In this episode of Enterprising Insights, host Keith Kirkpatrick discusses recent contact center product announcements, focusing specifically on news from Microsoft Dynamics 365 Contact Center and Cisco’s Webex Contact Center. He then concludes the show with the Rant or Rave segment, focusing on a contact center snafu.
You can grab the video here and subscribe to our YouTube channel if you’ve not yet done so.
Listen to the audio below:
Disclaimer: The Enterprising Insights podcast is for information and entertainment purposes only. Over the course of this podcast, we may talk about companies that are publicly traded and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors and we do not ask that you treat us as such.
Transcript:
Keith Kirkpatrick: Hello, everyone, I’m Keith Kirkpatrick, Research Director with The Futurum Group. I’d like to welcome you back to Enterprising Insights. It’s our weekly podcast that explores the last developments in the enterprise software market, and the technologies that underpin these platforms, applications, and tools. This week, I’m going to address a couple of contact center focused product announcements that have come out recently. Namely, Microsoft Dynamics 365 Contact Center and Cisco’s Webex AI-powered self-learning contact center. Then of course, I will address my weekly Rant or Rave. Let’s get right into it here.
There are a couple of announcements that are being made this week around contact centers. When we’re talking about contact centers, we’re talking about contact centers as a service. Essentially, it’s cloud based. The idea here is that what these companies, what these vendors are trying to do is really enable some of the functionality from AI, basically enabling AI within these contact center things to make customer experiences more seamless, less full of friction, and of course hopefully, helping out both the end customers as well as the agents who are serving them.
Let’s take a look at what Microsoft is doing. Microsoft is announcing the Microsoft Dynamics 365 Contact Center. They’re terming it a Copilot first solution. The goal here is really to engage or help customers engage and deliver exceptional customer experiences across all of the service channels. When we’re talking about this Dynamics 365 Contact Center, really what they’re talking about is this comprehensive vision for delivering service. It’s going to really encompass not only the contact center functionality, but also in pulling in things like CRM, generative AI, CDP, all of these different solutions, all within a common platform. The goal here, there’s really a few different things here that they’re trying to do.
One is make it easy to integrate all of the data from across the organization onto a single platform. That just makes things much easier for IT, and also in terms of actual utilization of the services. You’re not having to go out and grab data or processes from a bunch of different unrelated platforms. The other thing that they’re really leaning into, of course, is Copilot. As everyone is very much aware, Microsoft has leaned in heavily to delivering Copilot across their experiences. This is essentially what they’re trying to do is make it easy for customers and agents to interact with data and processes within the workflow that they’re using at the time, to really bring in that generative AI functionality. That can be anything from self-service routing to agent assist services, post-call wrap ups, analytics. You name it, they’re trying to incorporate AI, particularly generative AI, to make things work more smoothly.
A couple examples. Let’s say I’m a customer, and I have an inquiry, and I interact with a self-service application saying I need help with something. Well in many cases, if it’s an easier request, the idea is to let the AI take care of that request automatically. But what they’re also trying to do is intelligently assess what is this caller or this person coming in through an SMS text or an app really trying to do. Basically, they want to try to find out or use this technology to identify intent. In some cases, it means that well, the best case is not to actually have someone get shifted over to a full bot service, but actually get routed to a human agent who has the right experience, knowledge and tools to handle their request.
What the goal here is obviously to reduce that friction of going through an entire process with a bot only to be told, “Hold on, we’re going to connect you with an agent.” Their goal is to make sure that you’re able to get to that agent right away, because that will speed resolution, it reduces friction, and it makes the experience for the customer so much better. We all have those horror stories of going through this endless process, only to wind up with an agent anyway. Some of the other things that Copilot can do within a contact center framework. Assisting agents, so serving up knowledge to agents as they’re working on a case because the system will be basically incorporating all of the customer data, relevant customer data. When was the last time they called? What did they call about? What types of products have they purchased? What service agreements do they have? Then serving up that information to the agent so it’s at their fingertips, so they don’t have to hunt for that information.
