Balancing AI and the Human Side of Contact Centers with Melissa Copeland
The New Mix of AI, People, and Customer Service - (TechXY Turbo, Episode 8)
Melissa Copeland is the Founder and Principal of Blue Orbit Consulting, where she helps organizations transform customer experience and contact center operations. Since launching the firm in 2014, she has partnered with Fortune 500 companies to deliver measurable improvements in customer engagement, technology investments, and workforce management.
Named “America’s Top Contact Center Expert” by Forbes, Melissa’s insights have also been featured in USA Today and Consumer Affairs. She is a frequent guest on industry podcasts, including On the Brink with Andi Simon, where she shares strategies for creating extraordinary customer experiences.
Melissa’s career combines global corporate leadership with entrepreneurial vision. After years of extensive travel in executive roles, she founded Blue Orbit Consulting to focus on projects closer to home while continuing to influence businesses worldwide. Today, she remains a trusted advisor for organizations seeking to elevate customer experience and achieve lasting results.
Listen to “Balancing AI and the Human Side of Contact Centers” below, or on Apple Podcasts, Spotify, YouTube, Amazon, Pandora, or wherever you get your podcasts. The episode transcript is shared below the embed.
TechXY Turbo - Podcast Transcript - (E8)
Frank: Welcome to another episode of TechXY Turbo. My name is Frank Gullo and I am your host.
Today we’re joined by Melissa Copeland, founder of Blue Orbit Consulting, recognized by Forbes as America’s top contact center expert. For more than two decades, Melissa has been at the forefront of a quiet revolution in transforming how organizations think about customer service technology.
In this episode, Melissa and I explore how AI is reshaping contact centers in ways that might surprise you. We’ll talk about why so many technology implementations fail, how to balance cutting-edge AI with essential human capabilities, and what it really takes to turn a contact center from a cost burden into a strategic asset.
Melissa’s clients achieve 10x returns on their technology investments, and she’s here to share what makes the difference between AI that helps and AI that hinders. Melissa, so great to have you on TechXY Turbo. How are you doing today?
Melissa: I’m great, thanks. And you? Thanks for having me.
Frank: I’m great. Thanks. So to start, and we do this with every guest, this is a tech podcast. So let’s start here. How would you describe your personal relationship with technology?
Melissa: So that’s a great question. I love technology. I have always loved technology and figuring out new ways to do things. It’s changing so fast right now that I’m as overwhelmed as everyone else with how to keep up on stuff. But I tend to find it super interesting to figure out what works and then what things that I still want to do the old-fashioned way.
Frank: Okay, great. So your career focus is contact centers. So for those of us who don’t know the term as well as you do, or work in the field, first, can you define what contact centers are and what does it mean in the context of 2025, and how have contact centers evolved since the early days of your career?
Melissa: Sure. So contact centers are a central place that people reach out to when they have a question, a request, or a need from a specific organization. That organization directs all of those contacts, and in the old days, it used to be just from a phone number or multiple phone numbers. Today it’s from all different channels to a central place where they can get those questions answered.
The flow used to be pretty simple. You call the number. It was answered likely by a small group of operators who either took your question or transferred your call. In today’s world, it’s a mix of menu-driven self-service, chatbots, and people. And you might have noticed over the past couple of years that it’s getting harder to tell the difference about who or what you’re interacting with.
Frank: Yeah, absolutely. And for those who’ve heard the word, the term “call centers,” is that synonymous or are call centers different than contact centers?
Melissa: They’re used synonymously. A contact center is slightly more broad in the sense that it includes chat, SMS, sometimes web queries, email in addition to voice. And the traditional call center was thought of as only voice. But when you hear people, including me, use them, we use them somewhat synonymously today.
Frank: Okay, great. So building on that, so in the early days, contact centers were often seen as also cost centers. How has technology shifted that perception and what role is AI playing now in elevating contact centers to more strategic business assets versus, you know, the old call center?
Melissa: The traditional call centers going back as far as probably the 1970s were designed to be cheap and quick. Actually, sometimes they were just designed to be cheap, but they were for handling inbound voice calls. So even the metrics that people would talk about—and the 1970s honestly predates my work in call centers, but some of these metrics were holdovers to today—and they’re things like average handle time, right? It’s supposed to be short. Cost per call, it’s supposed to be low, and those emphasize those values.
