Designing great customer experiences is getting easier with the rise of predictive analytics—especially now that the pandemic has driven more of us online than ever before.
The COVID-19 pandemic has sped up digital adoption by several years and reshaped consumer behavior in the bargain. As consumers and customers continue to conduct most of their activities online, companies are using advances in analytics to improve performance in marketing and sales, operations, supply chain, and revenue management. However, when it comes to customer experience (CX), organizations continue to lag. During a McKinsey Live webinar, partners Kevin Neher and Victoria Bough shared what leaders can do to better understand what their customers want and to provide experiences at scale in a faster, smoother, more personalized way.
In a McKinsey study of 260 US CX leaders from 14 industries, 93 percent said that they continue to rely on survey-based measurement systems to understand customer preferences and behaviors and to improve satisfaction. While customer surveys remain useful for research, they cannot be the chief tool used to design and deliver better experiences. They provide insights into a very narrow subset of customers and don’t provide predictive analytical horsepower and insights that can be acted upon swiftly. They are not granular enough to reveal root causes of customer sentiment and not focused enough to be the basis of investment and strategic decisions.
However, with the rise of big data and predictive analytics, companies can now build a customer-management capability that is holistic, predictive, prioritized, and value-focused. For example, a leading airline built a machine-learning system that helped identify and prioritize customers whose relationships were most at risk because of multiple flight delays or a cancellation and offered them personalized compensation to save the relationship. A team of about 15 data scientists, CX experts, and external partners worked together for about three months to build the system. This step change resulted in an 800 percent increase in satisfaction and a 60 percent reduction in churn for priority customers.
To do this well, organizations need these four capabilities:
- ingestion of customer-level data, including operational, financial, and feedback
- several types of machine-learning algorithms—to understand and predict what influences customer satisfaction and future behavior
- communication of insights and suggested actions to a broad set of employees
- ability to weave customer-experience insights into day-to-day operations
Four ways companies can get started:
- Change mindsets. Everybody in the organization must understand that CX is of fundamental strategic importance—and think of it as a combination of design, analytics, and technology.
- Break down silos and build cross-functional teams. Excellent CX requires multiple groups to be part of an agile team that makes choices that create business value.
- Start with a core-journey data set and build on it to improve accuracy. Begin by focusing on basic customer-level data, even if the data are not perfect. Over time, recognize where data may be inaccurate or incomplete and adapt the data-acquisition strategy accordingly.
- Focus first on a use case that can drive quick value—one that can create immediate return, build momentum, and gain support. Begin by reviewing major sources of opportunity and pain points along the customer journey.
Questions and answers from the webinar
- What are the most common challenges companies face when embarking on their customer-experience analytics transformations?
When asked about the biggest challenge with the current system, one chief experience officer responded: “People associate CX with marketing, not technology.” That is changing as more and more companies embrace predictive analytics, and it’s up to CX leaders to encourage the change in perception. Predictive systems are not the domain of the IT department or a data-science team, and today’s CX leaders need to build a predictive customer-experience platform—not merely dabble in data. - How are companies navigating data ethics and unconscious bias when it comes to customer experience and predictive analytics? Any best practices?
Regardless of the source, all data collection, storage, and use should follow privacy and cybersecurity best practices. Organizations should follow regional data regulations and remove any variables related to protected classes, such as race and religion. All identifying information should be encrypted and anonymized before it is analyzed. Finally, regular risk reviews can help detect algorithmic bias in CX systems. Our colleagues have found that customer-data protection can serve as a source of competitive advantage as consumers become more careful about sharing data and avoid or stop doing business with companies whose data-security practices they don’t trust.
* * *
For more on this topic, please watch the webinar recording, read the article “Prediction: The future of CX,” and visit our CX featured insights page.
The contents of this site, including any statements, articles, graphics, charts, checklists, and other materials (“Content”) are for informational purposes only. The Content is not intended to be a substitute for professional advice or constitute medical advice.
This content was previously posted on McKinsey’s COVID Response Center and is subject to our commitment to the Open Covid Pledge under these Terms of Use.