In 2017, a report in The Economist claimed, “The world’s most valuable resource is no longer oil, but data.” However, utilities still haven’t tapped its potential. Only high quality data can enable an organization to costeffectively serve its customers on a digital platform, can support multichannel marketing and provide an elevated customer experience. It’s time for organizations to monitor, analyze and repair data quality (DQ) issues.
A Customer Golden Record (CGR) is commonly used to measure data quality. However, it gives a static view of the data health of an organization without indicating how or why loss activities occur for customers. To improve data capture, prevent data loss and understand a customer’s pain points, EXL helped a major utility firm create a dynamic version of the Customer Golden Record by integrating it with customer journeys.
The utility firm was struggling to improve data quality and was gradually shifting its structure to align with customer journeys. EXL’s expertise led to a diagnostic tool that helped them explore data capture and degradation in a more granular way. The goal was to create a journey level view for each data item, highlighting paths that led to data loss or lost opportunities for data capture. These performance metrics would be shared with journey owners to improve the customer experience and overall data health of the organization.
Our approach analyzed the issues, making sure they were detectable and avoidable in the future.
Identifying and Sizing the Issues
Customer profiles were tracked monthly for a year to measure how much data was lost or gained over time. We observed that around 3% of customer profiles captured one or more data items, while 10% had data losses.
For our analysis, we applied the concept of the Golden Record and divided the complete data of their customer base into categories - Green, Amber, Blue and Red, in order of descending data quality . The chart above represents the movement of customers from one category to another as they capture (blue flows) or lose (red flows) data.
Monitoring and Understanding Data Capture/Loss
We mapped the event closest to a data capture/loss to attribute a cause to it. To do so, we leveraged ML and AI on all text summaries of customer interaction, including webchats, emails, agent call summaries and system-generated summaries for changes made via online channels. We followed a modular approach for journey identification of each customer, as illustrated below.
By leveraging text mining techniques, the business was able to attribute a specific customer path/event to a data loss, thus prioritizing actions to eliminate these data losses in the future.
Our platform helped the organization shift from a static view of the Customer Golden Record to a dynamic view, with an end-to-end insight of customer data that highlighted areas of improvement.
- This capability helped us identify a prolonged technical problem with the front-end software, leading to marketing consent and title losses. An opportunity to market to 150K customers would have been lost over a year if the problem had not been identified and resolved. Overall, the platform delivered £3.9M in potential annual revenue savings for the organization.
- As the organization restructured around journeys, this new data perspective helped them create better data strategies and governance models.
- The analysis revealed how poor customer experiences like login failures, repeated outbound call reminders and incorrect names/ addresses on accounts contributed to data losses. On the other hand, driving people to online platforms for billing or general inquiries leads to an uptick in email address capture.
The real value was in changing the organization’s culture. We encouraged people to approach problems from a data perspective, to find innovative solutions through data quality. It’s a powerful approach, yet still not leveraged to its full potential.
Potential annual revenue savings of £3.9M by stopping the loss of key data items, increasing the marketing base and driving to digital and self-serve opportunities.
Customers better served by preventing data quality degradation
Potential annual revenue savings due to enhanced data quality