So around 2010 is when organizations started talking about customer experience and thinking about what I’ll call integrated service. It was still a number of years until the technology developed that could really support multiple channels. And so as the technology got better, interest in investing in a great customer experience became higher as well.
Interestingly enough though, AI in contact centers isn’t new. It’s been around for at least 10 or more years. Things like predictive modeling and machine learning power a lot of the functions behind the scenes in call centers. The newer things are the ability of the technology to interact directly with the end user who has the request, and that’s one of the things that help change the strategic value of centers.
From the perspective of being a strategic business asset, I’d say there are two big drivers recently. One, even though nobody likes to talk about it anymore, and it’s totally boring, the pandemic in 2020 really was a catalyst for change, both in consumer behavior and in companies for how they reacted to that consumer behavior. So companies were forced to enable interactions from anywhere, any channel, any time, and for companies to be able to service all that demand from anywhere, in any channel, any time, they had to deal with wildly different patterns than they were used to.
They had to invest on the spot, and voice became one channel of many and sometimes not the biggest one. So all of a sudden, a whole bunch of technology was deployed at scale. The second piece is the technology started to dramatically improve and further develop once there was scale in production. So organizations that long said that self-service wouldn’t work or that clients wouldn’t use SMS or chat for sales or orders found themselves proven wrong almost overnight.
So today, whether it’s conversational AI at the drive-through of your favorite fast food restaurant or a digital assistant helping your human customer service agent get you what you need, it’s faster and sometimes more accurate in the past 18 months than it’s ever been. And so that strategic business asset has really been the forefront of a real revolution that’s AI-driven in how centers can operate efficiently and effectively.
Frank: Thank you for that. That’s a very detailed answer with a lot of context. I want to push back a little bit, because I also think when I hear call centers, contact centers, I think of emotion and often fear and there is absolutely fear around job displacement today with AI and I suspect within the contact center space as well.
So from your experience implementing not just AI but automation solutions, what’s the reality? What is really happening here? Are we seeing AI replace humans? Is it changing the nature of offshoring contact centers or is something else more nuanced happening?
Melissa: I think we both agree that AI is replacing workers in many fields. That’s real. And that’s been going on for over a year. Job reductions are happening across sectors and they’re happening faster, I think, than a lot of people, including me, expected when we first started talking about this in 2023 and 2024. Customer service and call centers have been hit hard. And, you know, to give you some examples of what that means: we have one client that used to need hundreds of agents to answer calls about their specific third-party benefits. So think healthcare, think complicated, think emotional, all of those pieces coming together. So agents used to have to talk with the member, understand the issue, look up the contract and the coverage, and then net it out because the contracts and details were so difficult for anyone to interpret easily.
Today, that organization has digital assistants that are working with the agent to do those lookups and research very, very quickly. So what has happened in that center is a caller calls in and the agent can give them answers much faster than they used to. So an average call went from 12 minutes, 15 minutes or more down to single-digit minutes, six minutes, seven minutes, eight minutes. That’s a reduction of more than 30%. That’s a sea change, right? So the experience there is that they need almost a third fewer people to answer those calls than they used to by virtue of implementing these agents.
Now, the people that you do have, you need more and different skills from, so they still need the experience to evaluate what the digital agent is saying and understand if the answer is right, but they don’t have to spend all that energy synthesizing information. That organization in particular is redeploying 30% of that team to other types of work. So while there is a lot of fear out there, not every organization is trying to cut, although it’s cuts that often hit the news. So that reality is nuanced as well, right?
There’s also plenty of organizations that we work with that are feeling unexpected consequences. I’ll give you another example. One of our clients implemented a conversational AI application to help their teams schedule appointments. Now, in the case of this team, they were understaffed, right? And so this AI application is helping them do things that they were struggling to do otherwise. So in this case, they’re not hiring when they could have hired. Instead, they’ve had some great success with the chatbot.
