HITT Series Videos

HITT- AI in CX March 30, 2024

March 30, 2024


Introduction to AI and CX Training

So, for those of you who don’t know me, I’m Sam Nelson, VP of CX here at Telarus, and today in our high intensity technology training or our hit series.

I’m gonna talk about AI as it pertains specifically to CX or customer experience. And I do encourage you to ask questions. I am monitoring the chat here as well as the Q and A box throw them in the chat. This is supposed to be super dynamic.

And this is also live. So, if you see any random squigglies, guys, it’s live. This is the way I love actually presenting training. So, for the agenda today, I wanna make sure that I set the proper expectations with you all So first, I’m gonna provide you with some stats.

Then I’m gonna go into some pretty key components of, AI and CX. And then I’m gonna go into use cases. Now this is where you’re going to want to take these down and be able to talk to clients because stories sell. Right?

I’m gonna end with the big question that you can go to customers with today. And then the last thing we’re gonna hit is q and a as per usual, which usually takes, the most amount of time. So, but I like I said, do you got questions in between? Go ahead. Stick it into the chat, and we will go from there.

Key Components of AI and CX

Now, next piece. Okay. So I talked about stats. Stats. They out I think I heard somewhere that, like, all, I think, like, eighty to ninety percent of stats are completely made up. These are coming these are coming from Gartner to the Gartner of Forrester.

But with that said, some pretty interesting stats that I think are super helpful for just starting conversations.

Right? Your first one is going to be this lovely baby, seventy percent, and that’s the percentage of companies that will actually adopt AI in some type of form by the year two thousand and thirty, right, which is actually not far away.

It’s, like, six years away. Right? And then another big number is two point five billion. Okay.

Definitely not a small number. But that’s actually how much the global chatbot industry is expected to reach by You guess it? Twenty twenty five because that’s next year. Okay?

And then check this one out. Ninety five percent.

Of all customer interactions are going to be powered by AI by boom.

Next year. Another crazy stat. Right? The last one I’m gonna give you is eighty nine point eight billion a much bigger number, and that’s actually how big the global AI market is expected to be by twenty twenty five.

Okay. And don’t worry. This is all recorded. So if you miss something that I said, don’t worry about it.

And like I said, We are live. Right? So I’m just gonna go ahead and erase this baby while we get into some key components about AI as it pertains to CX.

Understanding Segments of Interaction

So Suddenly, it’s a Windex commercial. I like this.

Suddenly, it is a Windex commercial. Yes. Indeed.

Look how look how clean that ended up. Right, Doug? Am I right? Am I right?

I feel like I’m right back to the shamwow days. Alright. So, when we look at key components of AINCX, it’s really, really important to consider in what area of the interaction we are talking about. And when we look at customer experience, you know, you have to think about it in terms of an interaction Think about all the times that you’ve interacted with a company, in some way or form, whether that’s, through a chat bot or, voice or, email, or social media, or what have you.

Right? So I want you to think about it in terms of three different segments of an interaction. Now maybe some of you have actually seen me sort of talk about this, but I wanna reiterate it because, it’s been some time, but also, it kind of puts some structure around how AI is implemented, within an interaction that’s most relatable to both you as well as your customer. So I’m gonna break it down to three very specific segments.

Okay? It’s gonna start with the free interaction, and then you’ve got the during, and then you’ve got the post interaction.

Okay. These are kind of the three different areas, if you break down an interaction here, the three different segments. Right? The first segment, which is sort of your pre interaction, and I’m when I say pre interaction, I’m talking about, like, before talking to an actual person, right, is where you’ve got things like your IVA interactive virtual agent or assistant, as well as your chatbots.

Okay. Now during, alright, something during where AI is applicable is what people call agent assist. Now this is really neat, and I’ll go into these details, shortly, but it’s called agent assist.

Post is where you see a lot of the QA or QM technologies.

And that refers to quality assurance. Quality management, and I will also spend some time going deeper into those components. Now let’s start back up let’s go to the IVA chat bots. Okay? So IVA, again, stands for, interactive virtual agent, interactive virtual assistant, intelligent virtual agent, and tell so intelligent interactive or agent and assistant typically interchangeable.

