Speech analytics is the process of converting contact center conversations into text, analyzing that data, and gathering insights into customer interactions. When set up correctly, it can be a powerful source of information for driving highly actionable insights that benefit the organization in tactical and strategic ways.

At a high level, the uses of speech analytics can be grouped into

  1. Compliance monitoring
  2. Enabling tactical improvements
  3. Analyzing the root cause of descriptive metrics
  4. Improving business outcomes

As an organization matures in its use of speech analytics, it will go through a progression of integrating speech analytics into its overall analytics practice. The above four areas represent that progression in order of increasing complexity. Let’s take a closer look at each of these areas.

Compliance & Script Adherence Monitoring

Call centers are required by law to act in compliance with the rules defined for their industry. Failure to do so can result in stiff fines for the company. Speech analytics makes it easy to monitor for compliance by making it possible to analyze the full text transcripts of all calls. While all speech analytics tools allow for analysis of the full text transcription, tools such as Genesys Interaction Analytics go one step further and are able to differentiate between agent and customer audio, making it easy to focus on agent side interactions.

Another simple application of speech analytics is to ensure agent adherence to scripts. For example, if a cable customer is calling to get help with her DVR, an agent may be required to follow a script to ensure that the most basic troubleshooting activities are performed before the call can be escalated to the next level of support. Failure to do so can result in increased costs and negative impacts on call center performance metrics. Using speech analytics, contact center managers are able to measure adherence to scripts and issue subsequent training when they find non-compliance.

Tactical Improvements Through Call Analytics

Once managers master the basics of speech analytics, they are able to leverage it to address changes in agent level metrics such as an increase in average handle time (AHT) or reduction in first call resolution for groups of calls. For example, if a manager notices an increase in average handle times for a specific type of call, she can analyze the call text to look for changes in patterns. By segmenting the data, the manager will be able to isolate the root cause for the increase in AHT and address it by taking appropriate corrective measures. Similar actions can be performed to investigate and address changes to a number of metrics such as customer satisfaction, call transfer rates, revenue per call, etc.

Another common use of speech analytics is to identify call drivers impacting performance and adjusting IVR options, call scripts, or even business policies to improve the customer experience. For example, speech analytics could identify customers calling in at high frequencies about changes in flight schedules. Once the root cause for the increase in calls has been identified, managers can then adjust IVR options to provide quick access self-service options for these customers.

Root Cause Analysis of Descriptive Metrics

Organizations use descriptive analytics to measure performance at the product, business unit, or department level. For a contact center, those analytics include metrics such as total calls handled, transactions, revenue, customers retained, etc. These metrics measure the performance of a business unit at a macro level. Because all these metrics are numeric, they are not able to explain the reason behind the change in metrics. For example, consider a contact center where customer retention rate dropped by a few percentage points when measured month over month. While the change may be small in terms of percentages, it could represent a large change in actual revenue or customers saved. When such small changes occur, it becomes difficult to pinpoint the root cause for the change. Speech analytics makes it possible to dig into the data to a qualitative level where the reasons for losing the customers can be understood.

This application of speech analytics is more advanced because it requires joining caller data with business performance data using unique keys. Often such exercises require normalizing the data and storing it in data warehouses where it can be connected with data from other systems. As complicated as it sounds, combining data from previously disjointed data sources often results in highly actionable insights.

Improve Business Outcomes:

Predictive analytics is the practice of using statistical models to predict business performance results. In the most advanced form of leveraging speech analytics, analysts can identify patterns in calls to predict future actions and take preemptive measures to address them. Armed with this analysis, the organization can make an informed decision on what efforts can increase or decrease customer satisfaction.

Summary:

Speech analytics is a very powerful tool that can be used to methodically improve contact center performance along with customer experience. As an organization matures in its use, speech analytics can be combined with other data sources to deliver powerful insights that can shape business outcomes


0 Comments

Leave a Reply

Avatar placeholder

Your email address will not be published.