Customers like it. In fact, more customers like it than they expected, so they’ve offloaded some of the call volume. Appointments are booked. But they also realized as they started running the numbers that they’re not selling as much to customers anymore because the sales motion happened when a human was on the call with the customer. And that’s not happening in terms of the agents now. That will get fixed over the coming months as the conversational AI technology gets better, but this organization is in the process of thinking through, “Oh boy, what did we do? And while we did some good, did we also do some harm? And how do we balance that?”
That doesn’t necessarily make employees more comfortable. But it does give leaders cause to pause and think through, okay, when is a person really the right way to interact with a client and when is automated interaction the best way to engineer what’s happening?
Frank: Yeah, totally, and that’s why I wanted to have you on, I think that there’s so many different use cases, and even how we use these things in our personal lives. Like I hear people really melting down talking to automated agents for airlines when they’re at the airport, because there’s so much emotion, but yet that same interaction in a workplace is not as emotion laden. So interesting.
Blue Orbit specializes in contact center operations and technologies in this space. What emerging tech has you most excited about right now? And on the flip side, what do you think is most overhyped and what should organizations be really cautious about?
Melissa: I’m going to be contact center geeky, or more geeky than usual for a minute. My favorite of the emerging technologies is speech analytics with full transcription and analysis. To give a little bit of context, traditionally in call centers you would often have a group called quality, and the quality team would be a group of people listening to calls that agents were doing and scoring them against a rubric that said, “Okay, this is what we think a great call is.”
Traditionally it was done manually. Past couple years, there’s been some automation, there’s been some speech recognition and recordings added. But with new generative AI-driven technologies, companies can now apply transcription and analysis to 100% of their calls. So the impact there has the potential to be incredible. Right?
You may or may not want to do after-call surveys, but you will be able to tell exactly what happened on every single call and speed up the analysis so that you can identify topics, identify words, and identify process issues. I think this is incredibly valuable when we think about what the future of customer experience and customer journey mapping and thinking through how we can best serve our customers as an organization.
So that is absolutely my favorite one pushing forward right now. It’s also pushing the boundaries of how people think about the skills going forward. To go back a little bit to the last question you asked, this is another area where lots of jobs are going away in quality assurance and call centers. Because you can have the automation do a lot of what people were doing, but there are new jobs being created.
Now there has to be a people or a team with, you could probably use the term engineering or analytical, but with the ability to look at all this data and understand the patterns and the implications of those patterns for what can happen to improve the organization. And that’s a very different skill than the work that people were doing before. So that’s my favorite in terms of emerging technology excitement.
In the same arena, I think the most overhyped technology is the AI-driven individual agent coaching. Brought to you by the same folks that are doing 100% of calls and those analytics, there’s a lot of work to automate the coaching for agents.
That is the digital agent or the bot transcribing and listening to the calls. And based on experience of looking at thousands or millions of calls, it will make recommendations to the agent about what they could do better. Speaking faster or more slowly, not using jargon. “You missed this disclosure.” “You had too much hold time.”
That whole process of recommending the coaching and then expecting or providing automation so the agent can go get the coaching on their own and then change their behavior. That one, to me, is completely overhyped. From my perspective, people learn best in different ways and often from human interaction and modeling. And so I think a lot of the organizations that are doing that favorite stuff that I’m excited about, they’re really pitching this coaching piece, which appeals to companies that are trying to reduce the load on supervisors and managers. And just my two cents, I think they’re going the wrong way with that one.
Frank: Interesting. Back to the first part of your answer, I think that’s very applicable, because right now you have many people in corporate in offices are on many meetings, and what you’re seeing now is whether they’re using Teams, whether they’re using Zoom or whatever, they’re taking the transcription, they’re taking the audio, and they’re having these pretty quick turnarounds of summaries that before we just didn’t have, especially when you have lots of people in meetings. So I think the call center processes and expertise—there’s more people now in regular work who can probably benefit or are using that level of automation now. I’m seeing that now and it’s really changing the value of what you get from meetings, like night and day to what it was even a year ago. So that’s pretty exciting.
You also work with business process outsourcing, selection and implementation. Speaking of some of these advances in AI capabilities advancing rapidly, how does this change the whole build versus buy versus outsource decision in contact center operations?
