COVID-19 Impact: Recalibrating Human & AI Roles

As the global economy gradually reopens under evolving guidance and safety protocols for companies, employees and customers, business leaders are contemplating a vastly altered vision of the future of work from just a few months ago.

For many service and support operations, the sudden impact of COVID-19 exposed critical flaws and vulnerabilities in processes and operating models. Executive discussions are currently focusing on concepts like agility and resiliency, along with the mantra that “we can’t make the same mistakes again.”

What does the transition to the “new normal” entail for contact center operations? How will centers restructure service and support delivery models, agents’ skill sets, automation deployment and work-from-home solutions to emerge with the best possible outcome for future growth and success? For insights to these and other questions, I recently reached out to seven contact center industry thought leaders for their views. Our panel of experts includes Chris Arnold, Vice President of Contact Center Strategy, ASAPP; Tom Goodmanson, President & CEO, Calabrio; Tim Montgomery, Founder & Managing Partner, Alamo Cloud Solutions; Toby Parrish, Chief Operating Officer, Televerde; Jen Snell, Vice President of Product Strategy and Marketing, Verint; Paul Stockford, Chief Analyst, Saddletree Research; and Cameron Weeks, Co-Founder & CEO, Edify.

Q: What do you see as the top challenge facing contact centers as they transition to a post-pandemic service and support model?

Chris Arnold: There are two major challenges and opportunities: Digital expansion and work-from-home. In the last month, business, technical and in-house political barriers have fallen away to allow for improvisation and experimentation. Smart CX leaders are using this moment to pivot and expand upon their digital capabilities, which means meeting customers where they already operate—in asynchronous messaging/chat options and on more web and app pages. It’s not about AI or agents—the industry needs both. Augmenting agents with the right AI tools and platform will provide the ability to transform their structure into a digital-first organization.

COVID-19 changed the landscape of the contact center forever with work-from-home. Many agents will never return to the office environment and that changes how companies operationalize, what technology it deploys. WFH is not a transformation as much as optimization at this point. At leading companies, these opportunities will transform the cost structure of CX from an expensive legacy system to an agile digital environment where agents are augmented with technology that helps them to do their job fast and productively. An environment where agents can hand off basic tasks to automation and tackle the more difficult customer challenges will ultimately provide the best customer experience.

Tom Goodmanson: One of the top challenges that contact centers will face in the coming months is the decision on how to balance automated options and human agents. There has been discussion among contact center circles that COVID-19 could push contact centers further in the direction of chatbots and automated customer service. While the idea of “cheaper” automated solutions seems like the right answer, studies continue to show that customers prefer to interact with human agents, especially in times of crisis.

To best serve customers, the right balance must be met between AI and human responses. Contact centers need to analyze their interaction mix to gauge where bots prove a realistic replacement and where their implementation would come at the detriment of customer satisfaction and loyalty. In addition, contact centers should shift their thinking to see the use of AI and bots, not as a substitute for human agents, but as a supporting tool to improve customer interactions. Chatbots and AI should be used as a guide to empower agents as well as make them more accessible and knowledgeable for customers in need. Finally, certainly do not forget to do some type of automated quality assurance against the bots to help tune them to what the end customer really wants.

Tim Montgomery: Going back to the way things were done before. Customers proved that they’re ready to communicate in all channels—chat, SMS, email, video, social and intelligent voice. We must avoid falling back into the traditional prolonged approach to implementing new technical solutions and processes, and instead, move to a “fail fast and learn” model. In the “next normal” environment, the question is how quickly can we achieve a fully integrated, cloud-based “omnichannel” strategy? The answer should be measured in weeks, not years.

Toby Parrish: How limited is your imagination? Organizations have a golden opportunity to become a better version of themselves. The challenge will be not falling back on what we know or what’s worked in the past. Business has changed—forever. We must accept this and take steps to evolve. If we don’t, all the pains we are going through now and the lessons we learn along the way will have been for naught.

Companies that want to fully recover (which is all of us) can’t shy away from adopting new, emerging technologies. Before COVID-19, some companies found it hard to give up their legacy infrastructure in favor of new tech. This was a sacrifice made to reduce operational costs. It’s an antiquated corporate mindset that isn’t suited for our digital economy.