Using generative AI to generate emails or communications. Essentially, if an agent has to reach out to a customer, they can generate an email for them. Obviously, the agent still has the final word. They have to look over this content before it ever goes out. But it really is a starter. The hardest thing is to actually write from a blank page. This gives them the relevant information they need to actually initiate that communication. Then of course, something like call wrap up. One of the most difficult things for the contact center agent, whether we’re talking about someone whose dealing with someone on the phone, or through a text, or through an app, or whatever have you is when that interaction is over, they need to capture quite a bit of information. It’s not what they do best.
A lot of times, these people haven’t been trained on that. By using AI to actually listen in and capture all of the steps that we taking during that interaction, number one. And also, wrapping up all of the other notes about the interaction. Was the customer upset? Did they take these steps? What actually happened on that call? By using generative AI to capture all of that information, it saves a lot of time with that post call wrap up. It also improves accuracy, a lot of the time. Certainly, people try their hardest but sometimes it’s difficult to remember exactly what went on throughout a call, particularly if that interaction was quite long in duration.
Then of course, the other thing that Microsoft is talking quite a bit about is scalability and reliability. That’s critical in today’s world, particularly for services that are targeting large enterprises, ones that may have a fluctuating demand in terms of number of agents that you’re using or volume of inquiries. We look at something like retail, where obviously they’re going to see a real bump up in traffic and inquiries around the holidays. Both in the pre-holiday season, throughout the holiday season, and then afterwards for dealing things with returns and that sort of thing. That’s one of the things that Microsoft is really trying to market in these services.
I think, when we think about what they’re doing here, Microsoft is doing, is they’re trying to basically extend their market footprint. Obviously they’ve had a strong hold in the marketplace, when we’re thinking about Dynamics 365, in terms of productivity. Certainly, they’ve tried to make inroads in terms of being the ERP of record for small to medium-sized businesses. What this does is gives them another application that they can then in turn go out to customers with. And say, “Okay, we have all of these applications available for you, they’re all on the same platform. They’re all going to look to be working together very seamlessly.” Of course when you do that, the actual cost to implement should be lower.
That’s one of the things that is really coming up quite a bit today, is this whole idea of, A, how can we maximize our investments, minimize our costs. Of course, the biggest thing is, it’s time to value. When you think about the whole implementation process of putting in a new system, or even putting in a piecemeal system, you want to make sure that you’re able to get up and running very, very quickly. The reason for that is, when you see things like generative AI come in, you don’t want to have this long deployment cycle because by that point, the technology cycles would have cycled through a couple of different times. So you’re really behind, in terms of certain functions. It’s important to be able to quickly get value, to implement and get value out of each investment that you make.
The thing, I think, Microsoft, like a lot of these companies, has realized that, and that is why they’re leaning heavily into this idea of this platform play, making it a lot more simple to implement. Whether or not you use all of Microsoft’s tools or you have to integrate with another solution, they’re still trying to do what they can to make sure that that process is as seamless as possible. Now, I’d like to talk about another announcement that came out this week. That is Cisco. They actually announced there AI-powered contact center. Obviously, it’s branded under Webex. They’re talking a lot about their new AI assistant for contact center capabilities. Like Microsoft, they have similar use cases for AI. Things like agent summaries, call summaries, suggested responses, call wrap ups.
They actually also include things like coaching highlights. If an agent has an interaction and there’s something that could have been handled differently or better, they’ll actually get coaching through the application. Another one that’s pretty interesting is an automatic CSAT score. Which is, for those who are not familiar, that’s a customer satisfaction score. That’s pretty interesting because, if you think about what organizations are trying to do is quickly identify when there’s been an lapse in customer experience. If a customer is not satisfied after an interaction, they want to know and they want to know quickly. Of course, the most important thing is they want to try to identify the root cause of that. That’s where they can look back through the interaction.
What happened? Did it take too long to service? Was it an issue with the product? Was it an issue with the interaction that was going on? Being able to do that and identify that very quickly can be really powerful, in terms of delivering a better customer experience. One other thing that’s really interesting, and Webex is doing this, is they actually have this burnout detection feature. That’s really interesting because when you think about the life of a contact center agent, it is not an easy job. They have incredible volumes to deal with, in terms of interactions, whether we’re talking about phone interactions, or text, or web application communications.
