Closing the Metrics Gap with Agent Focus
With the plethora of agent performance data available today, it can be challenging to determine where to focus. That doesn’t apply to agent performance in isolation; it also applies to how that performance affects customer satisfaction.
For years, Workforce Optimization (WFO) applications have tracked agents with tools such as quality, performance, and desktop management. As artificial intelligence (AI) becomes more integrated into these and other applications, the data that emerges because more useful and insightful.
Nearly three-quarters of companies are suffering from a Metrics Gap, meaning they are falling short on adequately measuring and using key data points that could dramatically improve their customer experience. To fully benefit from customer—and agent—data, CX leaders must adopt a lifecycle approach toward identifying metrics, gathering the data, analyzing it, and taking action. At this point, only 26.2% of companies have adopted such an approach, according to Metrigy’s Customer Experience MetriCast study of 1,846 organizations globally. I have written a detailed report on this topic that you can find here: https://www.five9.com/resources/report-metric-gap.
Two main uses of agent-facing analytics:
Agent Assist – AI-enabled virtual assistants help agents in a variety of ways in real-time while they are interacting with customers.
Agent Analytics – Agent performance data shows where they are exceeding Key Performance Indicators (KPIs), and where they are lagging. Taking this data and pairing it with Voice of the Customer feedback can drastically strengthen agent performance and overall customer satisfaction.
The primary goals of agent assist platforms are to empower agents and make them more productive. While empathetic live agents can inject creativity and leadership into a customer interaction, AI rapidly computes and advises based on real-time and historical data.
For example, AI-empowered agent assist applications can do the following:
Listen to the conversation and deliver real-time transcripts for the customer data record, the customer, or context for future interactions. Meanwhile, the agents don’t have to spend time after the call recording notes or next steps—and during the call, they can be more attentive knowing AI is taking notes.
Provide real-time guidance on how to handle a customer. Suppose an investment advisor is about to recommend a particular stock or fund. The virtual assistant can track real-time macro-economic data with a screen pop that may suggest a different investment based on emerging market trends that the investment advisor isn’t seeing because she is talking to her client.
Validate agents are running through a pre-established checklist, by checking off the items for the agent once Natural Language Processing detects the agent speaks the script or offer.
Deliver reports containing relevant insights from the discussion. For example, does sentiment analysis detect a given agent’s customers are always happy or angry? Is the agent taking the virtual assistant’s advice, and either way, what is the outcome? This provides guidance on whether the virtual assistant is actually helping.
When contact center supervisors have AI-enabled eyes and ears on every agent interaction, the sky’s the limit on what type of data they can gather and use to effect change. The best part is that agent performance data doesn’t have to be reviewed only weekly or monthly; it is evaluated in real-time so improvements can happen immediately.
For example, agents may see that they are struggling with a new upsell campaign by viewing their own real-time analytics compared with their colleagues. At the same time, the virtual assistant may start making new recommendations in real-time based on what others are doing successfully that they are not. Or, AI may pick up on a problem that would have taken weeks to uncover, such as an irrelevant message for a specific geographical region.
Though there is plenty of analytics coming out of agent assist platforms, there is even more data available to help CX leaders evaluate individual and team performance.
CX leaders have long used dozens of KPIs to gauge contact center performance. Common KPIs include Call Handle Time (CHT), First Contact Resolution (FCR), and time in queue. But as agents are getting more upsell and cross-sell responsibilities, CX leaders are adding new metrics, such as sales performance to goal. Or, they are measuring success on each interaction channel and getting creative with correlations. For example, companies are correlating the average CSAT score with the number of simultaneous chats underway to pinpoint the maximum number of chats each agent can handle before performance degrades.
When leveraging metrics to their fullest, companies feed real-time customer feedback to agents. Agents who have received a morning full of only three-star ratings need to evaluate what’s happened and make changes in their approach so their afternoon shifts to five-star ratings.
All of these metrics help improve performance more quickly than ever if they are delivered to agent desktops. By analyzing performance and coupling it with gamification, agents are paying more attention to their performance—which ultimately improves customer satisfaction.
Get the big picture by downloading the Closing the Metrics Gap infographic.
To learn more about these metrics are and how you can close the metrics gap, download Metrigy’s free report.