I’d strongly suggest that all leaders engage their employees in building this 2.0 version of themselves. Get their feedback on what’s working and what’s not. Learn what they need to succeed, their biggest challenges, how they prefer to work, and training they are lacking. Then commit to deliver on what you’ve discovered.

Focus on reducing operational costs where it makes sense then reinvest that money in areas that drive growth.

This out-of-the-box thinking will set up your company, your engagement centers and your employees for our post-pandemic future of work.

Jen Snell: COVID-19 has exposed consumers to the realities of the supply chain and the back-end of businesses in a way that they never were before. Whereas before, your customer base probably took for granted what it takes to run a call center or what their requests actually entail, they have now come face-to-face with infinite wait times, juggling staff and changing work environments, as well as an unprecedented amount of delayed orders. All of which is to say that suddenly your average customer has more doubts about the fulfillment of their orders or responsive support than ever before.

Regardless of whether you’re an organization that has run into these problems first-hand, or you’re a business that has been able to effectively get ahead of each of these problems as they’ve emerged, both your customers and your live agents are going to have scar tissue from this period. Looking at a post-pandemic world, it’s incredibly important for organizations to concentrate on technology that will work in tandem with a fully distributed workforce. The contact center of the future is decidedly adaptive.

If your organization hasn’t already invested in AI-powered solutions to help support customers and live agents alike, I would stress that now is the time. AI solutions like Intelligent Virtual Assistants (IVAs) and natural language voice bots can help contact centers scale their support and allow live agents to focus their attention on where they’re most needed. IVAs can also help a company ensure consistent brand representation across their customer support channels, allowing enterprises to seamlessly roll out any updates to the brand voice that they want to implement or test post-pandemic.

Paul Stockford: Modernization. Most contact centers were forced to scramble in order to comply with recommendations of the Center for Disease Control (CDC) in one way or another. For the 52% of North American contact centers that are already in the cloud, half the battle was won.They were able to transition to a work-from-home (WFH) model fairly quickly in terms of technology. For many cloud contact centers, the challenge then became scheduling, quality management and training for a WFH workforce.

The results of our 2020 survey of contact centers show that workforce management (WFM) is at the top of the list of solutions that will be upgraded or replaced in 2020 with 18% of the market actively evaluating new scheduling software this year. Looking at it from another angle, there are still plenty of scheduling software solutions out there that are in need of replacing even if they didn’t report that they would be replacing it in 2020. I still hear people referring to their “Blue Pumpkin” WFM software. I expect that the reported 18% of WFM software to be replaced will likely be closer to 50% actually being replaced by the end of the year.

While our data indicates that 50% of North American contact centers had some part of their agent workforce working from home before the pandemic, it turns out that the vast majority of these contact centers actually have a very small percentage, less than 25%, of their workforce working from home.Even if those agents sent to work from home during the pandemic are called back to the contact center to work, contact centers need to be prepared for whatever may come next, including the possibility that customer service will be directly impacted again by another crisis.

Cameron Weeks: Historically, those working in contact centers have worked from the office. And then the pandemic happened, forcing us to work from home. Now people know they can work from home and management’s previous response that “it isn’t possible” just doesn’t hold up anymore. This will undoubtedly be a topic that contact center leaders will need to address, as there will be an increasing interest to continue to work from home post-pandemic. The good news is that it is possible to run a highly regulated contact center in a remote environment where agents deliver exceptional customer experiences.

Working from home has the potential to tackle a long-standing issue that has impacted contact centers for years—attrition. If people have the ability to work from home, their job satisfaction will potentially increase. Additionally, contact centers no longer need to hire based on location and can employ people from anywhere in the world.

Q: What potential gaps or weaknesses has COVID-19 exposed between digital and human service delivery? How can centers quickly address these issues to ensure short- and long-term recovery success?

Chris Arnold: COVID-19 exposed that most companies didn’t have enough web, app and chatbot capabilities to contain the voice queue volumes from reaching over 800% capacity and presented consumers with over three-hour wait times.

