Why Real-Time Automation is the Missing Link to Effective AI

Hi, everyone. This is Vicky Harrell, executive director of SWPP, and welcome to our webinar today sponsored by IntraDiem. We have Ted Lango and Josh Wilkins with us today. They’re gonna give us information about why real time automation is the missing link to effective AI. There’s our buzzword again for the day, AI. We’re look we’re getting sets on your bingo card. We’re we got you started already this morning. We want you to interact with Ted and Josh today using the chat on the bottom right hand side of your screen. And, we’re going to record this. We’re gonna send out the recording later on. And so even if you miss something or, you want to share this with someone else in your organization, you’ll be able to have the recording for that. If you see the chat on the bottom right hand side of your screen, I don’t know if Webex looks a little different to you this morning than it did as it did for me when I logged in this morning, but the chat is on the bottom right hand side of your screen, a little chat bubble, and you can send your message to everyone. Although, it looks like, now that looks a little different as well. But if you would put some chat in there this morning and let us know, where you’re listening from and how many agents you have in your organization. So everybody can find the chat this morning and make sure that everybody knows where to, where to put information as when they start asking questions and wanting your feedback. Alright. Having problems sharing? Yeah. My, my Webex just died. Sorry about that, folks. Just give me two seconds. Oh, okay. Okay. K. So Bob’s in the Twin Cities with three hundred to five hundred. Valerie’s in Memphis at Saint Jude with two fifty. Atlanta, a hundred and fifty. Iowa, a thousand plus. Bryn Mawr, Pennsylvania, the eighty five agents. Nicole’s with Bob. Okay. Dad’s in Detroit with about five hundred ish. Alright. Four thousand with Anne in Arizona. Charlotte has seventy eight degrees and question mark agents. I don’t know I don’t know what it the the weather the the temperature is here in Nashville, but I can tell you that it has been raining for a week, and we are about tired of it. So alright. So, again, if you find the chat on the bottom right hand side of your screen, this again, this is everything looks different this morning and with x for some reason, and put in the chat where you’re listening from and how many agents you have. But while you’re doing that, I’m gonna go ahead and turn it over to Josh and Ted and let them get started. Thanks, guys, for being with us today. Thanks so much, Vicky. Hi, everyone. I’m Josh Wilkins. I’m the solutions consultant with Intra Diem. I was a former customer in my past life, so I’ve used the tool as an agent on the phones to getting training delivered. I’ve used the tool to manage it at a platform level from the command center, kind of an intraday role, and I’ve worked with people like Ted in the the planning side of things for for many, many years. So I’m looking forward to the conversation today. Ted? Good morning. Thanks, Josh. I’m Ted Lango, founder of WFM Labs. I’m now also doing my own consulting with, organization called Kyoto Solutions, formerly with Intradiem. Had a great couple of years working with the team there and about twenty years of of call center operations. Like Josh was a a past customer, of Venture DMs as well. Happy to happy to see everybody today. Love to share a few ideas that we’ve got with you about data, AI, and automation. So I think, Josh, we’re gonna set this up and talk a little bit about why data alone isn’t enough, what does real time look like, and how to really set the stage for AI. At the end of the day, we’d love to hear back from you. We’ll ask a few questions. We’ll start actually kind of looking at the landscape of of data itself. We’re we’re, as an organization, I think, you know, swimming in data. Josh, if you wanna click on to the next screen, I wanna ask everybody sort of where they are on on data their data journey. Just personally myself, I’ve been sort of through this cycle where we start off working with spreadsheets, up in the upper left hand corner over time. We may get some of our data organized and, hopefully, to some degree, delivered via dash boards where we’re not having to hack together our data. Some of that involves automation, you know, around actually how we acquire the data. And then if we’re fortunate, you know, we can go past that and start to work with things in real time itself. And I I’d I’d say I’ve seen people who live across this spectrum. It’s kind of the first question I’d love to ask everyone here. You know, if we said up on the left corner, spreadsheets, that’s number one. We’ll say some, like, Tableau or Power BI, number two in the middle. Or are you running, you know, command centers where your data is flowing through real time dashboards? What type of world do you live in? Is it one and two? Have you completely gotten rid of spreadsheets and everything’s automated? Like, number three, in terms of how you work with data, love to just see where everybody is. If you can drop that into the chat, number one, hey. I’m still stuck in spreadsheets. I don’t have anything automated. Number two, and maybe it’s one and two, I’ve got some reports that are automated, you know, or, hey. We’ve left, one behind altogether, and we just go with number two and three. I see in a lot of twos and threes, a few ones and twos. I mean, obviously, I I I if you can, break away from spreadsheets and the manual nature of those, you know, I’m all for it. Getting your data itself automated, is a huge step in the right direction. I personally