Understanding and enhancing customer experiences (CX) is a priority for consumer-facing brands. The challenge, however, is customer journeys aren't linear: they span many channels, devices and organizational boundaries and may take place over days, or even weeks. For example, a wireless customer unhappy with his current service may research offers online, call the existing provider to ask for a lower monthly bill and, if unsatisfied, walk into a competitor’s store to select a phone and sign up for a new plan.

Surprisingly few brands are equipped to tackle journeys that cross multiple touchpoints. While adept at analyzing and optimizing discrete interactions within a single channel, most brands lack the ability to understand end-to-end journeys that span multiple touchpoints, multiple channels and multiple visits. 

By contrast, a disciplined approach to delivering cross-functional, end-to-end experiences can result in significant financial and competitive payoffs. McKinsey shared the story of one bank that redesigned its customers’ bill-paying, card-replacement and identity-verification processes using a journey-centric approach. These efforts generated a $2.3 billion lift in customers’ annual spend and a 28% increase in customer satisfaction, propelling the bank to the top three in customer satisfaction among 10 leading global banks within three years.

The Limitations of Current CX Metrics

Most enterprises are familiar with digital analytics, journey mapping, and NPS and CSAT surveys as ways to measure their CX success. Unfortunately, these techniques are limited in their ability to find CX pain points and drive improvements. They typically fail to do all of the following: leverage customer big data, determine the upstream root causes of pain points, quantify the cost of pain points, provide actionable recommendations, or measure the impact of CX improvement initiatives.

  • Digital analytics includes web analytics and mobile analytics. These tools analyze user behavior on web sites, mobile apps and interactive voice response systems to understand common flows, conversion paths, entry and exit points, and interaction metrics. These analytics can help to optimize the design of individual applications, but cannot improve the effectiveness of journeys that cross multiple channels.
  • Journey maps provide a visual representation of a process, showing the user’s questions, expectations and emotions at each point of the experience. Journey mapping can help document knowledge and perspectives about your customer experiences, but it doesn’t look at data, and doesn’t tell you how to drive improvements. 
  • NPS/CSAT surveys can reveal which of your customers are unhappy, what they are unhappy about and overall trends in customer sentiment. However, they are skewed (not everyone responds to surveys), do not provide root cause insight and do not tell you how to drive improvements.

Related Article: Why Customer Journey Mapping + Journey Analytics = 5-Star Customer Experiences

How Customer Journey Analytics Works

Customer journey analytics is a data-driven approach to understanding how end-to-end customer experiences affect business outcomes. Using a commercial product or in-house technology, an analytic platform processes millions or billions of events across touchpoints and channels to reveal how customers get things done and the pain points they face along the way. The methodology for generating CX insights and recommendations can be summarized in three steps:

  1. Collect and connect data.
  2. Understand CX pain points and their root causes.
  3. Implement actionable recommendations.

A journey analytics initiative is guided by primary use cases or areas of inquiry that ultimately aim to increase sales, reduce costs or improve NPS. For example, a telecommunications company launched a project to increase self-service containment, improve new customer onboarding, reduce complaints, and reduce calls coming into the call center. Through journey analytics the company was able to identify the root causes and come up with multiple solutions to address these challenges.

The first step is to collect data, which can come from any location, whether it is a data lake, an enterprise data warehouse or raw system logs. An initial project can start small, with a sample size of 50 million records from less than 10 data sources, such as: web data, mobile app data, call data, chat data, contact center data and NPS data. Later projects can expand the size and scope of the data. For example, the telco mentioned previously ingested 2 billion records from 44 sources.

To make the data useful for journey analysis, customer events that relate to the same individual have to be connected. In most enterprises, a unique customer identifier is not consistently applied across different data sources. However, link events (such as user logins and mobile check-ins) and link attributes (such as email addresses, phone numbers and loyalty numbers) can be combined to connect web sessions, mobile app sessions, phone calls, store visits, orders, complaints and survey feedback.

Learning Opportunities

By mapping together all the events of individual customers into individual time sequences, actual customer journeys can be discovered within the data (rather than relying on assumptions or suppositions about journeys based on anecdotes or intuition). The analytic engine looks for patterns across the millions of event sequences to identify CX pain points or opportunities. For example, the engine can reveal the common paths during the 30 days that lead to an order, or an attrition, or a complaint. Other journeys of interest could be the paths that follow a new account sign-up or a bill increase.

To understand the root causes of these CX pain points or opportunities, further analysis is performed to differentiate user segments that followed a similar path but resulted in a different outcome. The analytic engine determines the user attributes or journey characteristics that are the strongest predictors of the outcome of interest (such as a conversion or churn).

Once the root causes are identified, the analyst works with other stakeholders to develop actionable recommendations. The business case for these recommendations can quantified (in terms of financial costs and NPS) by measuring the number and other metrics of matching sequences in the journey data.

For the telco referred earlier, the team conducted three ‘deep dives’ that resulted in 119 actionable recommendations. As one example, a problem to be solved was the high number of calls regarding the SIM activation process. Customer journey analytics revealed consumers were calling because they didn’t understand the instructions that came with the mailed package, and were unable to complete the process online. This drove close to 150,000 calls per year. Through journey analytics, the team identified nine solutions to this problem, from simplifying the ‘Quick Instructions’ mailed with the SIM card, to instructing users to turn the phone off/on during the activation process, to pointing users to the correct web page if auto-activation failed.

Related Article: Future-Proof Your Customer Journey Maps With These 4 Techniques

Stop Thinking Channels, Focus on the Journey

Consumers are using more and more channels and devices when dealing with companies. To impact business outcomes, customer experiences need to be managed from start to finish across interactions and touchpoints. Customer journey analytics enables enterprises to understand and diagnose these end-to-end experiences, and thereby direct CX investments correctly.

Learn how you can join our contributor community.