What does your Quality Assurance (QA) process measure?

When you call customer service whether it be for your bank, your favourite clothing store or cable provider, you often hear the message, “This call is being monitored for quality assurance purposes”. You probably hear this message so often you block it out and don’t even consider its purpose. Within a call center, QA is an important process of any successful operation; for instance, it heavily impacts customer retention. To add more insight to QA, we got the chance to talk to Fred Shadding, Founder & Chief Advisor of The Call Center Source, about the biggest challenges, metrics, and the future of QA.

Why is quality assurance (QA) important in call centers?

Fred: In today’s competitive business landscape, the voice of the customer (VOC) and the manner in which it impacts how companies design, execute and deliver their products to market is a critical element to customer success. One of the key strategies in both assessing the reaction of VOC feedback and, enhancing how customers are handled in the call center is the use of impactful QA processes.

QA is generally designed as a procedural check and balance within the contact center to ensure that the customer experience and agent proficiency is delivered in the most effective and results-driven manner. Thus, the importance of a strong and flexible QA process is, in fact, paramount to the success of well-run centers.

What are some of the biggest challenges you see with QA today?

Fred: The primary challenge I observe is how companies determine the right mix of integrating customer feedback and reaction within the QA approach. It is a bit of a balancing act as companies seek to be sensitive to their customers’ needs yet, must maintain a service discipline, which can be repeated effectively.

QA really depends on the type of product, service, or customer that you’re working with.  That’s always the challenge with it because there is no cookie-cutter approach. Gone are the days when you have a standard QA checklist with 10-15 different criteria that you are evaluating the agent on and based on the weight of those scores, it is determined whether agents are performing on par. Due to the dynamic impact of poor customer interaction such as unfavourable social media posts and exposure, you have to augment it with some of the new tools that are out there like speech analytics and other metrics that can further analyze the tone, impact and the flow of the call.

Most leading support teams now incorporate external and other analytics in their overall assessments of agent proficiency such as NPS, CSATand other online reputation management metrics.

What are some of the QA metrics or KPIs you recommend?

Fred: I would say if there is a metric that provides a tangible measure of the call’s outcome then I would start there. To highlight, on a customer service call, metrics that are of importance would be CSat, NPS and sentiment (if a speech analytics application is in use). These metrics typically focus on the outcome and result of customer contact. This is important from a QA perspective, as it will initially signal if the call is viewed as successful or, may require further assessment. For a sales call, average speed of answer, closed sales ratio and abandoned calls would be useful to track as they would suggest collectively how well sales transactions are handled (i.e. are their calls answered promptly (ASA), were we able to close the sale (close ratio) and did they have to wait long (abandons).

Another metric I quite often see relates to scripting adherence, but I don’t like that as much because it tends to reinforce the practice of making the whole conversation a bit robotic. Certainly, scripts are necessary and in certain instances when supporting various industries such as Financial, Insurance and Pharma, verbatim scripts and disclosures are mandatory due to regulatory and libel reasons yet, whenever possible, as agents become more comfortable with their interactions, appropriate off-script commentary is fine so long as their performance isn’t impacted.

Lastly, is tone and empathy. ‘Did the agent open the call correctly’, ‘did they explain themselves’, ‘did they acknowledge the client’ etc. should be reviewed to understand the customer engagement and how involved the agent was in the conversation. Again, in the end, how the customer views their experience with the agent will determine their ongoing loyalty and/or whether they decide to purchase on the call. Therefore, the manner by which how the conversation is delivered should be tracked to ensure the overall customer experience meets with the customer’s expectation.

What’s the future of QA?

Fred: Currently, I’m a big fan of speech analytics technology because, from a QA perspective, you can automate several things in the QA process that we track today manually.

For example, if you have to say, a 500 seat call center, there are likely up to 25 QA staff members that are monitoring random calls and scoring them. With AI, you have the opportunity to listen to every call with automated analysis which generates very useful reports highlighting sentiment, tone, length and other variables that are trending which pop out to assist the call center management team in their customer experience enhancement strategies.

Clearly, the human element is always going to be there, but with this new technology, you can optimize these processes, whether it is obtaining information faster or, more accurately.  Further, through the use of these tools, I believe over time it will result in a more lean and agile QA staff to best address the needs of the center with greater efficacy.

What is stopping people from adopting AI in call centers today?

Fred: I believe to a degree, it’s a little bit of the unknown that every emerging technology experiences. Many people are still not sure how well it will work or how to integrate within their current QA approach. Then it is knowing the real value it can provide – it sounds great, but can it provide the ROI for the call center or the client to justify the spend?  I feel enough data now exists to justify the spend, even if for a 90-day pilot to trial and I am seeing more companies and AI providers promoting that route as an entry point for further adoption of the technology. Time will tell but I firmly believe the future is bright for AI in the call center industry.

About Fred Shadding

Fred works with several BPO partners and helps them align their services with his clients that seek his guidance and assistance in finding the right partner that fits with their needs. He is a 20+ year contact center veteran and ran sales organizations for several BPO’s focusing on business development, marketing and client service as well as establishing outsourced solutions in 15 countries. In 2016, he launched The Call Center Source to focus on serving buyers and suppliers of outsourced services to educate and facilitate the cultivation of effective and long-term outsourced call center partnerships.