can now anticipate what customers will need next and what to do about it – long before their competitors catch wind.

Outsell, outmarket, outperform

Our research on customer obsession

Moving from customer-driven to customer- obsessed

The customer data that turned around Uber

The customer data that helped The New York Times generate a big increase in digital subscribers

The customer-obsessed enterprise

What holds back data-driven companies from being customer-obsessed?

Looking ahead

Acknowledgments

Syncing your Virtusa Information

Syncing Complete

Exporting your personal Virtusa brand items

Export Complete

Loading your existing headshots

Please select any one of the images below

Downloading Secure Assets

Signing out

Successfully signed out. Redirecting to Virtusa Brand

Moving from customer-driven
to customer- obsessed