Skip navigation
All Places > Industry News and Best Practices > Blog > 2017 > September


Here we are again - 25 days out from LISTEN. Seems like just yesterday I was blogging about the exciting new venue that would play host to our annual conference for 2016. While our original plans were to relocate LISTEN 2017, the Opal Sands was so ideal for our conference, we are returning once again to the white sands of Clearwater Beach. We had a brief scare with Hurricane Irma but I'm happy to announce that CallMiner's Fort Myers staff, our office, and the Opal Sands (just up the coast) made out relatively well and we are excited to be bringing business back to the area - it's one of the best things we can do to help with recovery.


In addition to hosting at the same great venue, we are maintaining many aspects that made LISTEN 2016 such a success - tracks for executives and analysts, sessions covering fundamentals and more advanced use cases, relational round tables and plenty of networking opportunities. Best of all, we have retained the high level of content presented by you - the attendees and the real users of speech or customer engagement analytics. This year we have great speakers from organizations such as Americollect, Defenders, Mercedes Benz Financial ServicesOtter Box, Sirius XM, Vivint, Thomson Reuters, Bluegreen Vacations, Encore Capital Group, and more.



But in the name of continuous improvement we've made some great additions that you won't want to miss! We have two full tracks for analysts covering both fundamental and advanced topics and use cases. We have great new topics for both executives and analysts such as Stomping Out Fraud with Analytics and Voice Biometrics, Using Analytics to Improve Products and Insights Beyond the Contact Center, Identifying & Understanding the Events of an Interaction, and more. Without taking away from our breakout sessions, we've added more keynotes including CallMiner founder Jeff Gallino's view on the State of Analytics.  Jay Baer, best selling author and founder of Convince and Convert will share his views on Talk Triggers and the role analytics plays in identifying them. Joe Dudek of Quicken Loans will discuss how Corporate Culture can drive a better customer experience and innovation.


One of the additions I'm most excited about is an analytics hack-a-thon in the form of an escape room, dubbed The Great Escape Challenge. Analysts will be challenged to help track down missing secret operative Agent Honey Bee. Teams of six will have 45 minutes to use their Eureka Interaction Analytics know-how to solve a series of puzzles, unlock the clues, escape the room and complete the mission. Be sure to register yourself and your desired team members as space is limited - prizes for the team with fastest time.


playbook-min.pngAnd finally CallMiner will be releasing seventeen Eureka Success Playbooks to attendees. Playbooks are essentially recipes with step by step guidance on how to achieve key results with interaction analytics including numerous ways to improve contact center efficiency, customer experience, agent performance, compliance, and revenue generation effectiveness through increased sales or collections. These plays have been derived from CallMiner's 15 years of experience in the speech and customer engagement analytics space, and will help users achieve substantial return on investment.


Oh - and I almost forgot about the Halloween-themed Karaoke After Party! If you've already registered, we can't wait to see you there. If you haven't yet - what are you waiting for?!

Register here.



Product User Forums




See our highlight reel from LISTEN 2016...

Today’s customers have high expectations for the brands they work with including exceptional customer service. They expect more than a happy and polite representative on the phone. They want to build long-lasting relationships with the businesses they use. If you’re not interested in cultivating customer relationships, they are quick to move on to a company that is. If you want to increase customer retention and establish loyalty, you need to build a consistent positive experience.


Building a positive customer experience starts with measuring your agent’s performance. It is nearly impossible to review every customer interaction on your own. To do so, you would have to hire an extensive quality assurance team. When the majority of businesses are looking to cut expenses, this is not the most effective way to get answers. Instead, many businesses are turning to speech analytics and automated scorecards to measure and manage agent performance. With speech analytics, you can monitor, categorize, tag and score 100 percent of customer interactions. Automated scorecards allows you to consistently and objectively evaluate all agents against the same criteria.


This blog will focus on the scorecard itself- best practices and what you should be measuring.  What is a Scorecard?


Agent scorecards make it possible to see what is happening in every customer communication.  By capturing data and analytics, you have the necessary information to train agents to perform better.


