These are anxious times. Both from consumer customers concerned about rising costs and ensuring value-for-money and from businesses seeking to profitably meet their needs.
Lingering supply chain problems from the COVID-19 pandemic, exacerbated by climate-heating-driven destructive weather, global conflicts and political instability, but also by labor shortages, have contributed to increased product and service prices, low or no availability, and delivery delays.
But at the same time, consumers have strong expectations for excellent service. And when it doesn’t happen they become annoyed, even angry. Contact center agents hear and read that loud and clear in their tone and choice of words.
Contact centers cannot solve the underlying reasons why a product or service is not functioning, the wrong one was shipped, or why, or with service delays. But they can manage the experience customers (customer experience i.e., CX) have with them.
In such stressful times as these, it has become exceptionally important to assure quality service. And to ensure it is delivered whether the agents are working at home (which has become normalized in the contact center) or on-premise.
To get a handle on the issues - and on what contact centers can do to provide excellent quality CX - we had virtual conversations with these subject matter experts at several leading suppliers.
Dave Hoekstra, Product Evangelist, Calabrio
Manisha Powar, Chief Product Officer, Qualtrics Frontline Care
Dave Singer, Global Vice President, GTM Strategy, Verint
Elizabeth Tobey, Head of Marketing for Digital Solutions, NICE
Q. What are the top four quality assurance/quality management (QA/QM) challenges and issues that you are hearing from contact centers? What are their drivers?
Dave Hoekstra:
1. Inadequate coaching time. Business demands are making it difficult if not impossible to spend enough time with agents to coach and develop.
2. Too-small sample sizes. There just isn’t enough data to truly understand where the priorities are.
3. Remote coaching. QM leaders are being challenged to find ways to effectively coach and develop remote employees. While remote coaching can be effective, it lacks the impact of in-person coaching.
4. Shifting methodologies. As analytics tools become increasingly better, the traditional QM process is being less relied upon as a driver towards change, so the traditional QM department is starting to shift methodologies to a more analytics and conversational approach.
Manisha Powar:
The top four QA/QM challenges in contact centers revolve around maintaining quality and avoiding bias while handling high call volumes, adapting rapidly to market changes and customer expectations, ensuring consistency across diverse channels, and managing costs without sacrificing performance.
For many companies, QA methods in their contact centers are burdensome, time-intensive, and manual. They can also be susceptible to bias. For example, when evaluating call center performance on a random and limited number of calls, performance requirements can leave much up to the subjective interpretation of your auditors. This introduces the potential for human error.
In fact, a Qualtrics study found that a third (33%) of customer service agents felt their performance was not fairly evaluated and only 41% said they were incentivized to offer personalized, empathetic experiences.
These endeavors are also limited by the number of audited calls, hours in a day, and the audit timeframe. The industry is experiencing rapid change as consumer behavior and expectations continue to shift rapidly. And as contact centers grapple with how to best reduce costs to serve customers while implementing a plethora of new AI-powered agent-assist and coaching tools.
Dave Singer:
QA and QM programs in the contact center are generally too expensive and manually labor intensive. Companies are spending a lot of money and resources that are producing limited insights due to low evaluation coverage across customer interactions.
With limited insights, it’s difficult to coach agents and improve performance objectively because people don’t believe in the feedback due to the sampling bias that occurs as a result of manual processes.
As contact center volumes, engagement channels, and modalities grow exponentially, these manual processes cause contact centers to fall further behind the power curve.
Elizabeth Tobey:
The first challenge is that manual QA doesn’t paint a holistic view of CX operations and employee performance. Manual QA performed by supervisors makes it impossible to view and score every interaction. Therefore, supervisors are only scoring a small percentage of an employees’ work and don’t have a complete understanding of employee performance.
Artificial intelligence (AI) is transforming what’s possible with QA. AI analyzes 100% of interactions in real-time and historically, pinpointing areas for employee improvement. This leads to stronger, targeted coaching opportunities to boost employee performance. Additionally, when AI follows interactions in real-time, it can guide employees with suggested next-best actions, driving in-the-moment QM.
The second challenge is that employees don’t receive clear, data-backed feedback from manual QA. When supervisors don’t have a holistic view of employee performance across every interaction, they can’t effectively coach their employees. The feedback they do give employees is incomplete and can be perceived negatively by employees due to the greater chance for subjective feedback.
When supervisors have access to AI-powered QA technology, they can assess 100% of interactions and give employees data-backed, objective coaching to effectively improve employee performance.
The third challenge is that manual QA isn’t being done in real-time. This equates to missed opportunities for supervisors to resolve employee performance issues in the moment. This can not only lead to lower CSAT but also lower employee morale as their performance issues increase.