Melissa: In 2025, when it comes to the tech side of things, we counsel companies that, number one, rapid evolution requires flexibility. Because technology and AI capabilities are changing so quickly, one of the most typical structures for an organization to contract with a provider is a three-year contract or a five-year contract, right?
We really believe the three-year contracts right now are a bad idea. You really need to aim for a one-year horizon to maintain your adaptability and flexibility, and that protects you from being locked into a solution that is terribly exciting today and outdated tomorrow. Now, from a practical perspective for organizations, it doesn’t make any sense to sign a one-year contract. Really thinking through how you’re going to be able to shift what you’re doing with the partner that you select is one of the big things that we talk to folks about for their contact centers.
There are probably two or three others that we also wind up recommending. So one is definitely pilot before committing. Nobody funds these projects without an ROI and a pilot really helps you build a better business case. It’s also easier to pilot quickly with a buy decision on a commercial platform than it is to pilot something scalable on something that you’re building in-house. That pilot piece to make your ROI more concrete is a big one that we recommend.
And then we do recommend buying or at times outsourcing some of what you’re doing, just to get a feel for what the real return is and what the cost model looks like. When we think about building, building internally is the most viable for organizations that already have strong development teams and resources.
If the use case is already tested, building might make sense. If you have significant security compliance or very low tolerances with risk, then building may be the best or the only path that you can work through. So really thinking through what your needs are and going through kind of how you want to approach it are some of the things that we talk to folks about right now.
Frank: You focus on both customer and employee experience, and that’s interesting because some companies obsess over the customer, but forget about the employee. Vice versa. How does technology, in your opinion, impact agent engagement and satisfaction? Why should leaders and executives care about both?
Melissa: It’s a fascinating dynamic. It’s one of the reasons that when I kind of stumbled into call centers and customer experience many years ago, I stuck with it. Between customers and employees and both of those engagement aspects, the common theme from an employee perspective is people like to feel valued in what they do.
And value can have lots of meanings in the sense it can mean well cared for, it can mean respected, it can mean appreciated. So studies from Gallup, McKinsey, Aberdeen, and others show clear evidence that employees that are engaged deliver better service, turnover less, and drive revenue growth. Those gains can be by 20% or more compared to the production of their peers. When you start to layer over that technology, though, there are a few considerations.
The first consideration is that consumer taste in customer service for a particular brand isn’t the same for every brand and every person, but there are some common factors that consumers always care about: accuracy, efficiency, first time, and whether or not the experience was frictionless.
To your airline example, did I wind up knowingly arguing with a bot, as an example? Technology, when it’s easy and delivers on those vectors, is very acceptable to customers. For employees, our experience suggests, and it’s backed by research, that technology that reduces repetitive work, reduces stress and increases success in tasks is well-liked and well-accepted by employees. It also contributes to their engagement. It contributes to that feeling good and being successful at what you do. Think of that benefits example that I gave a couple minutes ago of employees that can deliver better, faster for their members because they have an agent helping them get the right answer.
The rub comes in with AI applications and potential job loss, how it’s messaged, what opportunities there are to interact with the technology before everything changes all around you, and whether employees consider both leadership in the organization and the technology itself trustworthy. All those things drive how employees react, and that’s why leaders really should be very concerned and thoughtful in terms of their decisions around how they want to incorporate employees.
When you think these things through, when you’re rolling out new AI-driven applications, you really have to consider: how are you going to sell it to the team, especially if there are workforce changes, and how are you going to get that buy-in? We find really having a strategy up front is one of the most effective approaches when you’re figuring out how you’re going to pilot or slightly before.
Frank: You said the word “change” at least twice in your answer, and that is a perfect lead-in to what I was going to ask you next. When you’re helping an organization implement AI tools or transition platforms, what do you think is harder to change? Is it the tech or the people? And based on that, what strategies work best for getting buy-in during these transformations?
Melissa: That’s a great question. Because sometimes the tech doesn’t behave as you and I both know the way we expect it to, but I would say in our experience, it’s people: 100% harder to change. Especially more recently, I think maybe, a scary number of years ago, 15 or 20, I would’ve said the tech is harder to change, because at that point it was.
