Date Published: November 20, 2014 - Last Updated 5 Years, 106 Days, 4 Hours, 24 Minutes ago
In today’s competitive marketplace, organizations are scrambling to capture and understand how to derive strategic value from their stores of Big Data. Various industry studies from 2012 to the present have consistently shown that organizations and their leadership are making greater efforts to tap the potential benefits of this mountain of collected information. Despite this interest and executive buy in, most are still coming up short on actionable outcomes and measurable return on investment.
While most organizations are still focused on the row-and-column data that can be pulled and tallied from a variety of internal management and transactional systems, their unstructured customer interactions are the key to providing insights necessary to paint a clear, detailed picture of customer attitudes and motivations.
Structured data is excellent in documenting what has happened in the course of a customer journey but it cannot answer why a particular event happened. For that, organizations must look to their contact centers.
Customer interactions represent the largest non-monetized data asset that many companies possess. When combined with traditional data from other sources, these interactions can provide rich context to help create and refine powerful predictive models to forecast future customer behavior.
Last year contact centers exchanged 31.8 exabytes of unstructured data, with 76 billion hours of human interactions – all explaining exactly what’s right and wrong with the companies, processes and products with which these callers interact. Recent advancements in neural networks, machine learning and cloud computing have made it possible and practical to leverage this underutilized asset.
Leading companies are moving beyond the use of interaction analytics as merely a tool to improve contact center operations. They are leveraging the customer attitudes and motivations revealed in the calls to manage marketing messages, make more sales, reduce attrition, and improve customer satisfaction.
Nexidia’s research has shown that customer behavior displayed during interactions before an event (like buying, downgrading or leaving) are remarkably consistent and are uniquely suited to add a very important dimension to big data modeling.
The predictive models designed to harvest insights from Big Data often produce correlations between “data elements” that are hard to understand – the relationships between cause and effect contain very little business context. Conversations captured in the contact center hold astonishing context (i.e., the exact words and sentiment of the customer), can be quantified – and most importantly – have proven to be extremely predictive of future behavior.
For example, a leading communications company wanted a means to identify customers who were likely to disconnect services so they could take proactive steps to avoid customer churn. Prompted by a 2012 industry-wide research report that indicated 22% of households with TV service were currently dissatisfied with their provider, the company wanted to determine how many of its own customers might be at risk, and take proactive steps to deter account loss. They relied on interaction analytics to gain insight into the clues hidden in their customer calls.
Interaction analytics enabled the company to uncover conversational elements in calls that were statistically indicative of future account disconnects. Key call phrases such as “still not working,” or “I have called multiple times” were revealed, providing a picture of churn threat. Other, less obvious words and phrases such as “please walk me through this bill,” or mentioning the price of a competitor’s product were also shown to be predictive of churn. By combining these insights with additional customer demographic metadata from other systems, the company was able to get a high definition view of their customers’ experience and future actions.
Using algorithm-based scoring, the company identified multiple groups of customers who were likely to disconnect their service in the near future. The organization capitalized on this actionable information by creating targeted outbound call campaigns to at-risk customers.
Customers who received these communications were four times more likely to remain with the company as those in a control group who received no outreach. Results showed that the sooner the outreach occurred, the better the retention rate.
Such outcomes are achievable across industries when leadership captures the vision of what their own contact centers can deliver when equipped with the right tools and processes. As more companies and executives begin to understand the value that customer interactions hold in predicting future behavior, the role of the contact center will be elevated and transformed from a cost center into a strategic business intelligence resource.