To address this, pivot from crisis mode to innovation and review the new digital options that you need to stand up. Also, optimize the digital channels you currently have and make sure all teams are on board with the changes. It’s time for companies to be brutally honest in a retrospective review of why the existing digital capabilities were not able to contain the tsunami of inbound inquiries. The strategy and approach contact centers take now to adjust to the “new normal” will deeply influence their performance in the coming years.

Tim Montgomery: Many organizations were forced to bring their digital and contact center teams into the same conversation. In most organizations, both teams are looking at the same problem, but through a different lens. IT is building customer self-service solutions and the call center is looking for ways to get creative with their legacy call processing tools. The post-pandemic contact center environment will require organizations to create digital strategies from the inside out with a focus on giving our support agents easy-to-use tools, just like the ones we provide to our customers. We often see transactions handled in contact centers that can be easily completed by customers online. Customers can quickly teach themselves how to process routine transactions and get answers to their questions within seconds. But when they come into the call center, that same transaction can take several minutes to process, which takes valuable time away from the customer. Also, consider the training time for the agent—several weeks for agents vs. several minutes for the customer. Our future digital strategies will be even more focused on internal solutions vs. external solutions.

Toby Parrish: For Televerde, our transition to readiness wasn’t an issue. Our delivery and performance have actually improved. As an organization with 10 global engagement centers, seven of which operate inside prison facilities, we have robust contingency plans that are always ready to go at any time. We also began to heavily invest in our infrastructure a little more than one year ago and to transform our MarTech stack from homegrown to best-of-breed. These combined efforts enabled us to successfully pivot when we needed to.

In terms of the business community overall, the pandemic was a wake-up call. It exposed a lack of digital readiness across many organizations. Companies weren’t as far along in their digital transformation as they may have thought they were, and that became clear when the time came to pivot and they couldn’t.

To ensure recovery, companies must reinvent their business models and processes—digital-first combined with the human touch. When you bring data intelligence, marketing technology and the human touch together, you can actually accelerate revenue for your customers. Key will be building out your MarTech capabilities to ensure your teams are reaching and interacting with the right decision-makers, while simultaneously freeing them up to collaborate and strategize with your customers. This transforms your contact center into a savvy, forward-looking engagement center.

Smart investments in technology need to become a need-to-have (vs. a nice-to-have) and leaders must commit to going all-in on disaster recovery preparation. Plan and prepare so you can pivot on a moment’s notice.

Jen Snell: COVID-19 completely changed the trendline on the kinds of questions that customers were asking of brands, especially during the first few weeks. For example, at COVID’s onset, Verint IVA and conversation insights informed our travel client of questions from travelers wondering if their trips were canceled or if there were service interruptions due to the virus.

The same was true for the majority of contact centers, where customer queries suddenly narrowed down to one or two topics but at an unprecedented volume. For example, one of our Cloud IVR customers saw a spike of over 150 million minutes through their voice channel that was supported at scale. However, this shift in customer behavior also left many contact centers scrambling for solutions to scale across their digital and voice channels as well as support their live agents, especially if they were unable to identify the change in customer focus until they were in the midst of a deluge of customer questions.

That’s why it’s so important to be able to detect trends in your customer base as they are emerging. AI-solutions should provide management with insights from live-agent conversations and user interactions so you can update information and content in real-time across all channels, including digital IVAs and voicebots. Being able to identify large customer concerns as early as possible gives your business the time it needs to develop new processes and a cohesive roll-out of responses. It also positions you to beat competitors that are scrambling to retroactively address customer queries.

We were able to work alongside our customers to update their IVAs and natural language voicebots with new knowledge, ideas, actions and natural language understanding intents, so customers and employees could receive quick, personalized resolutions to their questions. From Verint’s conversation insights, our clients were able to glean critical insights and understand customers’ concerns so the solutions could address them quickly and at scale, providing seamless and exceptional service, even through a major disruption.

I would encourage all organizations to discuss with your AI vendor how they ensure their solutions are adaptive for emergent situations and informing crisis responses, just like this one, where new trends of questions crop up and overwhelm a call center if their AI is not supporting both customers and live agents appropriately.

