A (0:00)

Hi guys. Tom Leonard here. You've heard me break down Contact Center Technology on the podcast time and time again. Now it's time for me to help you with your challenges. As your Contact Center Technology advisor through our new company, xpiva Digital, I can help you with things like CCAS selection, any type of AI implementation and nice studio and integration services. It's the same honest advice you hear on the Geek, applied specifically to your operation. Visit Expedia digital.com to get started. That's Expedia digital.com so welcome back everybody to another episode of advice from a call center geek. The Call Center Contact center podcast. We try to give you some actionable items, take back in your contact center, improve the overall quality, improve the agent experience, hopefully improve your customer experience as well. For those of you who do not know me, my name is Tom Laird. I am the CEO of XPV Interaction Marketing. We are a USA Contact Center BPO located here in in the United States. Also the CEO of Auto QA O T T O Q A which is a AI powered platform to fully score and give analytics and insights into all of your Contact center calls. I know I've been, I've been saying this same kind of thing in the last probably four or five episodes, but you know, there's been so much going on here when it comes to again, just personal life. Like I coach high school basketball, I'm coaching my kids 5th and 6th grade basketball, work with Expedia, looking at kind of some of the AI things we're doing there, looking at Expedia Digital which we're doing, you know, manage services just on the nice platform. So again, if you are a, a nice customer and you're having trouble routing, you're having trouble with studio, you have questions about the platform, you know, we're here to help with that. And then also again, Auto QA has just been doing amazing, you know, multiple customers. Now we've rolled out redo which, you know, if you don't know about it, check it out on the, on, on auto QA.com really cool stuff that we're doing. And again, I think that's a good lead into what I want to talk about today. We need to kind of think and be more creative about AI in the Contact center, every single CCAS platform. And again, I know I've done a newsletter on this, I've done multiple posts on this, but this is kind of the, I guess maybe the most robust conversation I want to have on this is we're just the CCAST platforms. These technology Major technology partners, they're, they're building the same stuff and it's kind of boring to be honest. Right. There's a race to, I guess, you know, use our agent to take over your customer support. Like that's the race, that's the only race that people are running when there's a ton of different things that we could be doing. And I think we got so excited early on that people AI could take over for the contact center agent that we forgot about all this other stuff. That really is a cooler use case in 2025, 2026 for AI. You know, it's just funny I, I see so many of these companies, you know, they talk about you're falling behind, you're falling behind and the only way to get ahead is if you use their version of the, their magical AI, right? Don't use anybody else's but they, this is the AI. And again, I'm going to do a video at the end of the year on kind of the, what I, I see as the real. What I see as kind of not real what you probably should be investing in and looking at in 2026. So kind of like a lead into the, to the next year. But what I want to do today is what I've been really harping on and talking about if you've, if you've been following me and I want to get into it at a much deeper level. But these are the tools I wish CCAS companies were actually building. Like this is the stuff when I talk to my friends, excuse me, in the industry, when I talk to other BPOs, when I talk to other call center managers, anytime I bring this stuff up, they're like, dude, do you have that? Where can we get that? How do we do that? And the companies that actually realize that there's more than just agentic chatbots to take over your support and give a, a customer a, another option for self service when, when we realize there's more to it than that. Those are I believe, the companies that are really going to thrive. And if, if the nices or the Genesis or the Talk Desks or the Five Nines or the U Jets or any of these, these big Amazon Connect, any of these big CCAS platforms actually put resources towards this, they would race ahead because their platform becomes easy to use. I don't care what anybody says, it's not easy to use. AI is not easy to implement. Not yet. In 2025, 2026, when you're using these, these, these large platforms, right, it's very, doable it's not like it's brain surgery, but it's not a, it's not a just, you know, set it and forget it kind of thing. There's a lot that goes into it. But anyway, so here's, here's, I don't know, 5, 6, 7, depending on how deep I want to get into this, of things that I would like to see. Ccas, this would be my AI Christmas wish list, right, for these CCAS companies. Number one would just, it would, it's probably not being built because it would put a lot of people out of business, but I don't think it's that difficult to do when you have the resources of some of these large CCAs. And number one is kind of an AI flow or routing builder that you could prompt in English and really deploy quickly. Meaning for those of you. Well, every platform has their own kind of programming language or it's not really programming, but it's a way that we have to tell the ccast, right? This is how I want calls routed. This is the business rules I want to use. And right now you can't do that in English. You have to have somebody who's at least skilled or trained in kind of that language. And NICE can, that's what we use. They call it Studio, right? So in Studio you can create objects and you can go in there and make sure that you know, you're routing and you