Artificial intelligence (AI) has been making its way into call centers for some time now, handling everything from removing sensitive customer data mentioned during calls to providing call center agents with suggested phrases and upsells to use while talking to customers. Speech analytics in the call center is one of the most innovative applications of the technology.
How is speech analytics used in call centers?
Speech analytics software is one of the most exciting forms of AI in call centers, allowing agents to record calls, classify phrasing to detect sentiment, and analyze data to improve customer service. Using this software isn’t as simple as connecting it to a system and letting it go, though. Instead, call centers must take steps to understand how this powerful software works, the many ways they can use it, and how to fully integrate it within the company.
We’ve compiled this list of tips, benefits, and best practices for speech analytics for call centers from industry experts to help companies in all industries make the most of their speech analytics software.
Speech analytics call center tips and best practices
1. Learn exactly what customers need based on what they say. “Gaining customer insights is one of the most common use cases for speech analytics. There are many valuable insights hidden in the unsolicited feedback in customer interactions, and speech analytics can help you extract that value and better understand your customers, their needs, their sentiment about your products and services, and what motivates their behavior.” - Speech analytics 101: What is speech analytics?, CallMiner; Twitter: @CallMiner
2. Trigger important actions automatically. “Real time speech analytic processing can be used to identify trigger words (or words that require special handling). For example, a sales call center may use speech analytics to monitor for words like ‘returns’ or ‘attorney’ or ‘attorney general.’ When those words are identified, a supervisor may be alerted or a special script may be presented to the agent to handle the situation properly.” - Speech Analytics in the Call Center, Call Center Advisor
3. Avoid buying speech analytics software with features you don’t need. “A speech tool’s standard features and functions should be clean and intuitive and satisfy the bulk of any user’s functionality needs. With the right tool, beginning users should be able to generate basic but powerful insights within the first week of deployment.
“That said, more sophisticated features can generate additional insights—but only if the user has been trained to operate them properly and interpret the results accurately. I’ve seen too many buyers line up the features of different vendors’ offerings and base their decision on the number of ballot boxes that can be checked.
“Don’t let bells and whistles divert your focus from the performance of core functions. Just because a particular feature is technically possible doesn’t make it practical.” - Scott Bakken, Tips on Selecting and Deploying Speech Analytics, Contact Center Pipeline; Twitter: @CCPipeline
4. Measure agent performance with speech analytics software. “Speech analytics automatically detects agent activities and behaviors that lead to successful interactions by automatically tracking and assessing calls.
“On the other hand, the technology allows managers to see specific difficulties contributing to poor agent performance patterns. Agents could work more efficiently as a result and request help where needed.” - 10 Speech Analytics Call Center Use Cases, CX Today; Twitter: @cxtodaynews
5. Don’t forget your other customer communication channels. “If you measure customer experience on only one channel, you’re missing out on an enormous amount of useful information. It’s essential to have a complete picture of the customer journey.
“By analyzing customer interaction across every channel, you can ensure that you don’t miss any useful feedback. In fact, customers are already giving their feedback and advice on different media. Most businesses merely aren’t listening.” - Evelyn Johnson, 6 Top Ways We Should Use Speech Analytics, Dataversity; Twitter: @dataversity
6. Audit speech analytics systems frequently. “Before deployment, review a sizeable sample of calls and associated transcriptions (approx. 300 to 500 random). This allows you to gain a sense of how accurate the transcription is and helps identify what keywords or phrases need to be added or tweaked. This is not a one-time activity! Commit to ongoing, periodic auditing and tuning.” - Paul Tessier, How to Implement Speech Analytics – Actionable Tips from Top Industry Experts, CustomerThink; Twitter: @customerthink
7. Don’t be afraid to boost your investment in speech analytics if it’s working. “Speech analytics is an investment – scope your projects so that the costs are more than justified by the returns. Build a pilot program that has proven return.