AI enables in-the-moment QA, coaching an employee in real-time with next-best-actions that lead to optimal results. AI can do this because it is trained on the best interactions. Additionally, AI can serve as powerful copilots to supervisors, alerting them in real-time when they need to intervene in interactions. This real-time QA means issues can be resolved faster instead of compounding over time.
The fourth challenge is that supervisors spend too much time analyzing the data with manual QA instead of coaching employees. Before AI, supervisors would have to sift through and listen to interactions to judge employee performance. This is a time-consuming process that doesn’t generate an accurate picture of employee performance over time.
“When supervisors have access to AI-powered QA technology, they can assess 100% of interactions...” —Elizabeth Tobey
Now AI automates this process, listening to every interaction for supervisors, tracking employee performance over time and pinpointing areas for improvement. This gives supervisors time back to focus on providing rich, targeted coaching to their employees.
Q. Conversely, what are the top four opportunities, in best practices, methods, and technologies, that you are seeing to enable superior QA/QM?
Dave Hoekstra:
1. A more targeted approach. This is based on what the customers are saying. Instead of a random interaction, focusing on the interactions that contain the information needed for improvement has a significantly higher success rate.
2. Auto-aggregation of topics. By having conversational topics grouped together and performing analytics on those topics, contact centers can stop guessing and start solving.
3. Reducing the form for the agents. QM forms can often get out of control, which impacts everything from agent stress levels to handle times. Reducing the number of variables in a QM form can actually have a positive impact.
4. More productive use of time. That means spending less time listening to individual calls and more time looking at the data from the analytics is what top tier contact centers are doing.
Manisha Powar:
We are seeing clients deploy new AI tools for predictive analytics and automated quality checks, robust virtual training modules for skill enhancement, omnichannel contact center platforms that ensure a more uniform CX, and advanced call routing technologies for better issue resolution. Each strategy helps unlock superior QA/QM and all together they can have a transformative impact.
“...customers still prefer human interaction when it comes to complex issues or emotional engagements that require empathy.” —Manisha Powar
Dave Singer:
The real opportunity for quality programs lies in the scalability of AI and CX automation to achieve comprehensive coverage across all interactions, channels, and modalities.
When combined with your company’s proven workflows, methodologies, and best practices, AI and CX automation allow you to embrace the best of what’s new without disrupting the value of your existing programs.
With the rise in Generative AI (Gen AI), for example, companies think they don’t need anything else to stand up a quality program. However, companies are rapidly discovering that while Gen AI provides insights, the question becomes:
- What do you do with these insights?
- How do you ensure Gen AI will deliver what you need?
- How do you test it?
- How do you calibrate it?
- How do you make sure it stays right?
Using Gen AI can only work “right” when it’s coupled with a company’s established quality processes and practices.
Q. Are you seeing any differences in customers’ views of quality experiences between automated and live agent channels?
Dave Hoekstra:
There is still some distrust in the contact center of auto-quality management, which aims to have AI evaluate every call with human-level accuracy. But as the software evolves, so will the results and with this, trust will increase.
By shifting the focus from catching the agents doing something wrong to understanding customer trends and pain points, contact centers can actually place the impact of change in the right place: which is customer satisfaction.
“There is still some distrust in the contact center of auto-quality management...” —Dave Hoekstra
Manisha Powar:
There is a difference in customer perception about quality when it comes to automated and live channels. A Qualtrics XM Institute study on channel preferences found that consumers still prefer to connect live either over chat or over the phone with agents across a variety of scenarios. These include technical support, resolving an issue with a bill, and booking a service.
While automation is appreciated for its efficiency in handling simple queries, customers still prefer human interaction when it comes to complex issues or emotional engagements that require empathy.
Dave Singer:
When it comes to automated versus live agent channels, consumers generally prefer to interact with a live agent as they don’t believe that chatbots will give them the answers they need.
While it’s true that contact center agents can provide a more empathetic and personalized experience, it can take longer for a company to fully realize this level of customer service and consistently provide it across representatives.
“...there's a significant opportunity to use AI-powered CX automation to grow and enhance the quality of CXs...”
—Dave Singer
So, while consumers might prefer talking to a live agent, there’s a significant opportunity to use AI-powered CX automation to grow and enhance the quality of CXs through automated service channels to delight customers by providing great self-service.
Elizabeth Tobey:
AI has advanced to the point where customers are beginning to not be able to discern whether they are speaking with a live agent or chatbot: and this is seen as a positive for both employees and consumers. This is only possible when chatbots are using purpose-built AI for CX.