know, hey, I want this to route three times, then go to this agent. Then you know, look for these hours of operation they go like all of that kind of stuff. And it can get extremely complex depending on when you go to self service to, you know, to an agent. Then you're looking and you want to send it to specific agents based on certain criteria. You can round off CRM data like it's, it gets robust, right? Normal people can't use the platform to its fullest extent. So there's no reason that we, we. There can't be an AI agent or some type of AI tool inside of Studio. And again, I'm not picking on NICE because no one has it, but I just know NICE the best, right inside of NICE that you could say, hey, build me a script that checks if the caller is a Repeat customer within 30 days. You know, it sends returning customers to the same group that they just interacted with. Maybe it prioritizes kind of the top tier, the gold tier members, sends password resets to self service and escalates any call that exceeds 90 seconds in queue to maybe a retention specialist or Some type of someone who's going to handle a more difficult caller router after call to voicemail and an email and transcribe it to the account manager. Like all of these things I want to just be able to tell it to do that and then have IT build its workflows out for you, right? Just like Claude code. If you need some type of API connectivity or API endpoint, it asks you for it tells you where to go, get it right? Managed services doesn't have to happen. Like I'm in managed services. I would still prefer to see something like this but this would take a massive swath of people probably in companies to doing other things. But this is the, the probably the number one thing that I think I would love to see built that would allow access to your platform to everyone, right? You don't need to be a huge IT technical person. A lot of these smaller contact centers that we're working with, you know, for auto, they don't have this, right? They don't have that real technical person. So they never use the platform to any extent. There's no reason that this can't be done. I'd love to see that being with. To be. To build, right. So again I just, you know, the real opportunity we everybody keeps still thinking is. Is the. The another chat bot but something that builds routing intelligence with English I think is. Is a. Just vastly more important and more valuable in 2025, 2026. You know, CCAs down the road, maybe not. But the second thing is something that I've been harping on for a really long time and we have it in a small kind of micro way, right? Which is, you know, kind of an AI workforce. And I hate the word orchestrator, but I'm going to use it an AI workforce orchestrator that just goes beyond forecasting like having a mission control. Having AI actually run your call center. You set the business rules, right? And in English you say we need to keep service levels above 85% for our sales skills, 80% for service queues and 75% for general question. You know, maintain an ASA of 30 seconds for this. We need to have sales conversions here. Please prioritize call to our highest converting agents, right? Keep occupancy below or above 72% for the call center. Right. Having all of these business rules set and then having the AI actually move agents inside out, use virtual agents when they have to. No more of this just having somebody staring at a dashboard moving agents in and out when they start to see queues, right? We can start to do. And we're looking now at like predicted queue times but adding in your workforce management, right? Seeing when, when queues are probably going to happen, be able to be preemptive agents to bring them on. Right. That 30 seconds or two minutes before we think that we're going to have a predictive way queue looking in the IVR for how many calls are in there. Moving people before the actual service level takes a beating. Like all of that stuff I think is vastly more important. And where we are as a contact center as a whole, even people that are using, you know, a lot of agents for AI still have a lot of agents on the, on the human side. Right. And, and there's no reason why AI can't be introduced to use to actually run the center. Now you guys have heard me talk about again, I'm going to go back to nice but workforce intelligence WFI where you can basically do that a little bit, right? You can say hey I have new agents on the phone. If they go over six minutes of handle time, please then to lower their priority levels so they don't take as many calls, please make sure that I get an email. If service level goes under 62% please move reserve agents in if the predicted wait time is going to go under or the predicted SLA is going to go under 75%. Like we could do this in kind of many ways and you can build that out, but it's really complicated. It's really robust. I'd love to see somebody actually build out the tool that connects wfm, connects your staffing, connects your, your, your acd, right. And is able to then manipulate and take all of that data, learn from it and then staff your contact center appropriately to make sure that we're hitting certain SLA service levels ASAs. I mean even getting into QA and looking at what agents are doing, what right to prioritize certain specific agents for certain things. Right. That goes into studio right where we talk about. So 1 and 2 are kind of like tied together to have really everybody talks about we're an AI powered contact center. I don't even know what the hell that means. That means you have some AI chatbots. That means you're using Agent Assist. I mean I do it too. That's what we say and that's what we're doing. But I don't have this. This is a true AI powered contact center where AI is actually running the center. I don't see why we haven't thought that. And somebody's not really going to town on that especially because the pieces from a workforce intelligence standpoint are there and just as a cheap