All really, it’s very high volume. A lot of times, they’re asked to handle multiple interactions at once. Let’s face it, a lot of times, particularly if we’re talking support, you’re not dealing with happy customers calling up to say how happy they are. Almost nobody does that. You’re dealing with angry customers who are frustrated, who are automatically gearing up to having a negative experience just because of their past history of calling up agents and not getting what they need. What they’re trying to do with this is identify what it is that … If there are certain things going on in interactions that are indicative that an agent’s performance might be slipping, or they might be slipping into a bad habit. Whether it’s sentiment, or tone of voice, that sort of thing, they’re able to identify that.
I think the goal with all of these things is to just assess when an agent is having an issue, so they can be pulled out of a heavy rotation for difficult calls, or whatnot. The idea really is to make sure that agents are taken care of because it is a very stressful job. Being a contact center agent isn’t on the top of anybody’s list, generally speaking, in terms of jobs that people are clamoring to do right now. They are having a labor shortage, so anything that companies can do to retain good agents. And also, keep good agents from becoming bad agents is certainly welcome in the industry.
One of the things that I found pretty interesting as well is that Cisco actually released a few numbers here. I believe this was from their early beta users. They actually surveyed them to identify how well everything was working. A few of the stats here. I think they said that, in terms of AI customer feedback in the contact center, they were able to generate, using AI, they were able to generate a 3X faster response rate to customers. That’s really important. Again, we live in an instant gratification world, where you want an answer right away. You can’t wait, customers won’t wait. Whether it’s fair or not, speed is everything. That’s really interesting to see how AI can improve response time.
The survey also found that 80% of the responses of agents actually will reply more accurately with suggested responses. That’s really important. When you think of an agent who is asked a question on the phone or through text, if they’re not able to look up the correct information, a lot of times they may just respond with something that might not be correct. If AI is able to suggest a response that is correct, that is grounded in company data … When I’m talking about grounding, I’m talking about the AI model will only use the data that has been specified as being this body of knowledge that it is allowed to grab from. That’s a really important thing, particularly when we’re looking at certain support things. Where if we’re talking about a customer who has a question about an insurance policy, or something where there are real consequences to giving a wrong answer. That’s a really interesting stat.
Another really interesting stat is that, of these early customers, they also said that agents, 93% of the agents could get up-to-speed with customer history and context of the call faster with virtual agent and drop call summaries. What does that mean? Well, that means that there’s just much more information provided in context to these agents about the interaction with the customers, with their history, which is really great to see. For example, if I call up a customer service agent and I don’t get through, whatever, and the call is dropped because either I drop it or something happened technically. Well in the past, an agent wouldn’t know that. They wouldn’t know that I tried to call three times and I was having a problem getting through, where I had too long of a delay, or whatever the issue might have been.
Now these agents are actually provided with that information automatically, so they can open the call with, “Hello, Mr. Kirkpatrick, I just want to first say I’m so sorry you’ve been having difficulty getting through to us. Please tell me what your problem is.” Right there, you’ve automatically knocked down one of the big barriers to a positive customer experience and that is demonstrating empathy for that customer’s situation. You automatically acknowledge, “Hey, I know you’ve been having issues getting in touch with us. How can I help?” That’s a real game-changer, when you think about the typical nature of customer experience and experience support, where you don’t know who you’re going to get, and they don’t know you, and it seems like they don’t care about you.
Then, I think the last one is really important here. They said that 80% of agents, they have clear noise-free and distraction-free conversations with this technology that Webex has included. It’s called Noise Removal and Voice Optimization. What is this doing? Well, it’s essentially removing the static, the background noise, everything that gets in the way of being able to clearly understand what the speaker on the other end of the line is saying. Again, it’s amazing how simple that concept is. That if you have a clear line, communication is easier. It’s absolutely imperative today, when we do have voice conversations, that people can understand each other.
I think the other thing that comes up, and sometimes contact center agents, they may not be speaking in their primary language. Sometimes when you have accents involved, it makes it more difficult for people to understand each other. If you add in background noise, then it can become really, really difficult. When you remove it, it just clears the path to better understanding and communication between people. I think that’s a really interesting and powerful stat that came out of this early test run with the service. Again, I’m going to be writing some more about both of these services over the coming week, but I just wanted to highlight a couple of things there because I do think that we’re seeing AI now become a real catalyst for really improving customer interactions.