have lived through where spreadsheets are sort of always yesterday’s data. Even when I automate and pull things into Power BI, I may just be looking at stuff that is, hey, from the last interval or from yesterday or from the last week. It’s great that I’ve automated the collection of that data. But if I can look at things in real time, that really gets me to a springboard to think very differently about how I work as an operation. Josh, some thoughts from you just on data. Yeah. No. Thanks, Ted. I love thinking of it in terms of day old donuts. You can still eat them, but you don’t really want to if you have a choice. So I I’ve I’ve worked in worlds where you kind of running your command center to fill in those Excel sheets. How many people have kinda gone through this journey and sort of stuck where, you need an army of people to now deal with the data? Does anyone sort of analyze themselves into a whole, so to speak? And I know a lot of people who still have to go back, and it’s like, hey. We’ve automated things, but we’re still back to leveraging spreadsheets to do that type of analysis. Yeah. Kinda like taking your forecasting out of the WFM tool and doing it in Excel on on the side of your desk. Right? Awesome. So where I was getting to with that is from a a command center perspective. If you have all of these data points, you have all of these entries into doing something about your business, about about your, your situation. If you don’t have the army of people to man those, it really tends to be a a a an issue of overwhelming. All I’m doing is sitting there in the DMV, so to speak, with a thousand customers in the line. I can’t talk to them fast enough. You can either ramp up the staffing side of things or change the process or the the technology that you’re using so you’re not just trying to solve that problem with people. The irony of this, as we kinda go to the next slide, is even if you’ve worked your way up and you feel like we’ve evolved our organization to where we’ve automated data, we have command centers, you know, we think we’re kind of on top of it, and we really have the views we need. We’re seeing right now a trend on CX that is as low as it’s ever been looking back over the past decade. At the end of the day, we’re still missing something. Forrester published this not so long ago that shows that CX, most really our COVID time, has just gone down year after year after year. So as you kind of reflect back on how we as a workforce management organization are working with data, we may feel good that we’ve gotten out of spreadsheets, that we’ve got more information to make decisions about. We may have a command center where we feel like we’re looking at things in real time. There’s still something missing here. We just see CX continue to decline, decline, decline. Now that’s not all on us, but workforce management plays pretty critical role in ensuring that we’re available for the customer when they wanna be there. So we wanna kinda dig into a little bit of this and contrast this with, you know, what are the priorities that are out there today to try to turn this around? Josh, I don’t know if you have any other thoughts on the the trends around CX. No. I’d love to kinda jump into the next piece here and open with a question, and and we’ll get to the the content here. But how many of you folks have CX as a priority from an OKR perspective? You talk about it with leadership or what whatever whatever the situation is. And I know CX is a broad term. It’s kind of a loaded question, but I’m trying to trying to see how many are speaking the language throughout the day kinda kinda thing. I get I guess turn it into a yes, no. Like, is CX a priority for you guys? Yes. Or has it dropped off? I mean, is is improving customer experience a part of the conversations that your leadership is having with workforce management? You know? Or is it like, yeah. Now CX is on a decline. Just going down, down, down, and we’re gonna let it roll. Yeah. I mean, it’s kind of a yes, of course, answer, I think, for everybody. That’s the loaded side of the question. Right? Of course, it’s a good of course, it’s a a focus, but is it budget season, and can you actually do something about it right now? And that that’s really where this slide comes from. This is a Salesforce state of service survey where they basically asked a a bunch of, leaders to what what focuses were within their company. Customer experience was one of the top, is actually the the top focus for for, different businesses. Improving data quality and reliability is another one, a close close follow there, Workforce skills. So it kinda all ties into the purpose of our talk today is setting up your environment to make sure that you have the the people processing technology to do something about all of this. The all these gobs of data is really the first step before you can start doing something significant with AI or any kind of automation. And while the priorities on the left there, improving experience, improving the quality of data, optimizing operations to reduce costs. Take a look on the right. Well, one of the interesting things here is, hey. Our top challenge, either major or moderate, is keeping up with customer expectations. Customer expectations continue to increase year over year, and that decline that we saw on the previous chart is we’re not keeping up. At the end of the day, if customer experience continues to be a priority, but keeping up with expectations continues to be a challenge, you know, that there’s something that still isn’t connecting the pressures to reduce costs. So, yep, we always I don’t think it’s ever gonna go away that we’re here, and we’re in the