  1. Best Practices - Deciding to implement an agent scorecard system is a big step. It requires forethought and planning to get it right. Here are eight call center scorecard best practices to use in 2017:
  2. Identify Specific Goals - Every call center works with different customers and call types. It is important to identify your reasons for using agent scorecards at the beginning of the process. Are you concerned with compliance, customer satisfaction or agent efficiency? Each requires unique techniques and processes to implement correctly.
  3. Take it slow. - Rushing into any new process can backfire. The purpose of agent scorecards is to create better call outcomes and improve your reputation. Once you know your long-term goals, create a plan that breaks down the necessary implementation steps and information you need to gather for success.
  4. Score 100% of customer interactions. - It is tempting to think you can manually score your agents. The reality is, it’s nearly impossible for most call centers and those that take this approach only end up with information on a small percentage of interactions. Using speech analytics and automatic agent scorecards assures that you will collect data from 100% of customer conversations. Data from all conversations makes it easier to identify trends and outliers.
  5. Embrace the technology. - Automated scorecard technology opens the door to a wealth of knowledge and possible statistics. Don’t try to force your existing process into the technology. If it’s already broken, new technology won’t fix the problem. Instead, review the possibilities and pull what matters most to your goal.
  6. Establish criteria. - Figuring out what to measure is one of the hardest steps in the process since there are numerous things you can measure. The best approach is to establish criteria that match your original goal.
  7. Include call center agents. - Using agent scorecards impacts your call center agents directly, so it is beneficial to include them in the process. By including them in brainstorming and planning, they are more likely to take ownership in the process instead of giving push back.
  8. Monitor and adjust as needed. - Implementing an agent scorecard system won’t fix performance overnight. Monitor the initial results and make changes as necessary. Call centers often overlook is the different types of calls they manage daily. It is hard to score sales calls with the same criteria as service calls. In these cases, centers achieve better results if the customize scorecards to each call type.
  9. Use the information.  - Finally, you need to use the information you capture. Establishing a new process and doing nothing with the results guarantees failure. Look for trends in the scorecards to identify common weaknesses across all agents. Meet with individual agents to review their scorecards and offer suggestions for improvement and hold them accountable moving forward. Use data on calls with the best outcomes to train the entire team.
  10. What you Need to Measure - You may still be questioning what criteria to measure on your agent scorecards. Well-rounded agent scorecards will include all of the measurements listed below. The ones that directly impact your long-term goal should be weighted more heavily. This also shows agents that you are reviewing every facet of their conversation, so they are not tempted to slack on aspects you are not measuring.


Agent Professionalism

  • Politeness
  • Empathy
  • Understanding
  • Insufficient validation

Customer Satisfaction

  • Agitation
  • Silence
  • Stress
  • Keyword mentions


  • Introduces offer
  • Up sell percentage
  • Cross sell percentage
  • Asks for the sale
  • Overcomes objections

Agent Effectiveness

  • Average handle time
  • Silence percentage
  • Number of calls handled
  • First call resolution percentage


  • Greeting number
  • Disclosure language
  • Risky language
  • Closing conversation

Speech analytics software records every conversation and transcribes them into a searchable database. In addition, it uses speech and voice analytics to listen for long silence times, certain keywords you establish, and inflection to identify upset or angry conversations.



When you add automated call scoring to the software, it captures all of these details in near real-time. Performance metrics can be displayed on dashboards for supervisors and even agents to see. For supervisors, trends data can indicate what should be emphasized in training the team as well as individuals. For agents, this data shows them what they need to tweak in their future conversations without management having to micro-manage each situation. 


Final Thoughts

Agent scorecards are quickly growing in popularity. They are the easiest way to analyze agent performance and identify trends that need to be fixed to improve the overall customer experience. Call centers planning on adding call scoring to their processes can ensure success by following  these best practices.


What agent scorecard metrics do you find most useful?


Eureka Essentals iStock_000003106884Small.jpg

Definition of Call Center Statistics

Call center statistics are data gathered about enterprise call centers that empirically illustrate both the internal and external relationships these centers have within the business environment in which they operate. They cover a variety of other areas including technological change, changes in consumer attitudes, and competitive considerations.


This takes a number of different forms and results in a diverse range of situational observations. The most obvious statistics hit on structural realities, such as the fact that nearly 3 in 4 businesses were already expanding their use of virtual agents in their call center teams throughout 2015. However, this is just the beginning. Taken in tandem, this compilation traces the contours of an ever changing, yet vital, landscape and seeks to impose clarity on an area where inefficiencies are all too often - and incorrectly - passively accepted as inevitable.


Examples of Call Center Statistics


Of course, no two organizations are alike. What may be a vital piece of information for one may be obscure if not meaningless to another, just as management best practices will vary from call center to call center. However, the compilation below is intended to be as universal as is feasible. There is a sampling of facts and figures from the key areas mentioned above: structural, technology, consumer attitudes, and competition.