Really, the contact center is the first stop on this journey with interactions anymore. It’s rare that you have face-to-face interactions. I don’t know anybody who writes a letter to companies anymore. I assume that they complain, they tweet about it now. This is a really, really good step for the industry, for any company that utilizes a contact center. With that, I want to move to my Rant or Rave segment. This, again, is where I pick something in the market and I either champion it or I criticize it. Today, I was trying to be topical, I do have a contact center rant.
This one is related to … Well, I’ll give you the situation. I took my car into the dealership for service a couple of weeks ago. I actually had a very good interaction there. They took care of what needed to be taken care of. They said to me, “We have two open recalls. Would you like us to take care of it?” Yes, thank you very much, I would love you. They acknowledged, repairs were done, all good. That was about two or three weeks ago. Then about a day or two ago, I started getting multiple phone calls to my home phone, to my cellphone. I got a text. Actually, a couple of texts and a couple of emails. All asking me, or letting me know that I had an open recall on my car. I thought to myself, “I don’t know if that’s true because I was just at the service center.” I just thought that was strange. Then I finally picked up on one of the calls because I was tired of it interrupting me.
I talked with the young lady who called and said, “I’m sorry, you have an outstanding recall.” I said, “Well, can you check that because I was just in?” She checked, it took her a little while. Then she acknowledged and said, “Oh, you’re right, sir. I’m sorry. Thank you very much. Goodbye.” A couple of things here that I’m going to rant a little bit about. Number one, they should know this information. This should all be in their CRM or CDP, or whatever they happen to be using to contact me. The fact is, it was acknowledged that they took care of those previously open recalls. It shouldn’t have even been a phone call. This is what these systems are for, is to keep track of all of that so you’re not wasting the time and effort, and annoying customers with basically irrelevant calls.
The second thing would be when this young lady did call me and I explained that no, I already had the recall taken care of, there was a missed opportunity there. I’m not saying I want to be sold every time I hop on the phone, but I do feel like they could have said, “Okay, I’m sorry, sir. Did you know we’re running a special on this? Did you ever think about this?” There were a number of different upsell opportunities. Of course, the other thing that was missing was the, “Well, would you like to schedule your next service appointment?” Very, very basic. Not hard to do. It’s rare that you get people on the phone these days.
Again, I think we’re looking here at two things. A training issue and a data issue. Those are two things that AI probably isn’t going to be the solution. I think it comes down to really making sure that people are trained and there are the right processes within the systems to handle those things. Really, this can make all the difference in the world in terms of customer satisfaction, making sure that customers feel like in exchange for all the data that we give all of these companies that they’re actually listening, and being active, and being present, and understanding who we are and how we’ve interacted with the companies in the past.
Anyway, that is my Rant for the week. Not terrible, but I do think it’s an interesting aspect in the market. That we do have the technology to handle these things, and yet in many cases, it’s still not being done. Well with that, that’s all the time I have today. I want to thank everyone for joining me here on Enterprising Insights. I’ll be back again in about a week with another episode focusing on the happenings within the enterprise application market. Thanks, everyone, for tuning in. Be sure to subscribe, rate, and review this podcast on your preferred platform. Thanks, and see you next time.
Author Information
Keith has over 25 years of experience in research, marketing, and consulting-based fields.
He has authored in-depth reports and market forecast studies covering artificial intelligence, biometrics, data analytics, robotics, high performance computing, and quantum computing, with a specific focus on the use of these technologies within large enterprise organizations and SMBs. He has also established strong working relationships with the international technology vendor community and is a frequent speaker at industry conferences and events.
In his career as a financial and technology journalist he has written for national and trade publications, including BusinessWeek, CNBC.com, Investment Dealers’ Digest, The Red Herring, The Communications of the ACM, and Mobile Computing & Communications, among others.
He is a member of the Association of Independent Information Professionals (AIIP).
Keith holds dual Bachelor of Arts degrees in Magazine Journalism and Sociology from Syracuse University.