middle of that optimized operations to reduce costs, yet that’s one of the biggest challenges is the pressure to reduce the cost to serve. And while we wanna improve our data and the quality of it and the reliability, One of the largest challenges on this same thing, actually, the last two categories on the bottom right are poor data quality or reliability and disconnected data sources themselves. So these two polls themselves, along with another one coming up is, like, this is the state we’re in right now. Post COVID, CX keeps going down and down. It’s obviously a priority for all of us. We understand that. We still have the cost pressures. And at the end of the day, we still struggle with with data itself. That brings us right into this. And the reality is is people wanna talk to people. You can have as much automation and chat bots and front ends to take a lot of that workload off. Those tend to be simple functionality, simple transactions, and all of us know quite a few of those folks will still call in to verify or to fact check with a human to make sure that that last piece was actually done the way that they they wanted. So there’s a level of maybe distrust or or hesitation as we sort of start changing the the into the brave new world. And that that really ties into everything we’re talking about is is we need to have the the groundwork ready before you can start making these changes to enable things like automation or an AI agent that’s gonna take workload out of your call center. I I think we’ve seen poll after poll after poll that just keeps saying the same thing. This number is, you know, from an aggregate of polls that we’ve taken, read. It’s like everything always continues to say the same thing. Agents rather or, customers rather talk with agents than with bots. Now, you know, transactional things, yeah, we love automation. We don’t wanna have to talk to an agent to say simply make a payment if we can do it online. But all that stuff has, over the years, gone away to where it’s like the transactional stuff. Yeah. We know we can do it on our app. We can do it online. That stuff is gone, but what’s left is still gonna require agents. And, I continue to challenge people thinking, hey. Call centers are gonna be gone in two or three years. That’s, you know, what we heard, all call centers are gonna be gone. AI, you know, is here. The bots are here. They’re gonna take everybody’s job. It’s not happening. Yes. The population as a whole may shrink over the next five to ten years, but people are still gonna wanna talk to people. And that means we’re all gonna be employed doing the same thing that we’re doing today, maybe on smaller pools of people, but people still need to talk to people on the things that matter. Yep. It’s a good point. And orchestrating all that work, whether it be an employee or an AI agent or whatever the future holds, it’s still gonna require the fundamentals that that we manage through. So that that’s a good call at Ted. Tying into this next piece here, Chad raised a a good point about all of the buzz. I wanted to ask from a a project perspective, how many of you folks have a or sorry. Yes, no question is better. Do you have an AI project in flight right now in your business? Yes or no? We’re kinda we’re we’re laying the groundwork for the the whole concept here is that the AI projects tend to fail if you don’t have the right groundwork, footwork, footing to make sure that you have the data to feed the people process technology to do something about the automation that you’re gonna get from something like an AI platform that will. And chatbots do count to an extent depending on how you’ve deployed it. You know, they have, like, the ten year old version where it’s not so much thinking. It’s more of going through, different workflows. There’s an in between with the the future state where that’s getting a lot more intelligence. So I I would count that one, Valerie. We we we talked about, at the beginning of this data an awful lot. And even as you see AI projects get released, they’re there. They appear to be successful. Behind the scenes, whether you see it or not, a vast majority of these projects fail. They fail primarily because of the data side of things. At the end of the day, we could be saying, hey. We’re training a, a chat bot that’s gonna be an intelligent bot that’s gonna work off these knowledge bases that are gonna work with customers. And most companies, are still still have a lack of appreciation that you don’t build these things, deploy them, and then say goodbye. The data itself and the knowledge of these bots and the different technologies, it shifts over time, and that data needs to be refreshed, redeployed. It’s a very intensive process to keep these projects going. So kind of a warning sign here is as you see projects released and all of a sudden next quarter or next year, a bot’s come out that appears to be taking volume out of our call center. We say, oh my gosh. That’s great. We can start planning for less headcount. Don’t count on that bot just being there, taking that traffic, and, you know, carving out five percent just like as though you had deployed a new feature on an IVR. These things drift. They’re gonna require us to really monitor, are they performing? And they’re not gonna monitor they’re not gonna perform just in the same way like a traditional IVR utilization does over time. So as you see those projects come out, you know, keep an eye on them. Data behind a lot of the AI related projects drifts over time. Those models might perform great today. In two months, they may not perform. Or you may have, something deployed that all of a sudden does something crazy and leadership