Smart Tools, Contact Center Metrics and Business Value – Time for a Rethink
Nov 22, 2019
Contact centers have always played a unique role as a barometer of business performance. As a direct interface between a business and its customers, contact center agents are positioned not only to resolve customer issues, but to identify underlying causes of problems and enable corrective action. Consequently, defining appropriate metrics for contact center performance is a strategic imperative.
For example, let’s say an agent encounters a high volume of contacts about an easily fixable issue. The agent who repeatedly resolves the customer issue will be performing well in terms of the traditional metric of Average Handle Time (AHT). However, if the agent communicates the nature of the issue to the business so that it can be fixed at its source, that agent – while perhaps performing less “efficiently” – will be contributing greater value to the business.
Today, the rapid integration of automation, Artificial Intelligence (AI) and machine learning into contact center operations is transforming the relationship between human and digital labor. This transformation is having a dramatic impact on how businesses define metrics to gauge both organizational and individual performance.
At a basic level, the value proposition of intelligent automation tools lies in optimizing the division of labor between machines and people. Specifically, digital tools assume the repeatable, definable, rules-based tasks. People, meanwhile, are freed to spend more time doing the sophisticated reasoning and creative work that (at least for now) only humans are uniquely qualified to do – applying logic and context, absorbing new sources of information to make decisions, etc.
Within a contact center, performance metrics can determine how effectively this machine/human partnership is functioning. Consider, again, the agent productivity metric of AHT. As mentioned, a low AHT has traditionally indicated agent diligence and efficiency in closing sales and resolving issues. In a highly automated environment, however, AHT can show that smart robots are underutilized. If routine tasks are automated, human agents should presumably spend more time on each interaction, since each interaction a human is handling should be complex and require time and special attention. As such, a low AHT suggests that automation opportunities are being left on the table – and that human talents are being squandered on mindless tasks a machine should be doing.
Ongoing analysis of exception rates can help call centers continually extend the use of rule-based automation. By assessing the frequency of specific exceptions, call center managers can define thresholds of when exceptions should be automated. For example, if an exception occurs, say, once a month, it’s probably not economically viable to configure a bot to handle that exception. Instead, it makes more sense for a human agent to address the issue. However, if the exception is occurring several times a week, automating that particular exception might be feasible. By continually identifying opportunities to automate exceptions, organizations can drive ongoing improvement in terms of increasing productivity as well as creating additional bandwidth for human agents to apply their skill sets.
As the adoption of smart technology matures, contact center operations, roles and metrics will continue to evolve. Automation expert, Wayne Butterfield of advisory firm Information Services Group, predicts that over time, the contact center role will become increasingly specialized, value-added and highly compensated. Exception handlers who identify opportunities to expand and enhance automation, for example, will eventually put themselves out of jobs. For contact center agents, ongoing refinement of automation will mean a decreased demand for technical skills, and an increased premium on people skills, creativity and empathy.