“If the return is good, then you will want to put more money into it in order to gain the greater benefits. Look for suppliers who can support a phased approach and offer you a choice of cost models.” - Jonathan Wax, How to gain measurable benefit from Speech Analytics, Professional Planning Forum; Twitter: @ForumOnline
8. Extract data to improve self-serve options. “Speech analytics can indicate the percentage of unsuccessful self-serve calls, break down those unsuccessful calls by category, and determine the percentage of calls in which the agent educated the caller about self-serve options. Companies can use the resulting insights to improve specific self-serve options that many callers have found problematic.” - From speech to insights: The value of the human voice, McKinsey & Company; Twitter: @McKinsey
9. Keep your compliances in check. “One of the key benefits of speech analytics software is its ability to automatically score and analyze 100% of calls, resulting in improved agent compliance and reduced risk of fines and lawsuits associated with non-compliance.
“But it’s much more than that. While manual sampling of recorded calls or contacts provides little to no prevention of non-compliant behavior, speech analytics software tracks every call for violations and risky language.
“Real-time call monitoring solutions such as EurekaLive take this one step further, tracking specific language and acoustic characteristics while the call is still ongoing. This allows managers and supervisors to review occurrences of compliance infractions and intervene when necessary.” - 3 Ways Speech Analytics Can Improve Call Center Compliance, CallMiner; Twitter: @CallMiner
10. Remember that accuracy isn’t the sole metric on which to base the success of speech analytics software. “Perhaps one of the biggest misunderstandings regarding speech analytics is that transcript accuracy is the sole measure of the solution’s effectiveness. Yes, accuracy matters. But so does recall or completeness.
“Recall measures the number of times a searched word or phrase is used. It’s just as valuable as accuracy, especially when speech analytics is used for risk mitigation or compliance and when catching every instance of a word or phrase matters.” - Speech Analytics: The Myths, The Truths, And Best Practices, USAN; Twitter: @USAN
11. Use screen recording and speech analytics simultaneously. “Screen recording allows the call center to record or capture images of what a call center agent is doing during a call.
“This can be particularly useful in regulated environments (e.g., such as when PCI or HIPAA is a concern). It can also be useful for training and quality assurance purposes. A trainer can review not only how the agent sounded on a call but also what the agent did during the call.
“Good screen recording and voice recording software work together, allowing a trainer, supervisor or agent to replay a call with the audio synched to the video of what the agent was doing.” - Call Center Recording, Call Center Advisor
12. Consider echo cancellation techniques to improve speech recognition and analytics. “Echo cancellation is used in telecom and telephony applications to improve the quality of voice communication. It prevents echoes from being created or captured. It can also remove the echo in post-processing.
“Echo cancellers operate by first analyzing the incoming voice stream and simultaneously monitoring the return voice stream before it’s being transmitted. If the echo is detected, the echo cancellation engine subtracts it from the transmitted or received audio signal.
“Here are some of the basic steps you can take to prevent echoing during voice and video calls.
- Wear headphones: One of the fastest and easiest ways to eliminate echoing is by simply using headphones. Since there is no sound coming out of the speakers (as the speakers are in your ears), the microphone is not likely to pick it up which avoids echoing. That can be applied to both one-to-one video calls and group conference calls.
- Adjust your volume level: When your speaker volume is too high, your microphone can pick up the voices of other call attendees causing an echo that you might not even be aware of. Turning down the volume and properly adjusting your microphone sensitivity settings ensures your microphone will only pick up your own speech.
- Enable half duplex communication: Another quick way to eliminate echoing is by using half duplex communication, where only one speaker’s audio is transmitted at a time. That means having all the participants of a call mute themselves when they are not speaking. The downside of this approach is that the call could not feel like a natural, uninterrupted conversation.” - What is Echo Cancellation?, LiveAgent; Twitter: @LiveAgent
13. Fill out your pages with common questions your software detects.“As your offerings and the market evolve, callers may pose new questions about your solution. Ask agents to keep a list of questions they hear often, or survey a random sampling of recordings to identify queries not currently addressed through your existing script or FAQs.
“Encourage agents to add these questions to your knowledge management platform so that when other agents hear a question that’s new to them, they can quickly check the platform to see if it’s ever been asked.” - Madelina Jacobson, How to Improve the Customer Experience in Your Call Center, Bloomfire; Twitter: @bloomfire
14. Share the data wealth with other departments, too. “The data speech analytics captures isn’t exclusive to one area of your business. There is plenty of information to support other department efforts as well. After your testing and auditing are complete, and you see progress toward your goals, you can look for trends that can help other departments outside of your contact center.
