Next-Gen Analytics for the Contact Center
Six ways analytics can propel the contact center with real business impact across channels.
Customers have multiple contact options today beyond voice calls: IVR, online self-service, chat, click-to-call, online forums, mobile apps, social media, and more. And usage is growing. According to Forrester, there has been a 12 percent rise in Web self-service usage, a 24 percent rise in chat usage, and a 25 percent increase in community usage for customer service in the past three years.
By the time a customer picks up the phone to call customer service, he or she may already be frustrated with an experience from another channel. More than 57 percent of inbound calls come from customers who first attempted to resolve their question on the company website, and more than 30 percent of callers are on the website at the same time they are talking to a rep on the phone.
As the number of interaction channels grows, it has become even harder to maintain a consistent customer experience across interaction touchpoints. IDG research reports that only 15 percent of CIOs say their companies do "very well" in maintaining a consistent user experience across all communication channels. Since most interaction channels are siloed in today's contact center and channels do not share information with one another, most customer experience breakdowns happen at the boundaries between interaction channels.
We know that customers switch brands when faced with a poor customer experience—62 percent of U.S. consumers have switched brands in the past year due to poor customer service, and one in four cite being shuffled from rep to rep with no issue resolution as their top reason for switching brands4. These poor customer experiences and channel breakdowns can seriously impact churn and in turn, revenue.
It's an issue that can't be ignored, especially as today's contact centers are expected to take a more active role in impacting customer relationships and generating long-term loyalty. But it doesn't have to be an insurmountable challenge for contact center leaders.
Smarter contact analytics=smarter business
In the complex and challenging contact center world, analytics hold the key to working smarter and having real impact across channels on customer experience, satisfaction, and overall business. Data gathered during a service interaction can offer incredible insights to the rest of the organization, allowing them to understand customers more clearly and influence proactive and predictive customer strategies. We believe there are six key principles—driven by technology-enabled analytics—that can guide contact center leaders in their organizational decision making:
1. Get to the root of the interaction. When customers contact your company, even if for apparently mundane reasons, they become a valuable source of information about your business and operations. Most contact centers desire deflection of such contacts to low-cost channels and do not necessarily pay attention to root causes. Root cause analytics can highlight fundamental issues and inform solutions that if implemented can reduce overall interaction volume and significantly improve customer experience.
Root cause analytics may reveal the need to make changes outside the contact center, such as coordinating marketing communications across channels, redesigning bill layout, making Web self-service more intelligent, and issuing proactive alerts.
2. Self-service can, and should be, intelligent. Both companies and customers desire self-service tools. They exist as FAQs, searchable knowledgebases, or even avatars. However, many self-service options today are dumb-—they do not provide the right answers or provide too many possible answers, leaving customers frustrated. Advances in text and speech analytics and machine learning can help self-service tools to select highly accurate responses to customer inquiries. These are highly automated systems that are self-learning, so accuracy improves over time. The one pre-requisite for these systems is a rich knowledgebase.
A great example of intelligent self-service that you may not be aware of is email spam detection. Have you noticed how accurate your email is in self-classifying spam so you almost never see it? This is enabled through text analytics combined with machine learning.
3. Empower and transform your front line. Too often frontline contact center employees are under-paid and plagued with high attrition. They work in a high-stress environment and are treated only as a cost by companies. Today the front line should be viewed as employees who can earn the trust and loyalty of customers during critical moments of truth when customers expend a lot of emotional energy. Success during these moments can drive significant positive business outcomes for the firm.
Examples of these moments could include events like a missed flight to attend an important meeting, a lost credit card when traveling overseas, or a stalled vehicle late at night in an unfamiliar area. During these moments, smart technological solutions alone won't cut it, because complex solutions are required, and because technology alone cannot create an emotional bond. However, emotionally intelligent frontline employees, empowered with the appropriate technology and data at their fingertips, can excel in these situations and create sustainable differentiation. Take the time to determine who within your organization has the emotional intelligence to succeed as brand stewards during the most fragile service interactions.
4. Discover the right channels for customer interactions. Many firms have begun to right channel contact center interactions, or assign specific interactions to specific channels. But analysis is required to make the correct workflow decisions. It is instructive to look at four factors: customer channel preference, customer profiles and behaviors (commonly available in transactional and other data), interaction complexity (e.g., interactions that require significant diagnostics to solve), and moment of truth criticality, the outcome of which will have significant implications for customer loyalty and trust.
Only very high complexity interactions that occur during extremely critical moments of truth should be handled by frontline live employees, with the rest moved to other low cost channels like self-service, chat, or SMS.
5. Follow and inform the customer journey, intelligently, across channels. Today's contact center world is siloed by channels, and sometimes even within a channel. This results in an inconsistent, frustrating customer experience. Many senior leaders understand the problem and want to improve the multichannel experience, but are limited by behemoth legacy infrastructures that support the contact center.
Firms can enable a scalable, massively parallel processing, cloud-based analytic solution that includes three advanced customer insight capabilities.
To make the most of this solution, first companies must enable real-time and batch data ingestion (both structured and unstructured) from multiple channels, touchpoints, and enterprise systems like CRM. This will facilitate a live customer profile and context as customers move along their journey with the company.
Next, get predictive. Analytics-enabled recommenders can implement a multitude of analytic engines, such as interaction reason prediction, sense and respond, next interaction predictor, or next best offer optimizer. These tools are designed to provide the best response in the best channel and takes into account context, customer behavior and history, customer preference, and issue complexity.
Finally, make sure channels are connected. Use channel connectors that seamlessly connect via open APIs/custom integration with various channels to orchestrate response. This applies to self-service channels as well as live channels where the actions may be orchestrated in the IVR or at the agent's desktop. And be sure to allot for future experience channels, to be as nimble and flexible to changing customer behaviors.
6. Pre-empt customer interactions with analytics. Most customer service interactions are inbound. So what if we were able to pre-empt them and solve the issue before the customer reaches out to interact? This would not only reduce interaction volume, but also delight customers. Predictive analytics can help. They can be used to predict interaction reasons even before the interaction happens.
For example, if analytics showed that customers who had Problem X (e.g., install issues with Internet service) also tended to have Problem Y (e.g., wireless network setup issues), then a customer who reached out with Problem X would be prompted to check and solve for Problem Y during the same interaction, thereby reducing repeat call incidence, improving FCR, and improving customer satisfaction. Or, a company that monitors its self-service interactions can identify an issue alert customers who may be experiencing similar issues, even if they haven't contacted the company about the problem.
Next gen analytics at work
Recently, a large telecommunications provider wanted to increase call deflection and improve first call resolution (FCR). We analyzed its call volume, taking into account both where customers were in their lifecycle with the firm, as well as their value to the company as determined by product mix owned. This analysis led to several recommendations that would lead to a reduction in total interaction volume, a deflection of interactions to low or no cost channels, and a 50 percent overall estimated annual cost reduction. In addition, an analytic engine designed to predict likelihood of a repeat call was developed, thereby enabling an improvement in FCR.
And a leading automaker's contact center was exhibiting deteriorating NPS performance, which had been identified as the key metric to influence. Data collected from customer surveys determined which customer experience improvements would lead to a significant boost in NPS. Predictive analytics identified prioritized improvements that would lead to an NPS improvement from – 29 to +70 (see the above chart).
Contact centers today are ready to lead a paradigm shift and drive transformative customer experiences for companies and customers. And they will need a strategic approach to analytics to get there.