Interactive Virtual Agents and Chatbots

But that’s the technology where, like, you call in And they say, you know, tell me what you’re calling about. Right? And you know you’re talking to a virtual agent. It’s not a live person. And you say it in whatever language you speak in or whatever accent or however you prefer use slang or whatever, and it’s up to that actual bought, AI bought to determine what exactly it is you are calling about and then route you to the right department or the right menu to best help you.

Chat bots, same deal. Right? They are dipping into FAQs and what have you. So, like, you’re using a chat bot, on a website, right?

That is where, you’re going to start typing stuff. It’s gonna return answer to you, but as soon as it can’t, it’s usually going to transfer you to, another queue or to a live person, or something of that nature. Right? And guys, I’m seeing a couple of hands up.

So, if you have questions, go ahead and pop him into the chat. There was a question. Will you be sharing some slide decks? I personally am not sharing slide deck.

This is a live hit training So boom, there you go.

Agent Assist and Coaching Cards

When we look at agent assist okay. So we’re talking into we’re going into the during component. Right? When we talk about agent assist, We’re talking about things like coaching cards.

Right? So when I am let me give you an example, on the weekend. Right? I actually volunteer for nine eight eight, which is, actually the National Suicide hotline.

Right? I spent about four hours, true story, spend about four hours, Saturday and on Sunday, contributing to this cause. And when I am interacting as an agent, with an actual person, these coaching cues are coming up on my screen, actually coaching me, telling me what to potentially say next. It’s actually popping up very specific cues, as to potentially what could happen next, what actions I should take, things of that nature.

So true story, but that’s, something like coaching cards. But something like artificial intelligence is not only picking up what customers are saying and how agents should respond. It’s actually also proactively providing resources, and then based on the success of those resources leveraged, it is also proactively updating knowledge bases.

Resources within a company’s environment to improve what the experience is like for end customers. Right? Because if it’s updating all of these different coaching queues and knowledge resource, base, knowledge bases with those resources. Right? That means that agents actually have the most up to date information that they need to provide the best experience humanly possible.

Quality Assurance and Quality Management

Right? And so The next one, here is QAQM.

Right? Now quality assurance, quality management. Now this is super super important.

Excuse me. And the reason for that is when people say when customers say that they want to learn what the voice of the customer is, It’s really interesting because they say, yeah. We we, you know, we listen to a few calls. We follow-up and read emails.

And we, determine what the sentiment is and our customers love us, or the quite the opposite. Our customers hate us. What do we do about it?

So QAqM, the job QAqM is literally that. It’s to assure quality.

And to really look into how to best manage that quality. So let me give you an example, in a company that’s got about two hundred to three hundred contact center seats. You’re going to have roughly maybe twelve supervisors or QA analysts. Right? And their one job is to go through all of these different interactions that come into a business contact center or unified communications or whatever it is, right, just a business and take apart and analyze what happened in that call. Everything from did an escalation happen? How did the customer sound?

How did the agent react? And that entire process takes forever.

I’ve actually done it, and it is dreamly painful, because you literally have to listen to a call, probably almost double the speed, and try to determine what it sounded like and actually pick out different moments within those calls. And the problem with that is that as a human, You can only take so much of that within a whatever day that you work, right, nine hour a day or so.

And it’s virtually impossible. To, actually analyze all of the interactions that come into a business. That’s just how it is. And so now there are AI tools out there.

That will actually go in, and completely analyze what’s coming into the business and surface the most important insights for the business.

So when we look at that, it’s all about insights. Now This is really important, okay, because if you think about it, the c suite is making business decisions based on insights. Now how accurate are their insights if they’re not using a tool, and they’re just depending on a small group of humans to analyze hundreds of thousands of interactions that come into a business every single month. It is literally impossible. And so as you can imagine, you got of these businesses, who are making big decisions without actually reading the data or the voice of the customer.

You experience this all the time when you are stuck on a call, on hold for more than ten minutes. It is very clear that that company has some work to do with regard to its processes and they’re not levered they’re probably not leveraging technology, to to help them make those data driven decisions.

So hopefully that makes sense.

And so That said, that’s kind of the breakdown of the three. Okay? So about, what, sixteen minutes into this call now I’m gonna do the Windex commercial again. But next, we’re going to head actually into some pretty interesting use cases.

Use Case 1: Conversational AI in Travel and Hospitality

I’m not going to share slides. Again, guys, for those of you who joined a little late, this is our hit training.

And I am just using the light board this morning.

So when we look at things like use cases, okay, use case number one.