“How can the data you track support product management, marketing, sales, human resources, and operational efforts? When you find ways to improve more than one department’s processes, you increase the value of speech analytics. It will require more work and a pitch to get other departments in line with using the software, but the value it delivers to the customer experience, and bottom line can be very convincing.
“Some businesses choose to align all departments at the beginning of the speech analytics launch. Which path you take depends on your company goals and whether upper management is supportive of a broader effort.” - 10 Best Practices You Need to Make the Most of Your Speech Analytics (PDF Download), CallMiner; Twitter: @CallMiner
15. Consider data reliability. “The primary factor that goes into choosing the right speech analytics software is the reliability of its data. Data reliability, in this case, would include the accuracy and completeness with which conversations are analyzed and stored.” - Find the best Speech Analytics Software, Software Advice; Twitter: @SoftwareAdvice
16. Help agents understand that analytics tools are there to help, not harm. “Emphasize to your agents that this tool isn’t intended as a ‘gotcha’ device. Your agents will likely ‘sit up straighter’ knowing that all calls are being analyzed, even while they harbor suspicions about the new speech tool.
“Take pains to assure agents that speech analytics will impact them positively—both in identifying ‘champagne moments’ to celebrate and helping build cultural trust within your organization.
“Consider building the first speech project based on a positive business issue while identifying other process and performance issues for future projects that you can roll out once trust is established. If at some point constructive criticism is necessary, be sure to deliver it respectfully to maintain morale and head count.” - Scott Bakken, How to Implement Speech Analytics – Actionable Tips from Top Industry Experts, CustomerThink; Twitter: @customerthink
17. Build a strong analysis team to optimize your strategy.“Leveraging the full power of your speech analytics tool requires performing a variety of administrative, analytical, and fact-finding tasks on an ongoing basis. The learning curve can be steep and confusing without the presence of internal or external speech analytics experts who know how to apply operational subtleties and strategies for capitalizing on raw data.
“Whether you tap internal resources or rely on outside consultants to manage your speech analytics initiative, maximizing the technology’s capabilities requires certain skill sets. Start with the stakeholders in departments that will be impacted, and build a team of inquisitive and curious analysts who have a passion for analytics.
“If you choose to manage the process from within, your ideal candidates will be able to toggle between left-brain (logical) thinking and right-brain (creative) thinking. Individuals who can synthesize speech analytics data from both viewpoints will be able to analyze and articulate business opportunities and make the case for necessary changes. These whole-brain thinkers are a bit of a rare breed, so here are minimum requirements for three essential skill sets.
- Administrator. The Administrator is responsible for deployment of the speech analytics tool and manages all server and database connections.
- Business Analyst. The Business Analyst is responsible for executing all the analytical tasks necessary to compile, analyze and deliver key findings of speech analytics projects to upper management.
- Interactions Monitoring Analyst. The Interactions Monitoring Analyst is responsible for listening to and reviewing the content of selected calls based on criteria provided by the Business Analyst.
“Depending on the depth of your internal resources, one staff member could perform multiple functions—but it’s best to train two resources in the event you’re unable to retain your lead.” - Scott Bakken, Tips on Selecting and Deploying Speech Analytics, Contact Center Pipeline; Twitter: @CCPipeline
18. Leverage speech analytics for up-selling and cross-selling opportunities. “Contact centers may be able to use speech analytics to identify opportunities to upsell or cross-sell to boost revenues. This software may also pinpoint the types of responses that specific customer demographics provide and make changes or improvements as necessary. Contact center speech analytics can help agents and supervisors create customized upsell or cross-sell tactics that are designed to target each customer’s needs.” - Speech analytics: 6 benefits for your contact center, RingCentral; Twitter: @RingCentral
19. Define specific use cases for speech analytics data. “These could range from targeted coaching, automated quality assurance, understanding customer sentiments, managing the workforce, reducing fraud, reducing collections, or increasing sales. One internal help desk provider defined ten use cases that helped it unlock 20 to 30 percent cost savings and a customer-service improvement of more than 10 percent.” - From speech to insights: The value of the human voice, McKinsey & Company; Twitter: @McKinsey
20. Choose speech analytics solutions that offer sentiment analysis. “Customer sentiment analysis is a tool that scores each customer interaction as positive or negative by analyzing certain keywords and phrases used by either the customer or rep over the course of a conversation.