plug. Not a cheap plug, but like something I always say if you, whether you have Nice or 5, 9 or Genesis or any of those platforms, they have a WFI type tool. If you're not using it, it's normally totally free. Make sure that you, you ask about that and, and you start using because we use it all the time. So I'd love to see this kind of like WFI on steroids. Okay. We've talked about, you know, having predictive routing forever, right? I remember, you know, I don't know, probably six, seven years ago at Interactions at Nice, there was a whole thing on, on really using personalities, right. And using Personas to route calls. What that never really, never really came into fruition. But I think the idea is there and the idea is good, right? But we need to make the, that next available agent that, that type of process that needs to go away, right? We can get with AI predictive routing that's not procedural, right. We can use sentiment and tone to, to understand emotion of that customer. We can, you know, no agent empathy and, and we can match customers to, to agents that historically deliver the best results for that type of, of personality type. We can look at the notes and see if somebody's been irate we their tone when they're in the ivr, right? We can look at, you know, their conversion likelihood. Like all of that analytic data is there, right? We have sales and retention trends. We kind of understand what customers buying patterns are. We understand we can look at what they've done on social media, right? So to even to, to assign lifetime value to, to certain customers and again this is somewhat being done within certain sales tools, but within, when you add in kind of that routing aspect, getting them to the right place, we shouldn't just be going to the next available agent, we should be going to the matchmaking of that customer, right? And I think that would be, that's, that's really cool. Again, using you know, using CRM data, using their, their current sentiment that we, or maybe they're talking through in the ivr, their conversion history. You know, we consider the agent's tone, their energy, their performance that day, looking at their, their metrics for that day, right? Are they selling? Are they, are they into it? Like all of this stuff we can kind of start to really hear and listen and then make AI powered decisions on what agents should be available or would have the best opportunity for success, right? That's, this is more Difficult, but with the tools that are out there, it's doable. And if somebody really wanted to put their time into it, this would be an amazingly powerful tool to make sure that we're trying to get the agents to the right place. Cooper, thanks, man. That's the new. The background is the new chat GPT 5.2 kind of their new image thing, which is pretty cool. It's much, much better and it spells things right. The fourth thing that I would love to be able to see and I think would be very valuable. And we have started to build an agent for this just internally. It would never be able to be sold, but to have AI dashboards or dashboards for our contact center built by prompt. So right now everything is very static. And I think you have most of you, a dashboard, you know, you have your sla, you have your average speed of answer, you have your agent kind of profiles, right, of what your agents are doing. You have your, you know, kind of your, your workforce, management and kind of your, your, your historical trends. Like you, you have all that in a dashboard. But you know, I, I hate to, you know, have to be able to pull up, create dashboards, look at filters, especially if there's certain things, again, in a BPO world, our customers ask us, and they have these dashboards too, but they'll ask for specific data, right? Or a specific question. Like I want to just be like, hey, you know what? Show me the handle. Time by sentiment over the past seven days compared to conversion rate, right? Add in call volume and service level by skill. Drill it just to my sales team. Show it by the hour. Highlight where CSAs CSAT has dropped more than 5% week over week. What skills has that happened in? And then we get to the point where it's proactive, where it starts to tell you some of the issues that are happening or some of the good things that are happening as well. So we move again from not just having a static thing that we have to look at and make our own rationale to asking questions, which is kind of the, you know, the, the first part of AI, that kind of generative model, right, where it's going to kind of generate answers and then get it to kind of the agentic model, right, where it's telling us, hey, service levels were down yesterday in this, our occupancy was here. You know, there's a lot that we could do, I think, with that. And then, then it ties back into our, our AI powered contact center, right? Hey, I'm going to make these changes now. Should I have them make that change? Right. AI powered contact centers don't exist. So when everybody says they're an AI powered AI driven, it's all garbage, right? All they're saying is again, they use some chatbots, they use agent Assist, they probably have auto qa. I say we are an AI powered contact center. That's what we're doing, right? This again is truly, you know what, what I think contact center managers would want to see people who run contact centers, right? They want AI to make their life like, hands off. That's the whole point of this thing. And what we do is we just make AI more difficult. I don't know, it's frustrating, but I think that would be very, very, very cool. And again, to have it start to prompt us for things that it's starting to see instead of just right now, it just highlights things in red. Imagine saying something like, hey, show me where we're winning today. Where are we bleeding? What's the trend for tomorrow? How is, is is call volume forecasted to go up this week? Is this forecast to go down? Right, right now we got to