Today the tech is not that hard to change. It might be hard to continue to evolve it or continuously keep up with the pace of the change, but the tech does okay in terms of the change factor. People though—nothing makes or breaks an initiative, particularly an AI initiative, faster than people.
You have a few factors in any organization that drag down the rate of progress. When you look at a technology program holistically, what does it take to accomplish the objective? How long does it take? Who does it affect from the customer all the way through the organization and how does it affect them? You have to think about change as both stressful and real.
So the fact that your best employees can and will get jobs elsewhere if they don’t like what’s going on should be a fairly significant motivator for folks to really think about how do we generate buy-in from our best employees and from as much of our team as possible. You can’t always get everybody, but getting as many people as you can, as an old boss of mine used to say, in the boat with you from the beginning, will make it that much more powerful as you push forward.
I’m thinking about some of the words I was using, and you’re right, I am using the word “change” a lot. I want to make sure to clarify that change is bigger than communication. One of the most successful techniques is sharing the why and the strategy and where it goes beyond the communication is offering employees ways to participate.
Training’s a big one, upskilling or really teaching people how to do what you want them to do. One of the things that we’ve seen this year that has been, perhaps less successful, has been when organizations just toss out, “Hey, here is your Gemini license. Here’s your ChatGPT license, here’s your Copilot license. Figure out how to do things better.” That has tended not to be very successful in terms of adoption or in generating enterprise-wide change. It can generate some good ideas, but organizations that really think through how do we want our employees to use the technology, how can we support them in learning something new and really applying it for an outcome, that really helps you get buy-in, up, down, and all around the organization.
Frank: Thank you. Great advice - I’m taking some notes here.
Let’s shift a little bit to business value and ROI out of some of these initiatives. Your clients achieve benefits of up to 10x their investment, and that’s a bold claim and impressive outcome. Can you walk us through a real example where integrating the right technology, process, and training delivers that kind of return? And what were the key factors that made it all work that people can take away for their own initiatives?
Melissa: Sure, I would love to - I love talking about this. Some of our best examples come from new tech implementation, whether the clients is an IT team or a contact center team, a lot of what we do in these engagements is help clients achieve results faster and bigger or better than they would’ve on their own. And so the natural question then is, well, how do you do that? In a few ways.
We help them avoid pitfalls that they wouldn’t otherwise know about. When you have a team, and most teams don’t have the luxury of staff anymore that can really focus on a new tech implementation—if you haven’t done it before, you don’t always know what can go wrong. And so we really emphasize, “Hey, you know, in our experience, or we’ve seen this before, or here’s how we can head off that.”
We also take a flexible approach to check and adjust as we go. If something doesn’t work as planned, we don’t want to wait or just keep doing it or ignore it. We develop an alternative plan that we can implement quickly. And so when you combine those two approaches and a collaborative work style, we really focus on the client success as our success.
Here’s a real example, which was the first part of your question. I think one of the best examples recently is a program we worked on this summer with a Fortune 100 company that had an opportunity to save over $15 million annually by transitioning from one type of technology to another. It was a relatively straightforward approach, but really complicated when it came to implementing it because they had global assets worldwide and teams everywhere that were impacted. Every single site had to make an adjustment to how they were working and know what to do if something wasn’t working. The client before we were hired had planned that it would take them 12 or more months to achieve the full migration to the new platform and realize those benefits. They hired us because they knew they didn’t have enough resources to manage the program and really support the joint client-vendor team.
With our team integrated into the client team, we were able to speed up all of the program implementation into about seven months, and we were able to transition some additional parts of the business that they didn’t originally think that would be possible based on our expertise and background in that particular area. We delivered acceleration that really dropped money to the bottom line for them. That’s how we get to things like 10x and that’s just one example.
I wish every program was that smooth and exciting. It doesn’t mean things didn’t go wrong. We had a few huge hiccups that almost sent the whole thing sideways and a lot of the opportunity when things get rough is finding solutions and communicating and falling to the backup plan so that we can keep executing to not lose ground. Having team members—and that’s the role we play—that can really focus on what are those plans and when do we need to shift and helping provide the data quickly to let leaders make decisions. That’s how we get to the ROI.
