Let's Get Real: The True Story of AI in CX & Your 2025 Game Plan

00:00Tom the Laird here, you've heard me breakdown contact center technology on the podcast time As your contact center technology advisor, through our new advice from a call center geek, a weekly podcast with a focus on all things call center. We'll cover it all from call center operations, hiring, culture, technology, and education. We're here to give you actionable items, to improve the quality of yours and your customer's experience. This is an evolving and creative minds and ambitious people, like- like this guy. Not only is his passion call center operations, but he's our host. He's the Interaction Marketing Group, the Call Center Geek himself, Tom Laird Welcome back to another episode of Advice from a Call Center Geek. The Call Center, Contact Center podcast, we try to give you some actionable items to take back into your Contact. Center, improve the overall quality, improve your overall customer? Experience and now technology as well.

01:14My name is Tom Laird. I'm the CEO of Expivia Interaction Marketing and Auto of the two, one, ones across the country and the two. One ones are kind of like the non proud to be able to have, you know, a force that can, kind of Help there, right? And I do a couple of weeks, I think I kind of just did one, really just a couple of weeks ago on the outsourcing side, but there's so much just just non truth, misinformation, I hate that word, but misinformation, people scamming people trying to get into people's heads, a lot of FOMO, that's happening as it comes to AI. And how's it relates to CX. I did my largest, it was a, top 10 most viewed post that and I've ever done on LinkedIn, talking about the word that everyone is using is complete garbage and it is hyper personalization.

03:28Now if you go, and hopefully some of you guys can do this right now, go on to Clod or go on to ChatGPT and type five predictions for customer service in 2025. the number one thing that will pop up is hyper-personalization, whatever. he I he To contact center technology companies, all using, I think they're using ChatGPT to do all of their content. And so, as a contact center professional, and those of you who, a, lot of you who are following me are, that, you're trying to figure out what do we need to do in 2025. What is the... the thing that we're seeing, where is the industry really moving? And you're getting lied to. You're getting lied to by a. So many people, right? The problem is we are being driven in Chris Crosby, who I love, right? And Chris and I, a couple of years ago, talked about this as it related to kind of the beginning of AI. And I, I think it's It's only I'm not seeing anything that all And spending God theirs money on some generative chat-bot from my chatbot is amazing.ai company.

05:47stuff is where people are making huge mistakes, because they're not ready for it. And a lot of times we don't need to where you can, more answers to. So not everybody needs this. It's moving towards that way. We all have to be prepared for it. We can put our head in the, sand and I'm not saying that it is a, it's a, it's a nothing burger, right. It's still gonna be, be something that all of us are going a have to kind of work through as we kind of go through this. But you know, this rush that you have to get to there now, or you're falling that I'm the It's not true. I have said this before, and I think it's almost been cemented in me now, from what I'm seeing is, you know, AI, when, when it comes to that first touchpoint, customer facing AI, is really for enterprise, large, large enterprises now.

07:40Why is that? Number one, they have the traffic to get the ROI for it.

[Gap 07:48–08:18: no reliable transcription]

08:18figure out that ROI, because the volume is so massive. They have the data for it. They have the infrastructure for it, which most SM, SMBs, most even smaller enterprises are not ready, for this, are not ready for it. Again, the tools, again, Nate, just kind of threw a comment here on LinkedIn, but it's so important, BPO needs to, discover what they're good at, look to how to improve it, and find an AI tool that adds efficiency. And I agree, and we're going to get into that, right? Like, those are some of the mature, if we can the mature AI tools, you know, today are kind of the, the agent tools that I don't say back office, but not customer facing, facing, right? And we'll- kind of talk that through, but I want to kind of get into some of this stuff, right? I hammered hyper-personalization on LinkedIn last week, and I think all of these new phrases and terms are total garbage. They are terms that are created by people. People picked up by AI, they, regurgitated through AI on what is actually happening in the industry, and I don't even know what hyper-personalization is, right? I'm still trying to figure out what maybe what's right below it like premium personalization and then you have regular personalization like what is that.

09:46And ' some of the things, and this is in no particular order, this is kind of my level that hey, we bandwagon all of our customers are gonna want this all of our competitors are moving to it difference? Can you just tweak what you have for the right now from a from a self service? So that's that's kind of the, you know, I kind of the the deal. Please and understand that this is extremely expensive today. Set up fees are crazy expensive, your minimums for these, you know, for these large third party, you know, AI slash analytics, because most of the analytics companies have now turned into AI companies, Right, they're extremely expensive and even the the per minute usage fees that they platform or if you're going to third party. They're going, they're going to a third party guys are, they're going to integrate and they're going to be fine.