When AI is only trained on industry-specific data, not the open internet, AI-powered agents can understand and respond to needs with the same level of skill as humans: offloading repetitive tasks from agents and decreasing onerous wait times for consumers.
Additionally, chatbots are becoming more intelligent through dynamic re-skilling where employees are training AI and chatbots on how to act just like live agents. This process is also happening in reverse with AI helping train live agents to be better. This dynamic partnership is improving automated and live agent CX, making every interaction exceptional for customers.
Feedback Fatigue
Customers, it appears, are being inundated by surveys at every touchpoint and channel. Are customers suffering from “feedback fatigue” and ignoring or purposely giving incomplete, or worse yet, wrong answers?
Dave Hoekstra
Yes, customers are feeling the fatigue. Yes, customers are skimming through surveys and not giving quality feedback. Why not spend some time analyzing what they are saying when you don’t ask them? Conversational analytics tools are vastly superior ways to get feedback than surveys.
Manisha Powar
We have found that only a third of consumers give direct feedback every time they have a bad experience with a company.
Since 2021, the share of consumers providing feedback directly to the companies they buy from following a very bad experience has fallen by 7.2 percentage points. So organizations need to be smart about gathering feedback where customers are giving it, such as in call center conversations, online chat, product reviews, and social media posts, and taking action to address it.
Customers are often overwhelmed by the excessive number of survey requests they receive without seeing any tangible improvements, so it’s critical to respect their time.
This is where artificial intelligence (AI) and omnichannel analytics can be incredibly useful. Having insights into what customers are saying or doing outside surveys, like identifying feedback in call analytics, social reviews, or digital behaviors, allows organizations to hone in on more specific survey questions on topics or issues where additional insight is needed.
Why ask a question to a customer if you already know the answer?
Dave Singer
Today’s brands want and need more data to make decisions, which is good. However, it requires companies to keep asking their customers to give them more and more feedback. Ultimately, the goal is to alleviate the survey load on customers while enabling companies to obtain more accurate answers across a broader reach.
A tremendous amount of AI and analytics data is being used today for synthetic feedback scoring. This approach augments insights with customer experience (CX) data analytics tools pre-built into your contact center platform.
It gathers and analyzes customer data across every customer touchpoint and channel to reach a broader audience while promoting proactive listening and providing real-time actionable insights.
This proactive approach empowers companies to stay ahead of customer needs and preferences. In addition, feedback analytics train on your AI models for continuous improvement. Not only does this help reduce the amount of customer survey feedback requests, it also automatically performs “lower-level feedback” so companies can keep manual surveys short and to the point to avoid feedback fatigue.
Q. What are your recommendations for contact centers to ensure high quality QA/QM?
Dave Hoekstra:
1. Listen to your agents. No one knows their process better than them and incorporate their feedback into the process. If they are frustrated, they will tell you...if you ask.
2. Make sure your QM form is aligned with the customer experience, not with the agent experience. If your form says to say the customer’s name three times, the customer is probably annoyed with that.
3. Develop a “one question” QM process. What is the one question that needs to be answered for a quality interaction? For example, “Does the customer have to call us back to solve this issue?” is a great starter question and build from there!
Manisha Powar:
It’s essential to involve agents in the QA/QM process so they understand its value, integrate AI and automation for efficiency, invest in regular training programs, and leverage analytics for smart decision making.
Contact center leaders need to collect peer feedback and celebrate agents who demonstrate the company values. Ultimately, a focus on both employee satisfaction and CX leads to high-quality QA/QM.
Dave Singer:
For contact centers to ensure superior QA and QM programs, my recommendations are three-fold:
- Infuse AI and automation into quality programs to improve the scale of coverage.
- Keep the best practices and methodologies from existing QM programs to ensure a closed-loop process and the right level of escalations and workflows that companies need to stay compliant.
- Utilize AI and automation technologies to coach agents in real-time to drive better outcomes and improve quality by allowing organizations to scale their coaching quickly and easily.
Elizabeth Tobey:
Organizations looking to improve their QA should invest in purpose-built AI for CX and an interaction-centric platform.
Generic AI that is trained on the open internet will not transform QA: it does not have the knowledge and the data required to meet the levels required to provide real value. Purpose-built AI for CX, trained on industry-specific data and delivered through a trusted AI vendor, will yield accurate and in-the-moment QA that businesses can trust.
In addition, when organizations leverage an interaction-centric cloud platform, this provides a rich training ground for AI to learn and deliver more intelligent outputs. An interaction-centric platform, combined with purpose-built AI, improves every aspect of CX operations including knowledge management, workforce management, and employee and customer experience.