Q: What are the opportunities for contact centers to leverage AI-driven technology to adapt quickly to changing circumstances driven by COVID-19?

Chris Arnold: We’ve learned to live with immense complexity in the industry. Agents are now more isolated by working from home. That strong peer support in their brick-and-mortar centers, consistent coaching conversations for challenging situations, and limited access to all the systems and tools creates challenges. An artificial intelligence platform can play a huge role in augmenting agents, coaching them—based on results of the center’s best agents—and being their assistant to proactively unearth the right answers and solutions for customers. Secondly, with machine learning algorithms able to ingest huge volumes of data—whether it’s voice or text—an AI platform can provide critical business insights to identify where a company is missing the opportunity to contain and automate basic transactions that will impact an entire CX organization and help transform its cost structure to become a digital-first organization.

Lastly, an environment where agents can hand off basic tasks to automation and tackle the more difficult customer challenges will ultimately provide the best customer experience. Implementing AI will enable CX organizations to address systemic inefficiency by way of automating the more basic, highly transactional customer needs while augmenting the agent with a modernized, intelligent suite of desktop capabilities for the more complex and emotive interactions.

When a center takes the opportunity to use AI, it can reduce the cost of customers switching to competitors due to poor customer service, which costs U.S. industries $1.6 trillion. AI can prompt the entire agent workforce to say and do the same things as the top 10% of its agents. Elevating the performance of an entire team by closing the typical range of performance also drives significant benefits in customer and agent satisfaction while driving substantial improvements in business objectives.

Tom Goodmanson: It has been said so often at this point, but these are unprecedented times, and for businesses, it is hard to navigate without visibility. Running with a gut feeling will not work as it has nothing to be grounded in. AI-enriched technology can offer that visibility and certainty for contact centers. Service teams can act quickly and with data-driven confidence using AI-fueled analysis that aggregate insights from 100% of interactions rather than the 2% that most contact centers manage.

Such an effortless overview of the overall tone and quality of customer interactions allows contact centers to track and respond to changes as they happen instead of playing catch-up later down the line. Using fast, holistic AI indicators allows for near-real-time alerts on the channels, processes or teams that need attention, for instance, if there is a sudden downturn in customer sentiment across a segment of customer communication, you would be able to take a multitude of actions to get them back in line with customer expectations.

Jen Snell: By leveraging AI-driven technologies like enterprise-grade IVAs across service channels, contact centers have the chance to turn their greatest challenge into their greatest opportunity. It’s no secret that contact centers are facing unprecedented call volumes that have the potential to overwhelm agents and frustrate customers. However, by leveraging an IVA as a part of your organization’s support stack, you can easily scale response to meet the size of the demand.

At Verint, we’ve seen how IVAs can be deployed to meet an outsized demand even before coronavirus. During the 2010 Icelandic volcano eruption, there were worldwide disruptions of flight and travel schedules that led to an incredible uptick in customer queries. Another customer in the travel sector, whose IVA usually handled approximately 25,000 questions per day, answered more than four times the number of questions in a single day during this crisis. Not only that, but the IVA was integrated with the travel company’s fluid scheduling tool, and travelers were able to utilize the virtual assistant to find out the latest updates without calling in or waiting in queues.

IVAs make it possible to answer thousands of questions accurately and simultaneously. During this time, when every customer wants the ability to reach out to brands 24/7 and get answers, there’s nothing more valuable than AI that allows a business to scale their response without needing to increase headcount.

Paul Stockford: The industry should automate immediately.I’m referring specifically to AI-enabled intelligent agents for both customers and agents—especially WFH agents. On the agent side, intelligent agents, or bots, can supplement in-house training and also act as real-time assistants to agents in terms of gathering and presenting information to agent desktops during a customer interaction. The role of intelligent agents in their ability to assist live agents is being defined in a sort of on-the-job-training scenario. Customer service professionals are constantly discovering ways to improve the agent experience and are increasingly turning to intelligent agents to fulfill those improvements.Replacing a manual knowledge management task with an automated intelligent agent is a good example of how the role of bots will continually expand.