  • ● There is a 37% rate of attrition amongst call center employees during the first 6 months on the job (DMG Consulting)
  • ● 76% of call center operating budgets are tied to resource related expenditures (DMG Consulting)
  • ● Quality management only exists in 3 out of 4 call centers (DMG Consulting)



  • ● It’s estimated that humans will still be required in 1 out of 3 customer service interactions in 2017 (Gartner)
  • ● Cloud-based infrastructure providers now account for more than 18% of contact center seats (DMG Consulting)
  • ● Generation Y already prefers social media to any other customer support contact channel, yet 60% of contact centers have absolutely no social media capabilities (Dimension Data)
  • ● Analytics is believed to be the future of call centers, yet 40% of them have no analytics tools (Dimension Data)


Consumer Attitudes

  • ● Companies realized a 10-15% revenue boost and a 20% jump in customer satisfaction when they made experiences across customer journeys a priority (McKinsey)
  • ● Customer service apps increase the favorable view of a company in 72% of customers (Nuance)
  • ● 2 out of 3 consumers are willing to pay a higher price for excellent customer service (American Express)



  • ● More than 4 in 5 customers told someone when they had a negative customer experience (Maritz)
  • ● 3 in 4 companies view the experience they provide customers as a way to differentiate from their competition (DMG Consulting)


Benefits of Call Center Statistics

Call Center Statistics are a window into the current trends, best practices, and ongoing realities modern day call centers face on a daily basis. They are a compass for both upper and lower management to better understand the role these centers are playing, as well as the impact they are or might have, on ROI.


Beyond their meaning in isolation, paying attention to year over year fluctuations in call center statistics provides companies with valuable business intelligence. They allow for a proper evaluation of macro-level trends, leading to more informed decision making about strategic approach, resource allocation, and the elimination of inefficiencies.


What are Call Center Metrics?

As demonstrated above, call center statistics are an important tool providing valuable macro-trend insight to management. However, this just scratches the surface. Call centers pose a different set of challenges and realities for any given business. While the broader trends are useful, they should in no way be considered monolithic. An internal measurement and auditing of vital metrics is necessary in all cases.


These internal observations complement the broader scope and utility of call center statistics. When used together properly, they allow internal trends and processes to be monitored or even benchmarked against ongoing macro-level trends to ensure maximum efficiency and effectiveness at any given time.



Average speed of answer is, at the most basic level, about running an effective call center by finding the fastest path to having customers’ questions answered or issues resolved. This means understanding the metrics that need to be monitored, transcribed, and analyzed in order to glean actionable insights.


Average speed of answer is one of the most important metrics for call centers to measure. The concept is closely tied to (and often confused with) those of average handle time and first call resolution. However, there are important differences between them. The most critical is to understand that average speed of answer is all about understanding the needs of customers and being able to provide them answers quickly.


How Do You Measure Call Center Average Speed of Answer (ASA)?

Average speed of answer is defined as the average amount of time it takes for a call center to answer a phone call from a customer. Included in this metric is the time a caller waits in a queue. The time it takes to navigate through an IVR system is not factored in to ASA.


For this reason, measuring ASA requires a nuanced approach that ensures maximum accuracy. Here are two important tips for calculating it correctly:


Calculate the Average Properly

In its simplest form, ASA is calculated by:

ASA = Total Wait Time for Answered Calls/Total # of Answered Calls.


The idea behind ASA is to get an overview of general performance. For that reason, one of the most common mistakes made is to simply take an average of the aggregate data. But approaching the calculation in this manner will include outlier data points that can skew results. Be aware of this, and make sure to account for the effect of outliers when drawing conclusions from the measurement.


Customer Abandonment

Average speed of answer in isolation doesn’t give any information about the impact of the time frame necessary for a response. To make up for this blind spot, be sure to look at customer abandonment rates as well. Even if the average speed of answer seems reasonable, it will need to be improved if there are still high customer abandonment rates.


Why Is Average Speed of Answer Important?

Average speed of answer is important because it gives call center staff the information and tools they need to their jobs effectively. Customers value their time, and so an understanding of what they are experiencing when they call in is the first step to making them happy and improving overall customer satisfaction. If a particular trend is spotted that indicates an area where more could be done, it empowers management to provide better training and coaching for staff so an answer is readily available when it arises.


Another reason average speed of answer is important is its relationship with interactive voice response (IVR) systems. Because IVR works by leading a caller through a series of menu options, average speed of answer can be used to gauge the effectiveness of menu options. A well-constructed IVR will keep response times low and get a customer to the agent most equipped to answer their question, while a system which is poorly designed will lead to higher wait times and less targeted agent responses.


Speech analytics is one technology that cannot only assess ASA and other performance metrics, it can also detect issues with IVR routing and identify additional routing options.


Average Speed of Answer & The Customer Experience

Bottom line, average speed of answer is all about a call center’s ability to get a customer's issue resolved as quickly as possible. Understanding how long it is taking customers to get to an agent is at the heart of the value behind the metric.


However, it’s not enough to simply take an average. Proper measurement should consider outliers. It should also be concerned about the customer experience across their entire journey.