Frank: That’s a great and representative example. Thank you for sharing that. I want to shift a little bit to what I see as a reality in the market today with many organizations that are pretty strapped on resources and you’ve said yourself that few organizations today have that surplus of resources on hand for technology implementation. I’ve lived this and seen it myself, and it’s always a balance.
So in your experience, what do you feel are the common mistakes you see organizations make when they try, when they’re strapped on resources to implement AI assistants, transition platforms on their own without reaching out to expertise? And how do you help them avoid those pitfalls?
Melissa: I’m going to set aside change management for now since we’ve already talked about that, and that applies to a lot of initiatives, regardless of whether they’re AI or not. One of the things that we work with clients on is pragmatic planning and managing expectations. There’s an old adage in IT in particular that things cost twice as much and take something like three times as long. There is historical truth to that. So while I don’t encourage the 2x or 3x thinking, and I think it’s a bit less common than it used to be, it is hard for organizations to resist the internal pressure to say, “We can get it done now and it’ll be in two months, or three months, or four months.”
We recommend leaders be thoughtful about how to accomplish goals and what they commit to as the outcomes. I’m a huge advocate for bold goals. I just want organizations to be realistic about how long it takes to get to the bold goals, to do that and to avoid, you know, pitfalls and, hopefully, not too many negative consequences.
The other two things we talk about are internal resources and the going-forward operating model. So we see a lot of organizations recognize that things need to be different in the future, but they don’t do that practical task or job-level planning to make sure they have the right process, skills, and resources in place to support the new capabilities once they’re in, right?
In this case, I would recap the pitfall as “Wow, we got to go live and then we don’t know what to do next.” We really try and work with folks around what does your operating model need to look like? What’s different when we have this digital agent out there, we have this AI-driven process working? Who owns it? Who manages it, who fixes it when there’s a problem? Who decides what gets deployed next?
Doing some of that advanced thinking also helps organizations be much more successful. The last item that comes to mind that I would also highlight is we talk to organizations about being careful about relying only on professional services from the vendor during the sales process, we often say, you know, set your own strategy. Don’t let the vendor set your strategy for you.
We also caution teams to remember that the purpose of the vendor professional services organization is to get the platform or application working, not necessarily to ensure your success as a business. You have to be careful about believing the hype and don’t let the vendor call all the shots. Make sure that you have your own resources that are really devoted to understanding the technology, the capabilities, and how you’re going to operate it going forward.
Frank: Great advice, thank you for that. Moving on. So one of the things we’d like to do is we like to have our good friend, my AI assistant, ask a question of our guest as well. So I asked my friend Claude to analyze your background and generate a closing question.
So let’s see what it came up with:
Melissa, given your focus on integrating people, process, and technology, where do you see the biggest gap in how organizations are currently approaching AI implementation in contact centers? If you could close and give executives one piece of advice to bridge that gap, what would it be?
Melissa: Great question. Thank you. Part of what makes me smile is about giving one piece of advice, I have two teenagers and they wish I would give no pieces of advice, so I appreciate the invitation to give one.
I think the biggest consideration is to align your goals and your team first. Think big about what outcome you need. AI is a vehicle, not the end game by itself. It’s not a magic wand. It’s not a Magic Eight Ball. If you build it, that doesn’t mean that it will work exactly as you expect it.
When we think about aligning your goals and your team, I advise folks to test your foundation, engage your people, think big, and then use AI as the enabler to make that business outcome happen faster, not be the outcome itself.
Frank: Melissa, so great having you on TechXY Turbo. I’ve learned a lot, and we’ve only gone to the surface on contact centers, so if people have questions, want to engage more, where can listeners connect with you and learn more about yourself, Blue Orbit Consulting, and contact centers?
Melissa: Easiest way to connect with me would be on LinkedIn or you’re welcome to look up Blue Orbit Consulting on the web (@ https://www.blueorbitconsulting.com/).
Frank: Thank you so much, Melissa. I’ve enjoyed this conversation a lot. Have a great day.
Melissa: Great. Thanks Frank. I appreciate it.