12:35They're going to be. Some of them are going to be really cool, but your ROI is not going to be there. Some of them are going to be there. if you're a smaller enterprise, it's not there. This is where, and, again, selfishly a, I don't mean to sell, but the BPOs, I think, have a huge advantage. If you, a and you need to go, and you want to do all of this really cool AI stuff, the BPOs, if it all the time. The learning curve is not going to be there. I think everybody talks about this is the end of the BPO. I think that this is the time, the time where BPOs can really thrive because they can be that AI partner for so many customers that don't really know what they're doing. So, again, talk to a BPO, talk to a consultant. I think that you're going to to see that, again, like us, we're using all the tools, right? In real life with customers right now, understand what is working, what is not working, what you should have, have. the AI platform and most people. the most they have from a data set especially if your SMB is it may be a robust FAQ, right?

14:40We need to make sure that we're starting to build data and we're getting it to the point where you have, where we, where we have a knowledge management system, not just uploaded with all, we have a process then to keep things updated or to take out old data, to take out bad that's one of the major hold backs, right? You know, the infrastructure standpoint of integrations and the infrastructure standpoint of the AI from a KMS standpoint is really, really, really important. The the other thing that I think is real and that I'm starting to see companies do is panic on AI, not sure what the heck to do but they have to do, right? So they'll either do one or two things. Number one, either they'll make a mistake and they'll purchase something that they're still scratching their head on, I don't know what we're doing with.

16:02It doesn

[Gap 16:09–16:39: no reliable transcription]

16:39the things that you can get out of that. Looking at things like again, you've heard me talk about workforce intelligence. All the Made in CCaaS players have a tool that can kind of predict wait times and move agents in and out of skills to protect you to AI, right? So being able to route off data, right? Let's just say you're integrated into a Salesforce instance, and calls come in, and we can look at that data, and find out, like, let's say, we're, you actual customer, right? And who's the client? Send it to the agent, make sure a screen pop that says, this is, you, you know, Janie Smith's office, how can can I help you? Those kinds of things, right? It's really not that hard but it gets your mind thinking, okay, there's an integration here now between my telephony and-my data. This will start to build kind of that AI integration kind of repertoire that you're going to start to need that kind of skill set that you're the start to have.

17:55I also see customers moving finally and to get into the the very easy, instant ROI tools of the kind of the back office or the agent assistance type tools. So a these are number one agent assist. It still takes a little bit of work, It still takes some knowledge, But it is a- a tool that to move to a generative or to an AI assistant model, that's an unbelievably great bridge to get you to that point of understanding, So, if you make a mistake, a lot of times the agent can correct that, it's not customer facing, but it will also help the agent and assist the agent. at nauseam about auto to be able to fully. automate your quality, quality assurance using ai at scale is a massive, massive savings, right B

[Gap 19:35–19:56: no reliable transcription]

19:56to kind uh, put things into the CR I'm very easily, but they still screw it up or they don't Slides and PDFs your computer. See here. No, maybe I can't present here. I'm gonna share my screen and bring this up. This is- and I'm gonna post this but I think this is kind of what I've seen. Be working. There we go. Let me put this on this screen. And let me share this share screen. Okay, here we go. Alright, hopefully you can kind of see this. Let, let me flip it around for my LinkedIn buddies. Okay, so this is kind of what I have the scene as what has what has been working from a, roadmap standpoint for AI, for customers that we have. And this is also what I'm seeing work and people who are trying to be really, kind of methodical with this thing and trying Looking at your KMS, looking at your What API connectivity do you have?

22:52Are you willing to kind of go through using LLMs and using large language models? Having had that all scoped out, I think is really, really, really important.

[Gap 23:00–23:14: no reliable transcription]

23:14I know I have this as a hierarchy, but for some people they're going to move between steps. But just overall, You're looking at your advanced analytics. I think that that's a good next step, right, to start saying, can you, can you glean any insights into help you from a self-service or from even from an agent quality standpoint, right? But it's still, it's using the system, it's using technology, right? From a sentiment standpoint to to really almost take away, again, some of tools that you've had, right? If you are really good at analytics and you're really good at sentiment analysis, I don't think that you need NPS and CSAT service. You've heard me talk about this a lot. I've done multiple podcasts on it. So I I think that I think that that's important. And that's a really cool next step for those of you who are just trying to kind of get involved and see where we are with AI.