And this is one that I was actually directly involved in, so this is pretty cool. So it was a conversational AI use case in, in travel and hospitality. So it’s a travel company.

Well, it was more hospitality if you think about more like a how do I say it? Let’s just say that they put together a lot, a lot of vacation packages.

Okay? And the goal was to better manage operational costs. Okay? Now This particular business, if we think about vacations and such, if they’re selling vacation packages, okay, that means they’re experiencing some peak time, probably in the summertime, probably during the holidays.


And what happened during those times is that they had to hire a lot more employees just to handle interactions that were coming into the business.

And so what we went ahead and did was we helped this company, evaluate different conversational AI or in other words, some IVA interactive virtual assistance or agents solutions to potentially implement so that they didn’t have to worry about hiring and changing the, the workflow throughout the year. Right? They wanted to just have a base of really, really good employees, keep those people throughout the year, and then have something to help them manage those, like, pee ebbs and flows of business. Right?

And so essentially what happened, was in after after implementing the conversational AI tool, Right? They’re actually able to significantly reduce those operational expenditures because they had all these interactions, right, all these inter actions coming in. And what the tool did was it took all of those interactions and said, okay. I’m going to handle x y z, but guess what?

For these because they need somebody to actually speak to them, and it’s urgent or it’s a complex issue. These then went to, the small group of agents here who were able to handle those. So just to give you a visual, the importance of having something like conversational AI to really assist is to help with any of the mundane tasks.

And maybe not just the mundane tasks, but also conversational AI is so much smarter these days in that it can actually make proactive recommendations based on, a client’s profile. I know. It’s like super creepy. Right? But if you think about it, every time you’re on social media, you’re like scrolling through your Twitter or your Facebook or whatever it is, and you see those ads every time you spend three to five seconds on those ads.

It’s going to basically mark that as potential interest And then you’re going to start seeing more and more of those ads.

Another use case is when you’re talking to someone about something and then you see it in your feed, that’s not by mistake. Okay. That’s actually real. It is.

Yes. It is kind of listening to you, and it’s showing up in your feed. So as you’re going through your feed, now you’re gonna oh, no. I gotta skip it because Sam said don’t spend time on it.

But you can also test it. I’ve tested it several times. I’ve tested it at events where I say something to a person on purpose, And I’ll say it into the phone a lot of times and the next day I’ll get an ad for it. So, it definitely is something there.

So with that said, we could see that these AI tools are becoming smarter.

Effectiveness and Efficiency of AI Tools

But in that sense, they’re becoming much more effective in helping us in increasing overall efficiencies if you think about it. Right? Because we’re able to to delegate these these particular interactions to a group of human agents, they can focus on the more complex issues and really focus on you know, things like increasing client loyalty, all of these other things. Right? And, the beauty of this conversational AI is that it actually sat on top of what everyone was using.

So they didn’t have to change out anything. There was zero interruption in what was going on.

Now, the last use case I’m gonna go through, is one of our favorites, retail.

Alright? So retail is a fantastic use case. The retail such sector, anything having to do with support is really, really big. I would say the number one AI use case right now is actually anything having to do with support because of this purpose up here.

Where we’re taking some of those interactions and moving them to an actual live agent. Right? But only the ones that need to be. So let’s head into this one.

Retail is one of my favorites, and this was a really big candy company. I was actually talking to them initially. And the customer, wanted to. This was so interesting.

Collect feedback. And the way they wanted to do this was through surveys. Not the biggest fan of surveys. Okay?

Let’s say, I think it’s less than, again, stats totally made up, but pretty realistic.

With regards to surveys, I think it’s, like, less than two percent of people actually complete them. I mean, who really stays on the phone and can when, like, finishes the survey. Not me. Soon as I’m done with the phone, I hang up.

I don’t even like to use the phone. I like to text. I don’t even do surveys over text. Go figure.

Right? And so, that said, in this use case, I actually asked the customer, which was the head of digital engagement, I said, what’s your goal in adding surveys. He said, well, we want feedback. Right?

We wanna know, like, you know, candy shipments. And I’m like, you really think people are gonna stay on and talk to you about, like, the candy they like.