says, oh, we gotta take that down. So for three weeks, you just took three percent of traffic out, and then it gives someone a wrong answer and people pull the thing back out. So I fully believe AI is gonna continue to help both agents, and I think that’s where the the, you know, the gold is is agent assisted AI, I think, can go a lot farther than people realize to help our agents streamline what they’re doing. Deflection, yes, over time, but it is not a silver bullet, and it’s not going to change what we’ve said on the the previous slide here. That’s a good call out, Ted, because the last few conferences we’ve gone to, even the last two years at SWPP, last year was more focused on here’s all these great things that we can do. Here’s all this awesome technology that’s coming out. This year, the conversation was very much, okay. Let’s take a step back, make sure that our knowledge base is really key and and clear so that we can feed something like a knowledge management tool or a chatbot that’s gonna take volume out. So if you don’t have that foundation, all of a sudden you’re gonna have to stop, start again, look back, and think what went wrong. The better thing to do would be to focus on that as you’re launching, making sure that your your house is clean before you kinda lean into your neighbor’s yard, so to speak. Well, a couple of years, I think this is about two years ago, we developed a, in WFM Labs, a workforce management maturity model, and we’ve tested this and gone pretty deep in a lot of these areas. I wanted to bring it up again today and do another poll. I’ll walk folks through, you know, how the maturity model’s laid out here briefly. And then I’m just gonna ask if people wanna put where they are on that model, itself. And I wanna tie this into really the heart of our conversation here, which is, data getting your data’s clean, but then taking and using your data for automation, we believe, is a prerequisite to really becoming an AI rich organization. Workforce management organizations almost always start at level one. You know, this could be your call center twenty years ago when it just started up, or it could be a a new call center. It’s and it’s manual in its nature. It’s Excel based. A lot of it’s reactionary in how we decide hiring. There may or may not be a formal WFM team or it may just be one person, you know, who’s building schedules out on on spreadsheets. We define that as level one. Once you’ve grown past level one and you get off spreadsheets, what have you done? You’ve invested in, an actual technology like an IEX or an Aspect or a Verint or any one of them out there, and you start to really build out your schedules, you build out your team, you start to forecast into the future, and you go from being reactive and all on spreadsheets to really building processes, having the technology, and and and, foundation for doing workforce management, you know, in a more structured way, far better than doing things on spreadsheets. That’s what we call level two. That’s honestly between level, one and level two where probably eighty to eighty five percent of the industry sits right now. You know, we’re still doing things like prescheduling training, prescheduling coaching, hoping for the best. And then when the day comes up, we’re pulling a lot of levers to react to variance in our plan. Level three is where you introduce automation. We’re a data rich organization, workforce management. Our data is very structured. Data that comes out of the ACD in terms of handle time and call volume, who’s logged in, what are the queues, data that comes out of our WFM system. It’s all very structured. You know, unlike a lot of stuff that feeds AI, we can define handle time. We can define how many people are staffed here at ten thirty. We can pull that together and introduce automation. If you’re an Intradiem customer, you have some type of automation that is taking action in real time to make adjustments to your workforce management, you’re starting to play in level three. And we’ll spend more time on level three here as as Josh goes into our next section. Level four, we defined as an organization that starts to really look at the capacity planning. There’s two big failures in a lot of call centers, and one of them is, hey. We’re reacting. We’re reacting. We don’t have automation. We can’t plan dynamically enough. Another is something that a lot of us can relate to, which is building bad budgets and getting stuck with what I’ll call a fragile capacity plan itself. And we’ve proposed ways where now you can use a lot of tools, simulation, and do far more analysis on developing a capacity plan that presents a range of outcomes and starts to assign a risk rating to it. This is where we go from kind of deterministic models where we’re just adding up numbers to really simulating outcomes, that may happen in the future. And, really, when you go past that is when you’re ready to start bringing on things that are more advanced. You’re AI ready. You’re able to start to look at, things like employee attrition and not just trend it, but predict it. You get get very sophisticated in level five, pioneering about how you’re building very dynamic shifts. You can start to integrate things like AI with your agents and really dynamically work on a day by day basis. So, just a question for everyone. You know, you’re doing things still in spreadsheets. That’s a one. Are you in traditional workforce management software like a Verint and IEX, but still prescheduling things like training, coaching, doing all those things ahead of time? That’s a level two. Level three is you already have automation in your organization. You basically are not prescheduling