On the customer side, automated intelligent agents were, or could have been, instrumental in handling the tens of thousands of routine calls that were blocked or abandoned because contact centers were overwhelmed with calls following the general shutdown of the country. In those scenarios wherein bots already had been deployed, agents were freed to address complex customer issues, while those customers with basic questions or issues were helped by automation.

With the country’s unemployment rate approaching that of the Great Depression, filing an unemployment claim became a frustrating endeavor for claimants and an overwhelming problem for unemployment call centers trying to answer an unprecedented number of daily calls. A small company in San Francisco called DoNotPay.com developed intelligent digital agents to assist with this problem and offered them to the public at no charge. In this application, unemployment claimants interacted with an online bot that gathered information from the applicant, created a form that the applicant could print and send to the unemployment via snail mail. Why did this process work?Because the systems processing unemployment claims were put in place during the 1960s and could actually process an old-school printed form faster than an online application?

Automated intelligent agents are available to contact centers of all sizes and there are no technological barriers to delivering these AI-enabled bots. They can be delivered via the cloud to virtually any contact center and have already proven their value in terms of reduction in cost-per-call so payback is immediate and impactful. It’s time for the contact center industry to adopt this AI-driven technology and put it to work.

Q: How will the acceleration of automation adoption or expansion impact agent training and development in the post-COVID work environment?

Chris Arnold:The proper use of AI overall will result in the simplification of the agent experience, and that has an outsized impact on reducing training time and costs. By augmenting agents with artificial intelligence in a messaging environment that coaches them on what to say, when to say it and what are the next best actions—based on what a center’s best agents have been saying in terms of close rates and solutioning—you can deliver a 40% reduction in agent training and ramp-up times.

Tim Montgomery: Most of today’s workforce has been exposed to online learning and a “do it at your own pace” approach to training. In the new world, contact centers will realize the ancient teaching with PowerPoint in a classroom needs to be replaced with flexible models that aren’t the traditional“one size fits all” training strategies.With modular training and at-your-own-pace programs, agents will be on the phones faster, and training budgets can be more aligned to meet business needs. This creative, new-world approach allows for near-real-time budgeting to help contact centers retain the agents we cannot afford to lose and to move quickly to make investment decisions that will have an immediate impact on the business.

Jen Snell: Organizations have now seen first-hand the benefit of bringing AI technologies into their customer support stack. They have witnessed how solutions like Intelligent Virtual Assistants have helped scale customer support and communicate brand values on-command. Many of these organizations also have already leveraged AI for internal purposes, including live-agent support and training. However, I expect that even more organizations will be interested in using AI as a tool to train their workforce post-pandemic.

After all, onboarding and training are some of the most important processes for a company. These are processes that require transferring brand values and institutional knowledge, but in a way that still empowers and makes sense for individual employees. While obviously, you would not want all onboarding to be done through assistants, virtual assistants do provide a resource for employees that allows them to ask all of the questions that they undoubtedly have without interfering with operations that might otherwise be focused on getting pre-COVID procedures back up and running. IVAs can train to the same high standard regardless of whether an employee chooses to remain remote or return to the office post-COVID.

Additionally, virtual assistants allow employees to be more honest with their learning and what they’re still figuring out. Particularly now that many contact center workforces have been pushed to align on and appreciate the value that AI and AI-powered solutions bring to the arena of customer support, employees are more comfortable than ever before turning to AI solutions for help. While there may be some element of embarrassment about constantly going to HR with your questions, it’s guaranteed that a virtual assistant won’t judge you for your questions—no matter how many you ask.

Cameron Weeks: Looking forward to a post-COVID work environment, I believe that training will be incredibly different and that will be in part due to machine learning (ML) technology. ML has the power to provide real-time, tailored learning tools, based on the individual needs of the agent. For example, ML can identify areas of growth for an agent based on the data collected and analyzed from that agent’s work on any given day, and quickly and easily offer that agent training to address those specific areas of growth. No longer will agents need to sit and watch PowerPoints, but instead they will benefit from curated, bespoke learning models personalized for each individual agent. This will undoubtedly impact the customer experience for the better.

Q: Which AI-powered support tools will add the most value to the customer and employee experience as contact centers navigate a near- and long-term recovery?