“Historical sentiment analysis gives managers insight into customer service level and support over a specific time period, while real-time analysis lets supervisors jump on live calls and save business when needed.
“Sentiment analysis uncovers the root causes of customer complaints or dissatisfaction, making quick improvement possible.” - Rebecca Drew, Call Center Analytics: How to Effectively Analyze Call Data, GetVOIP; Twitter: @getvoipreviews
21. Understand the nuances of different privacy laws for compliance.“Privacy laws, such as the California Consumer Privacy Act (as amended by the California Privacy Rights Act that went into effect on January 1, 2023, (“CCPA”)) in California, other state privacy laws coming online in 2023, and the (UK) GDPR, require companies to notify data subjects of the purposes for which they are using data.
“Some uses of AI voice analytics may not be covered by existing privacy notices. Also, depending on the data collected and the use case, these privacy laws may require consent from the customers, mandate that customers be provided with opt-out rights or the right to limit the use of that data, and/or limit the ability of the company to share personal data with third parties, which may include a vendor who is providing voice analytic services. Additionally, data subjects in these jurisdictions will likely have rights with respect to this data—such as the right to access and deletion.” - Legal Risks of Using AI Voice Analytics for Customer Service, New York University School of Law - Program on Corporate Compliance and Enforcement; Twitter: @unylaw
22. Leverage speech analytics for root cause investigation. “Finally, speech analytics lets you capture impromptu feedback from calls, even before the agent has a chance to intervene with a formalised questionnaire or survey. For instance, if the customer says “it was very difficult to have…,” you get clear and honest feedback into a genuine problem area. In cases where the formal CSAT is unusually low, speech analytics aids in root cause investigation.” - Five Ways to Use Speech Analytics in a Contact Centre, CX Today; Twitter: @cxtodaynews
23. Understand your business needs first. “Before you explore the different options available, you must be clear about your own business needs, budget and preferences. Only then should you assess the software on the basis of their capabilities, flexibility, support and cost.” - Shweta, Call Center Analytics Guide (2023), Forbes; Twitter: @Forbes
24. Use speech analytics to reduce average handle time (AHT). “Many operations are eliminating wrap time almost entirely with speech analytics, automating timely processes such as:
25. Reduce overhead costs with speech analytics. “Most companies view speech analytics as an investment that helps in saving costs in the future. Top areas where such software could reduce the overheads are:
- Automated call monitoring and compliance checking reduce the number of employees required for auditing and quality check.
- Automatic compliance reduces damages and monetary losses due to non-compliance.
- Speech analytics software can analyze the common reasons why customers call, thereby allowing the agents to troubleshoot the problem at the root. For instance, a small glitch on the website might be increasing the number of customer calls. In-depth analytics software can identify keywords in the call audio to determine the call's reason. Thus, the user can rectify the glitch, reducing the call volume, thereby reducing the employees' work pressure.
- Better decision-making would increase call resolution rates, reducing the number of callbacks, complaints, and escalations.
- Since it becomes easier to understand customer needs and personas, the executive can quickly satisfy the requirements and end the call within a shorter time. This reduction in call time reduces resource requirements.” - Jone Samy, Benefits Of Speech Analytics For Customer Support Call Centers, eLearning Industry; Twitter: @elearnindustry
These expert tips and best practices will help you leverage speech analytics to improve operations throughout your call center.
Frequently asked questions
Speech analytics is a technological process that monitors live calls or call recordings for numerous tasks, like improving customer service or pulling words from conversations to inspire future marketing campaigns.
Speech analytics performs multiple functions in call centers, including:
- Marketing to customers: Speech analytics software can prompt agents to recommend products or services to customers at the right times based on the flow of conversation.
- Improving customer service: Speech analytics can pull out words and phrases to help determine sentiment analysis, which gives companies insight into how a customer feels about the service they receive.
- Training agents: Speech analytics can detect potential issues with agent communication or resolution, prompting supervisors to intervene.
Speech analytics specifically looks for words, phrases, and language used in a conversation. In contrast, voice analytics picks up nuances of conversation, like rhythm and tone. However, the two are very similar and often used together, so some companies use the terms interchangeably.