go and we have to look at different dashboards, different reporting. Let's have that all tied together. Let's be able to prompt a report. The next thing I think is super cool too is, is, you know, having agent Assist be reimagined. And I know there are some companies that are working on this, right? So. But have it turn into real time coaching, right? Again, it's kind of moving from that generative to the agentic model, right? Where it's telling you and then it's going to do a little bit more, right? So to be able to say, hey, you know what, you're speaking too fast for this customer. The frustration level, the CSAT is dropping right now. Make sure that you acknowledge, you know, before you solve this problem. Right? Your empathy improved 12% yesterday. Your sentiment went up blah. You're doing a great job with this. And again, I'm not saying that we're in the ear because that gets super annoying and you're just given tons of information, but instead of just kind of bringing information to the agent, let's take that to the next step and have a real time coach, right? Have somebody that understands the business model and the business rules of the company. So this can't be just generic. If you're a sales company, we're going to focus more on type of the sales things. If you're a customer, support and your empathy and those type of things are really important. In csat, we're going to focus on those type of things, right? And to be able to build that out and listen to how you want your agents to really talk, how you want your agents to actually interact with customers and then guide them right when they're going off the rails or give them a hey man, that's freaking awesome. You're doing great. So again, looking at AI as a coach, not just as a assistant, right? Not just as kind of a scripting tool that kind of pops things up, but to be able to, you know, agents see how their, you know, their words, their tone, their pacing affects customers, I think that could be really, really cool. I know that that's the next level of what's coming there. But again, it's a lot of outside CCAs, it's a lot of technical companies that are focusing on this that are, that are doing it. It's not with directly into your ccast. This is the next one is, is just a pain for me. I don't know how many other people deal with this, but I want real time billing analytics and transparent pricing within the CCAS platforms. Right. I mean I, I, I get very frustrated again with the whole I hate the seat licensing. I know people say that everybody loves it. I, I don't, I think it's confusing. I hate minimums. Thinking through what my minimum will be. You know, I want to see what skills are the, without me having to go be a analytic guru. That I do, and I do this every month is I look through our entire bill and we give that to finance and we know what skills and what programs are costing us the most and we try to, you know, look at that. But I don't want to do that. It would be great to not have these monthly bills, to have a real time bill of what is happening in the contact center. What is actually costing us money? What have we spent today? What is the forecasted cost by channel, by queue, by scale? What is the projected month end cost? Again, I can build that out. I can do that in Excel. I can take that data with an API and pull it and then manipulate it. I don't want to do that. Why can't it just be here? You know, you know what skills are driving the most expense? Do we have too much stuff on our platform? Right. What is the utilization of analytics for this specific skill or this specific customer? All of that stuff. And that's probably more of a BPO thing but again, to constantly looking at, at how we're, we're, you, we're Utilizing our agents. Right. So even like looking at utilization, you know, trying to identify waste better, you know, flagging anomalies, things like that, that I want to be in real time, I think is that would be a huge benefit to customers of the CCAST platforms. CCAS probably doesn't like it because you're going to be more knowledgeable about what you're actually paying for. But there's so much waste that does go into all platforms, especially in the BPO world when I have certain customers and it can get tough and I try to stay on it as best we can and finance does too. And I think we do a pretty good job of it. But there's still, you know, depending on what a customer wants, we still have agents that maybe take calls from different customers based on a type of day and maybe they have analytics on their platform. The other thing I hate is why can't we just use what we pay for? What we use. If I turn somebody on on the 3rd of the month, I got to pay for them for the month. Why? I mean, I know why, but there's no reason to do that. How about I just pay for what I use instead of paying for this month to month? Like again, that's the whole CCAs, that's kind of a. Everybody does it and I think it's, it's kind of garbage. We don't have to do it. They want to do it because they make more revenue from it. But again, if I had a seacast platform, I would not do that. I mean, and the one that doesn't, the one that really can explain that well about, hey, we don't do this anymore, it's going to be massive for them, but nobody thinks that way. It's just, it, it's crazy. What about you guys? Like, I'd love to hear anybody who kind of made this all the way through this 24 minutes, I'd love to hear some of the things that, that you guys would, would like to see. You know, I think we've, we've come a long way, right? I mean, real time QA is another thing, right, that, that we're working on with auto. We're not there yet. I think would be, you know, instead of waiting afterwards to be able to score calls as they're happening. And I think that that ties into agent assist and I know some of the agent assist guys are kind of working on that to kind of tie the QA and that together. I think that that is coming in probably in the next year