He was like, no. I well, but, you know, we need something. Like, okay. So, what we went ahead and actually did was put them through an evaluation of QA or quality assurance and quality management solutions to help them actually capture the entire voice of the customer. So this ended up being a VLC or voice of customer play, right, which was really neat, And so, they had, I think, two hundred agents, only about seven QA people are or quality assurance analysts, and not only were we able to tell them what candies were the most popular across the year or where they were having you know, interactions around shipping questions and what have you, but we are actually able to also implement real time agent assist So say, for example, if during a call, and believe it or not, people call about candy, they still call about candy.

You know, when someone calls out candy and mentions a particular candy type, it’s gonna pick it up and actually pop up a special around that particular candy on the contact center agents screen to talk to that person about the special so that they’re most likely able to take advantage of that special in real time.

Also the beauty of this use case, it sat right on top of what they were using. So zero interruption. We actually ran a proof of concept as well, which actually a lot of these AI companies will do just because this technology is so, so new.

Conclusion and the Big Question

Okay? So those are the two use cases I’m going to leave you with today. And then I did promise you, like, what’s the big question? And What I mean by the big question essentially is, you know, how do you even, like, start the conversation around AI?

It’s kind of scary. Who can I bring in to help me? Well, as far as who can help you, I can definitely assist, but even more importantly, are sales engineers, solution engineers, can help you. We’ve got three dedicated solution architects just to the CX practice.

Leveraging AI in Different Environments

Jason Low, Megan Thai, Mike Balarjan, or who we like to call Mikey B, but what’s the big question?

The big question is this. And it’s really how are you?

Notice I said how.

Right? How are you leveraging AI? Right? In your environment today. And this is a great question because if they’re not, this begs the question of how should I be doing that.

Right? But also it’s completely tech agnostic stick. So maybe it has nothing to do with CX. Maybe there’s an AI opportunity in other swim lanes like cloud, cybersecurity, advanced networking.

Right? This is just a great open ended question that you can ask your client to get the conversation started. And if you don’t wanna go in alone, that’s fine. And we’ll go in with you.

Hosting AI Conversations and Addressing Questions

If you get a meeting, we can actually host, help you host that AI conversation.

So it’s time. I know we’ve only got a few minutes, but I wanna make sure that I address any questions that came up, in the chat. I know that was a lot of information, but hopefully all found it helpful. So I’m just gonna start going through here. And, Doug, even if you wanna on too, we can chat.

Looks like we’ve got okay. We’ve got a lot of, like see, you guys are already using AI technology.

In the chat. It’s like AI.

Isn’t that amazing that they’re already using it here right here in our, in our chat window.

Exactly. Exactly.

Discussion on AI Tools and Analytics

Sam, there were a couple of questions that came up about, the use of analytics and the data, by different AI tools the questions are really twofold. Is it similar to the way that other, media uses.

They brought up Facebook, they brought up Google, things like that, or is it a little bit different in terms of how the data are used And then the second question was, how’s it best for the human follow-up to occur closing the loop on some of those results and the data that are brought by, these AI tools.

Oh, yes. Okay. That’s a great question. So the the first one okay. The first one was a little a little vague.

Clarify the first one for me, Doug, again. Yeah.

So are are the analytics involved from, AI tools similar to, you know, counting the number of clicks and whatnot that you might see from Facebook or Google, or how deep do the analytics go in terms of what the, tools are able to provide in a contact center environment.

Gotcha. Okay. Thanks, Doug. So, the data is nuts.

Analyzing Data and Insights from AI

So when we look at things like, quality assurance quality management tools that are actually analyzing the data. But even the conversational AI ones, right, they’re at the end of the day, they’re all analyzing the data. They’re really taking a look at all the different components, which which we really refer to as, like, the metadata. And that’s everything having to do with, all of the information around interaction, whether that’s your your name, your contact information, but also what specifically was discussed?

What keywords were discussed? Maybe how were those keywords used, and it dives really deep into, you know, what was the overall sentiment of the call it’s even down to did the agent express empathy. Right? Things like that that can be picked up now.

So it’s even more complex than ever before.

It’s still working on picking up sarcasm.

It’s really, really tough. But, the data goes super deep. And the good news is that a lot of companies already have a ton of data. They’ve been collecting all of this data for years. Right? Because that’s nothing new, but we’re now in the era of what do we actually do with that data.

And this is where we are today. We’re leveraging AI to look into those insights to make data driven decisions. Now, Doug, remind me of that second question.

Closing the Loop with Data and Agent Coaching

Second question had to do with, sorry. I was going ahead in the, other questions that came out.