Chris Arnold: We’re at a point in time in the contact center space where it’s not an either-or situation. It’s not automation or augmentation, and it’s not bot or agent. It’s all of it.

Agents are crucial to the customer experience industry, and when they are augmented by AI, it provides them a virtual exoskeleton that will amplify and accelerate their capabilities to be more productive and effective. At the same time, an AI platform that ingests vast amounts of data can produce action-oriented processes and results to drive out systemic inefficiencies that have driven the frustration and stress of both customers and agents. The automation element can be good for consumers who want the choice, convenience and control to complete basic transactions at any time and at their convenience without the friction of waiting. However,being able to gracefully transition from human-centered to machine-enabled parts of a conversation is sophisticated and not simple, it requires a lot of smarts from a technology perspective, so it’s critical that an AI strategy is well-planned.

A number of our customers are using AI as an umbrella technology that stitches together all of these elements and crucially expands across both digital and voice, which isn’t going away. AI is also driving digital engagement and supporting more complex and emotive contexts where agents are essential and can raise CSAT associated with emotive contacts. That omni-experience hasn’t been achieved, but with the proper use of AI, even the largest and most complex CX organizations will get there.

Tom Goodmanson: One of the main trends we will likely see on our road to recovery is a continuation of the work-from-home model. In a PwC survey from April 2020, about half of responding CFOs said remote work will be a permanent option for some at their companies. I believe contact center agents will be among those that begin to work from home more often. And with that in mind, AI-powered support will become even more important as it will offer contact centers the ability to remain flexible yet in control, even from a distance.

With agents out of the office, predictive AI insights can give contact centers and agents essential visibility into support strategies. Setting up an AI-driven command center for interaction evaluations imbues contact centers with an insightful preparedness no matter where agents are working. Contact centers can use this information to prepare their agents with targeted training and allows management to remain close to agents who may be at home but should not feel like they are on an island. Continuous evaluations through AI and supportive coaching will also pave the way to recovery and growth as they are a vital investment in long-term customer satisfaction, retention and upsell.

Another potential struggle for post-pandemic operations is agent scheduling. Especially during the height of the pandemic, agents often needed to change their schedule at short notice owing to personal circumstances. Even as we return to a sense of normalcy, an increasingly remote workforce will require this same level of flexibility. Rather than just looking to chatbots as a fix-all for the customer experience, contact centers should embrace bots as an agent-facing self-service tool that can offer them the flexibility to adjust their working day as it happens and enable more adaptive service operations.

Jen Snell: There’s little question that COVID-19 has proven that AI technologies are a vital component in a contact center’s customer service stack. And I don’t think that’s going to change post-pandemic. However, I think that long-term, we’re going to see the value of AI technologies become just as obvious when it comes to augmenting and enhancing employee support and training.

AI technologies like Intelligent Virtual Assistants can be used for just-in-time learning to give employees access to the information they need, when they need it. Given the newfound sense of autonomy that a crisis brings, your workforce no longer wants to have to wait for a response from HR—who themselves may be swamped with post-pandemic tasks—to a simple question. Nor do your employees want to skim through an employee manual, which is likely littered with outdated processes in a post-COVID world. With the right IVA vendor, you can ensure that your employees get the updated information they need, exactly when they need it with no additional hassle or stress. Virtual assistants really empower employees to take learning into their own hands, which is going to be incredibly important as some contact centers remain remote while others readjust to life in the office.

Cameron Weeks: AI- and ML-powered technologies are known to benefit the customer. Bots are very common and many customers enjoy the self-service capabilities these tools offer. I believe that the next value-add we will see from AI and ML technologies will focus internally around supporting the agent.

As people continue to work from home, and as businesses still try to grow, AI and ML will be instrumental in educating, training and information-sharing with employees in an efficient and effective way—regardless of whether they are in the office and in-person with their managers or working remotely.

When agents receive better support from their leaders, they will deliver better service to customers, so as contact centers leverage technology to better train and support their agents, the customer experience will benefit, too. It’s a complete win-win scenario. Agents are the front line of the customer experience and if we’re not empowering them to be successful, it doesn’t matter what we’re doing in regard to the customer experience.