or two, you know, I love what we're doing with Redo. Like, to be able to re. Have agents retake the calls that they, they struggled with or they didn't do or had a difficult customer to retake that call with the AI avatar is something that should be in, in every single platform. Like, I don't understand why that wouldn't be part of your, your QA process. And then to be able to build libraries like we're doing with Redo, right? To, to. To keep the. Hey, here's five different types of calls that all of our agents always screw up. So we're going to make sure that all of our agents take these calls before they go out on the phone so they don't make the mistake. Right? To be able to take 15 to 20 calls with AI before they go on the phone. Like, that's the stuff we're building out with Auto. Again, I can't do everything, but we're trying to do really cool things that are outside of the damn chatbot thing that everyone seems to be doing. And, and again, no differentiation. Pretty boring. I don't know. So anyway, this is my list. I'm gonna. I'm sure I'll be adding to it next week. I'm for sure. It's a pretty chill week. You guys know, the week of Christmas. So I'm gonna talk about kind of what is real, what is not real. I think in 2026, when it comes to AI, what you should probably be looking at from a. If you have budget for, for AI. And I hate saying that budget for AI, it should be budget for problem solving or some problems that you want to solve. But I know AI is such a. A huge thing that we gotta always talk about it and we always have to invest in it, even though sometimes we don't even know what the hell we're doing. So again, thank you, guys. I apologize for not being on as much. I mean, it's been a couple months. Like, it's bad. I've been really bad this year, but it's been a crazy, crazy, awesome year. I hope to talk to you more about this in the coming years. And not from a sales standpoint, but just some of the really cool things that we've learned, especially with Auto and with Expedia and some of the AI fails that we've had. We've had a lot. We've had a lot. We've had some really massive wins as well. So I'm going to try to go through some of the things in our BPO that we've really struggled with some of the things that I think are real are not real. Again, we'll continue to kind of move, move that list because that list is evolving very rapidly and yeah. So anyway, let me know if anybody's listening to this, please DM me, please follow me on, especially on LinkedIn. Just, just search me up Thomas Laird and any questions on Tick Tock as well on Facebook, on Instagram. Right. All of that stuff from advice from a call center geek is there. So check it out on Tick Tock we have probably 300, 400 videos. YouTube has about the same. And again, this will be episode, going almost to 250 episodes of the, of the podcast. So again, thank you guys very much. Stephen, let me see if I can put this on the screen here. Intraday. What if modeling informed by all the intelligence available in your. Yeah. Yes. So I mean that's kind of what we're saying, right? To, to, to have the intelligence of, of everything that you own. Right? To be able to. And again, that's, that's the, the promise that we were told, right? With AI, we were told, you know, it's going to be able to look at, you know, from a customer facing, from a customer standpoint, all the CRM data, all the notes, everything. It could go out onto the Internet, it could search instantly, you know, all their social media posts about certain topics. It could then be able to kind of predict, you know, where this thing should go or why the customer is calling. And I think if it can do that, then there's no reason it can't get into our contact center stack. Right? Everything from workforce management to the CRM to the ACD to any analytics platforms, any analytics tool to qa, have all of that data set and then start to make really good decisions on top of that. That's the AI people want, right. And again, maybe we're just not there yet. But I'm just tired of CCAST telling us that the AI that everybody wants is the chatbot, which I think some people want that. I think that's one spoke of the tool and it's a pretty important one. But there's all this stuff that's being left on the table that for no reason nobody wants to build. And I think if somebody actually starts to do that stuff, they're going to be massively successful, especially in the next three to five years. And again, think about how many chatbots are going to be available in 2027, 2028. I mean, I don't know, maybe there's a ton of them, or maybe they all just kind of move into one platform that's, that's doing great, but there's very little differentiation. It's the worst thing to, I believe now start to build. Everybody's doing it, and there's so much opportunity in these, these really cool kind of niche things that people actually want, people actually need. And it would make people's lives actually easier, right? Instead of just saying, hey, our ROI is going to be there or we're going to lower cost, but how about we actually use AI for other things? I mean, the ROI is going to be there, but how about we make life easier? How about we make it better? How about we make the agent experience better? How about we make really cool tools that are AI powered, that we can really change how we do customer experience? I don't know. That's me. But again, 30 minutes is. I'm way over time. So again, thank you guys so much. Hit me up if you guys have any more questions. I'd love to talk more about this and I'll see you guys next week. I think next week you'll find very interesting as well. You know, again, what, what I think in 2026 is real from what people are saying and all these vendors and when you go to ccw, what is real, what is not real, and what you probably should be investing in. And there's a lot of it. I'm not an anti AI guy. I'm for it, but we just got to do it the right way. Thanks, guys. Appreciate you.