Oh, closing the How how to best follow-up on some of those data with live agents and use that most effectively to close the loop with with maybe underperforming or differently performing, agents.

Yes. Thank you. Closing the loop. Okay. So this is essentially what we call the feedback loop.

Right. What do you do with the data once you have it? The beauty of the, specifically, around agent coaching, so let’s say, agent assist or even quality assurance quality management, right, is that’s actually where the supervisors and the queue analyst analysts live. So based on agent performance, they can actually create agent coaching, coaching forms, coaching sessions within those tools, and then deliver them seamlessly to the agent for whether that’s immediate feedback or during a one on one.

And so that’s where those tools are most effectively used, is housing all of those because they can actually pull specific examples from interactions and put them directly into coaching, sessions and deliver them virtually. And agents have the ability to also respond back to those particular coaching sessions if not done live. So hopefully that helps.

Resources for AI Opportunities

Great question came up, some of the resources that we have available. I’ll put two of these together. The first one was, what sort of materials do we have available right now? For partners who would like to either learn more or use, our materials in in pursuing their AI opportunities And secondarily, specifically, and I love this question, are we developing a matrix for AI providers?

Oh, man. You guys are, like, putting me on the spot here. We’re revealing all the secrets today. Right? No. This is being recorded.

So let me address the last one first because I think everyone’s everyone has that burning question. We are coming up with something that you can leverage. It’s coming out very soon, but, something that you can use to help guide you in a live, the customer interactions where you can actually navigate, answer questions, and it’ll come up with recommendations, on that kind of stuff. So that is that is coming.

Tools there are coming. But also we do have a multitude of tools. If you have not taken the AI training in Telarus University, I highly recommend you go there. It’s very comprehensive.

It actually talks about things like LLMs and what have you, more of a sort of generic AI foundational training for you. So I’ll leave you with that.

Cold Wallet Technology and Other AI Applications

Excellent information. We did have a couple of questions that came in about, cold wallet.

Devices that store cryptocurrency private keys offline.

Are you getting many inquiries from, our agents and their customers about cold wallet technology.

And is there anything that we can share with them in terms of particular suppliers who address that?

Yeah. Oh, cold wallet. Okay. So that’s more in the realm of, like, a cyber security, and maybe even, like, believe it or not mobility IoT, which is act actually now advanced networking. So I would encourage, you know, those folks to to look more into that swim lane, because as soon as that hits my queue, I actually pass that off to, like, a Jason stein and Jeff half code or even a Graham Scott, who can assist with that realm?

Very good.

Empowering Technology Advisors with AI Insights

Tremendous information that I I think most of our, technology advisors feel a little bit lost going into the conversation and you’ve given them some great resources to, better help themselves understand AI. Tilaris just released, and we’re gonna talk about this a little later in the call. A new guide to AI, and we just recently had our AI summit.

Key Takeaways from AI Summit and White Paper

Do you wanna just summarize some of the, takeaways from the summit and what’s in that white paper that they should be looking for.

Yeah. Okay. So a lot a lot happened in the summit. Which is really exciting. And we do have a white paper follow-up. If you haven’t seen it yet, definitely talk to your SPDM or who, you know, whoever your Telarus rep is.

But lots of different insights because the beauty of it is you actually get AI insights from all of the different swim lanes, and that was the purpose of AI summit. Was to actually help expose how AI is essentially leveraged in every environment, not just in CX, but also in cloud cybersecurity as well as advanced networking.

So I highly recommend you take a look at it, but also if you want to kind of pick it apart and, you know, come to us and and and, you know, discuss specific topics as they pertain to maybe some customers that you have. We’re more than happy to help there as well. That’s a great question, Doug. Thanks for bringing that up.

Conclusion and Call to Action

I think that’s pretty much it for the, the questions that we, have time for today. Last words to you, Sam. Great presentation. What would you like to send our partners away with today?

Oh, this big question. This question, ask somebody this question today, and I guarantee you you will get a meeting about it. Just learn the art of the pivot, ask the question, propose a date and time to ask, and, date and time to meet, and then bring in your resources and and let’s roll there.

Awesome. Sam, as always, tremendous presentation, we could sit and listen to this all day long.

Really appreciate when you join the call. We’ll have you back very, very soon. And if anyone on the call receives a shamwow or a bottle of Windek in the mail as a result of listening here today, we’d really like to know about it.

Thanks, Sam. Appreciate it.