How Does MiaRec’s Automated Quality Management Solution Work?

When deciding on a quality management solution for your contact center, you wonder how you can use MiaRec’s Automated Quality Management (AQM) solutions to get the best ROI.

At MiaRec, we customize our AQM solutions to meet contact center demands across industries, including healthcare, finance, retail, and more. We believe in catering our AQM solutions to your contact center’s specific quality assurance needs.

In this article, you will learn how MiaRec’s AQM solutions can streamline your QA processes. Discover how you can use MiaRec Auto Score Cards to improve agent performances and customer experiences. 

The Importance Of Automating Agent Evaluations With MiaRec Score Cards

As part of your quality assurance processes, you must assess how agents interact with customers. Measuring agent compliance, competence, and other key performance indicators (KPIs) ensures your contact center agents are providing meaningful customer experiences.While you can try to grade interactions manually by listening to every call, this process is incredibly time-consuming—most contacts with a manual agent evaluation process are only able to grade 1 to 5% of their calls. Instead, with MiaRec Auto Score Cards, you can automatically score 100% of your calls and quickly identify customer interaction issues and training gaps. By scoring all of your calls, you gain a more comprehensive insight into agent performances and how to improve them.

How Does MiaRec Automated Agent Evaluations Work? What’s Needed From Contact Centers?

MiaRec Auto Score Cards uses Moments Query Language (MQL) to analyze call transcripts and grade agent performances according to your scorecard criteria. Throughout the onboarding process, MiaRec works hand-in-hand with your team to customize scorecards, reflecting your contact center’s specific standards. We pride ourselves on being agile and flexible, adapting to your needs every step of the way. If you prefer to customize your scorecard independently, we will provide free manuals and open documentation, available on our website, to guide you through the process. You will need engineers who have experience writing expressions using syntax.

MiaRec uses query language to search for specific phrases in your call transcripts. With query language, you can make your search statements as nuanced as you need. For example, when asking “Did the agent say their name”, you can include specifications such as POSBEFORE:50 AGENT: (“my name is” | “this is”). This notifies MiaRec Auto Score Card to focus on whether the agent said either “my name is” or “this is” within the first 50 words of the call transcript. This is just one example; you can modify and adapt expressions according to your scorecard question.

Image: Screenshot of a MiaRec Score Card. This question specifics that an agent needs to say their name in the first 50 words of the conversation. 

Image: Screenshot of MiaRec transcript, which confirms the agent verified customer's name and telephone number. 

Once you have finalized your Auto Score Card, MiaRec can begin grading calls. MiaRec analyzes your call transcripts and looks for keywords and key phrases to answer your scorecard questions.

MiaRec Score Cards are evaluated on a point system. MiaRec Score Card tracks how many points a question is worth and how many points you earned per question. 

In the example below, the question “Did agent say their name” is a yes/no question, is worth 10 points, and is not weighted. This means that, if the outcome is yes, your agent can earn 10 points at maximum.

Image: Screenshot of MiaRec Score Card. “Weight” and “points” determine how much one question can impact the agent’s evaluation score. The default “Weight” is always 1, which indicates it is not weighted.

In order to calculate your agent’s call score, MiaRec calculates the total sum of how many points the agent earned per section and divides that by the maximum number of points they could have earned on the Scorecard.

In the example below, MiaRec looked at how many questions the agent got right per section (e.g. 0 points in the greeting section, 0 points in the verification section, and so on), totaled them up, and divided the number by the total number of points the agent could have received (80 points). Thus, 10 points/80 points is 12.5 points, rounded down to 12. 

Image: A screenshot breaking down a MiaRec evaluation score. 

By default, questions are given a weight of “1”, which indicates they are not weighted. If you decide to change the weight of a specific question, it will influence the question’s total point value. 

A question’s point value will be multiplied by its weight. In the example below, the question’s weight was modified to 2, and its outcome value is 20 points. We would calculate the question’s final point value by multiplying its outcome value by its weight. Thus, this question is now worth 40 points (20 points x 2 = 40 points).

Image: Screenshot of MiaRec Score Card. 

Your scorecard questions are not all equally important. For example, while your agents should introduce themselves, it is crucial that they follow your contact center’s compliance regulations. MiaRec Score Card gives you the option to customize your question’s importance by modifying its outcome value (points) or weight.

For example, let’s pretend you have a scorecard with 10 questions, all worth 10 points. If there is a particular question that is more important than the others to get right, you can make that question worth 20 points instead. 

Alternatively, you could have a scorecard with 10 questions. Each question is worth 10 points and they all have the default weight of “1”. If you want to make a specific question worth more than the others, you can set that question’s weight as “2”. The question would then be worth 20 points (10 points x 2 = 20 points).  

Image: Screenshot of sample scorecard questions. You can customize and add your own questions.

How Accurate Is MiaRec’s Automated Call Scoring?

Because MiaRec uses your call transcripts to grade scorecards, call scoring accuracy depends on the transcription’s accuracy. While our Speech-To-Text transcription services are as accurate as any leading transcription engine on the market, there is currently no transcription engine that can guarantee 100% accuracy. This is due to factors such as poor cell signal or speech disfluencies. Speech disfluencies are any factors that disrupt the flow of speech. They are caused by speakers; some examples of speech disfluencies are stuttering, filler words, and improper pronunciation. MiaRec is constantly training our STT transcription models to overcome these challenges and provide more accurate transcriptions.

These factors can then affect Auto QA scores. For example, an agent may have said, “Thank you for calling.” However, there was inconsistent signal throughout the call, which meant the sentence was only partially transcribed. This causes the Auto Score Card to incorrectly believe that the agent did not say “thank you.” 

While these cases are infrequent, we offer the option to override automated call scoring grades. If the Auto Score Card scores a question incorrectly, you can overrule its decision.  

Screenshot of MiaRec Agent Evaluation. The override option is on the right, inside the red box.

How Does MiaRec Auto Call Scoring Improve My Quality Assurance Processes?

By scoring 100% of your calls, you will have a more comprehensive understanding of your agent performance. Let’s pretend your agents consistently fail to answer a particular compliance-related question. Rather than have agents redo an entire training course, you can use your call-scoring insights to create training sessions that target specific pain points. 

Your MiaRec Auto Score Cards can highlight call script efficiency. As you review your scorecards, you may notice that multiple agents always escalate calls whenever a customer has questions about a return policy. You can modify your call script to answer common customer questions, reducing call escalations. These are only some of the ways you can improve customer interactions with MiaRec Auto Score Cards. 

Conclusion

MiaRec AQM solutions scale quality assurance processes so that you can get a 360° view into your agents’ performance. 

Now that you know how automatic scorecards work to better customer experiences, you can begin creating your own. MiaRec will help you every step of the way in creating a comprehensive plan that will streamline customer calls and gather personalized training initiatives to improve call script adherence, new hire performance, and customer service standards.

Contact our sales team for a personalized demo on how you can use MiaRec’s Voice Analytics & AQM solutions to transform your contact center.