call center metrics

3 Contact Center Metrics Improved by Predictive Analytics

Predictive analytics predict future events by combining various techniques that analyze historical and current patterns. Predictive voice analytics can have a major positive affect on integral contact center metrics, including customer retention, follow-up call success and quality assurance.

Customer Retention

One of the customer service industry’s main goals is customer retention, and experts believe that it costs more to acquire a brand new customer than to keep an existing customer. Predictive voice analytics, which analyze the customer’s voice during customer-agent interactions, can determine if the customer is at high risk for ending their relationship with the company altogether. It can then inform the agent that they need to put more focus on retaining the customer. On the flip side, predictive voice analytics can also tell which agents aren’t doing enough to keep the customer coming back. This is more effective than random checking for quality assurance, which can take a long time to identify poor-performing agents.

Follow-up Call Success

Often, the first contact with a customer isn’t the one that has a positive outcome (i.e. a sale); it’s the follow-up call that proves to be more advantageous. However, it’s difficult to know which customers are a priority for follow-up contact. Instead of leaving it up to your agents to determine which customers are worth a follow-up call, predictive analytics can analyze past interactions and study voice features to determine if the customer’s tone and behavior predicts a favorable outcome during the next interaction (like making a payment or finalizing a sale). Predictive analytics can create a ranked list of customers, organized by their likelihood to say “yes.”

Quality Assurance

Predictive analytics are a richer way of assessing quality assurance than traditional methods. Routine QA testing often ignores customer patterns, and it is also unable to learn in real-time. Predictive analytics, however, can analyze all types of data, both structured and unstructured, to give a well-rounded view of agent behavior and how it impacts the customer. All customer-agent communication is assessed in-the-moment, allowing the contact center to get an accurate view of agent performance immediately instead of having to wait several weeks.

Contact centers can’t just gather metrics to assess their current performance and then call it a day. They must also use what they’ve learned from the past to create goals for the future. Predictive analytics can help shape those goals realistically.



Contact Center Dashboard and Metrics

Software is an essential component of a high-volume contact center. From being an important element of sales to managing complex tickets, CRM dashboards and metrics play a vital role in your contact center.

3 Important Types of CRM Dashboards

  1. Key Performance Indicators Dashboard

The KPI dashboard allows the contact center to identify trends using data that’s been collected over a long period of time. Relevant metrics include average seconds to answer, call resolution, cost per contact, and service level.

  1. Call Status Dashboard

The call status dashboard monitors important metrics that can change from minute to minute, including call abandonment, call status, call volume, and current service level. Active and waiting calls is part of the call status dashboard, measuring the volume of calls to the percentage of callers waiting to talk to an agent.

  1. Contact Center Metrics Dashboard

The contact center metrics dashboard provides the analytics required to manage the contact center as a whole, including call status and call resolution. The goal of this dashboard is to help management and agents coordinate when a call “storm” is on the horizon.

4 Chief Metrics to Measure

–Average Handle Time: Measure the amount of time spent per call, including on administrative duties like submitting a call report. This metric can be used to measure both an individual’s performance and the overall performance of all the contact center’s agents.

–Call Abandonment: Measure how many customers disconnect before they’re connected to an agent. According to, if a customer can’t resolve a problem or reach an agent in three key presses, they’re likely to hang up. By monitoring this metric long term, you’ll be able to see patterns that can be fixed with additional staff or technical adjustments. By monitoring this metric in real time as well, it will help you identify problems that are currently happening, which could prevent a small number of dropped calls from becoming a bigger problem.

–Call Resolution: Measure the outcome of calls to determine how well inquiries are being resolved, specifically whether or not they’re being resolved during first contact. The call resolution metric is a barometer for contact center efficiency and customer satisfaction. While there are some problems that may take multiple calls to resolve – like a complex technical issue – this metric’s red flag is the number of current open tickets.

–Service Level: Measure your contact center’s ability to follow through on arrangements made in the service level agreement (SLA). SLAs promise to answer X percentage of calls in X seconds. This must be measured in real time so that you’ll know right away if there’s a problem. Service level can be affected by a number of issues, like an unexpected surge in call volume, unexpected service outages, or a shortage of agents. In some contracts, financial penalties or loss of contract can occur if SLAs aren’t met.