24:23The next step, once you kind we'll figure it all out for you. And again, I'm not saying that this is impossible, this is crazy hard, this is, but it has to be exacting. You have to make sure that you're getting it right. Right, so again, that's the next, that's the three. If you have, in 2025, right, if you have your infrastructure built out, you have your integrations you feel good about, you've developed your KMS, you have some type of advanced analytics where you're really looking at customer data, you're looking at agent behaviors, And then you start to implement some of these agent level AI tools, like auto QA, auto summarization and agent assist. You, I guarantee you're ahead of 95% of every single organization out there, just with those three things. Where I'm talking about, you know, bots and, and, and, and, you know, virtual agents and all that yet, right? Okay.

25:44So I think once you get through there, you start to think about self-service. you start to think about what do we need to do to make it better, right? Now that we built out this KMS, can we start to implement some of these, this knowledge into what we already have, right? Looking at intelligent FAQs, looking at the really just more automated flows within our self-service. Are we at a point where we think we have everything built out where we can fully automate things? And if you do, then let's start to talk about customer facing AI. You've gone through kind of your security apparatus with your team. Everyone feels comfortable about what can be said, what can't be said, what you can use from a self-service model. And then we can start to implement some of these LOMs and start to put our kind of toe into kind of either a visual or a audio kind of bot, right?

26:43And start to look at some of these and more advanced self-service solutions. I will tell you though, I believe in 2025, maybe five to 10% of companies can really find value in the customer-facing AI compared to what they would have from a regular self-service model, right? If your self-service is you feel on point, I don't think that there's gonna be that much that you're gonna find from a customer-facing AI platform than what you have now. Now, for some of these, again, the bigger you are, I think the hierarchy works with you too. The bigger you are, the quicker I think you can move to this stuff to find the ROI from it. But remember, you need to have either the know how to do it internally, or you, you're going to pay for the nose externally with a third party, at least in 2025. And I don't,t again, know if you're going to get all the tools that you really, really want.

27:40So again, I will, I'll post this if you guys find any value in that. We flip this back around, my TikTok buddies. Okay. So that's kind of how I see again, And where 2025 is, for our customers, if they wanna go to customer facing AI, we have the tools, we, we can do it tomorrow, right? But we wanna make sure that we, we're talking to them about what do they need to do to really make sure that things are good. And I think it's really difficult if you're just talking to tech guys, you're talking to sales guys who don't have your best output to kind of really they're just trying to make you kind of make that sale, make you kind of have a little bit of and then and then move on you know I think that's kind of what I'm seeing seeing as well. Let me stop the screen here. Okay all right if you want something weird that's a little bit more kind of out there you know I have been wrong and admittedly I have been totally wrong on the metaverse I thought we would be much farther along I think maybe AI kind of stopped some of and I know that you know kind of the Apple and the meta and the the goggles that they have come out haven't been great. They've been really I think, you know, I've kind of a while and I've, last year I kind of.

30:08kind of roll. So something that's kind of a little bit out there. I wanted to at least throw one of those kind of things. It's everything I have. Again, it's kind of boring, right ai overall in CX for where we are right now and what you should be doing and where the industry is at. It's kind of boring. We are at an infrastructure build portion. We're at the bottom of this thing and I think it's going T get really cool, right? But so many SMB small enterprises, you're just We're not need to take this next year to start to get ready so that we can start to employ some of these these really cool things. The reason that I don't think you're falling behind is not like just a slope or step you go up with and you get a tool, then you get something cooler, then you get something cooler. You can wait, build, build, build, build, build, build, build, with some tool that's going to come out in a year from now or in six to eight.

31:09months from, you know, six to eight months from now, you don't have to keep going with technology today, right? Start to build your stuff out. And then you can catch up really quick and maybe even quicker and maybe some of the technologies that are okay. They went right to the kind of the end after things. Things have been kind of hashed out a little bit and you're getting something that you know is going to work in the ROI is going to be there so you know this this falling behind thing I think is garbage now if you're not working towards it then you probably are falling behind but as long, as you're you're starting to kind of think through this you know starting to build out starting to play with some of the basic tools learning about he's in it, okay, you're going to be okay. No matter what any AI, my chat.bot is great.ai, sales guy's going to tell you, and going to freak you out with, or even what your CCaaS, all these tools that they're coming out with, they're all AI-driven, right?

32:08It's going to be okay. All right, so again, if you, anybody would like to the talk to me, I'm an open book. If you'd like to talk to me about where you guys are at, Let us know in the comments below. You know, more than happy to kind of I help with that. And again, super proud and super excited about how OTTOQA is going. We are building a site that is pretty awesome, really big names, really small names. Everyone from doing 500 forms a month to a thousand forms a day. So it's been an unbelievable story and we'd love to be able to talk to some of you as as well if you're interested in using one of those tools that we talked about from an auto QA standpoint. But other than that, I hope everybody's new year kind of keeps it, I'll talk to everybody soon.