training, not prescheduling coaching. You’ve given up on adherence because you’ve got automated ways to deal with adherence, and you’re like, that’s no longer worth dealing with. That’s level three. Level four is you’re doing actual, simulation for your capacity planning as opposed to taking all your data out, spreadsheets. That’s level four. And then level five is is what we call pioneering where you’ve gone, all that. And just like John said, you may be at two, but you might be playing with some of level four, capacity planning. We think of these as gates. You can do a level two and a level four to some degree. We see most organizations evolve, you know, through these gates. It looks like a lot of level twos right now, which is what I would expect because we’ve done a lot of surveys across the industry. And as I said, eighty to eighty five percent of folks today are still rescheduling and attempting to do things ahead of time. Build a plan that has all those things set up ahead of time. I’ve said enough on this, Josh. I don’t know if there’s anything you wanna add on this slide. No. No. It’s great context. And and I I like the idea of them as gates. You kinda require the having a lot of automation to be able to do a good job with number four. As an example, if I’m building a really robust forecast, I have my plan down to the interval and really, really accurate. Anything that changes throughout the day, like absenteeism or, an event happens where all of a sudden thirty percent of your call center is late. There’s a power outage in a work from home area. Whatever it is that’s gonna change your demand or your capacity, reacting to that manually is really gonna take all of your Monte Carlo simulations, that accuracy, and it kinda throws it out the window because you can’t do anything about it. And that’s really where this next piece comes in where you can start taking actions without needing to take the action in the first place. You just kinda do the setup, build the logic on what you want to happen, and then automation will take it from there. And there’s a lot of different flavors of that. The one we’re talking about is really kind of leveraging the data that comes out of the ACD and within the WFM system and then taking intelligent actions throughout the day so you don’t have to have an army of people kinda manning that that queue. And we we think as well of this as gates because as AI is coming and it’s coming, it will introduce degrees of variability into your business. And at the end of the day, if you’ve got your plan built out and it’s rigid and you feel good about it, there still is variability that’s gonna come into play. AI won’t necessarily just take that out and make it, you know, a better thing. In fact, level four is a function of AI. It’s predictive AI. A lot of us think of AI and just go right to LLMs and chatbots and things like that. But a big piece of simulation in in level four is using machine learning Mhmm. A a function of predictive AI to really get better at how we think about planning. What we what we feel pretty passionate about is you really wanna be AI ready when and call yourself level five cutting edge. You not only need the simulation around your capacity planning, but you need to break out of the mindset of trying to build these rigid plans that preplan all that activity and shift into where I can start every day and know I’ve got automation that’s gonna, you know, have my back. I’m gonna be able to pull some people off in those areas that we have lulls. I’m gonna be able to put people back in, shift between channels, and really ebb and flow. As you think forward into AI and what that environment’s gonna look like, we think now, more than ever, it’s super important to have a lot of our functions automated that we do in workforce. So tell us, Josh, about it. No. It’s a great call, Aletha. Thank you. If you if you think about Ted’s example of all of those moving pieces, if you’re doing that without automation, what you’re really doing is you’re leaving it up to the people at the end of the day to make those communications, send a Teams message out, a Slack message, whatever it is. There’s a lot of lag time sending those communications, someone reading it, taking a second to process, saying yes, responding, acknowledging, doing the work, whatever it is. When you do automation, you can kinda skip quite a few of those steps, and you would be shocked at scale how much efficiency that adds into the the equation. Even something like delivering a training session, which I’ll get into in a second, doing it in a ten minute block during your lulls where you have your overlap period or people have logged in for the evening shift, things like that. Again, you’d be really surprised how quickly you can harvest that time and do something productive with it, all without needing an army of people watching it like a hawk and then trying to forang a bunch of a group of people to go do something different. So from a technology perspective, what IntraDiem does is we integrate into your ecosystem. We connect with your WFM and ACD systems. We do the ACD side in real time, and we send events back to WFM as they’re happening, so essentially in real time as well. What we’re doing is we’re trying to take the information from these systems, like Ted was talking about earlier. There’s so many data points coming from all these different platforms. There’s value in each of them. Being able to measure and manage and do something reliably from any of these data points, it’s a superpower at the end of the day, again, because you don’t need that army of people to do something about it. Focusing on the WFM and the ACD side of things. On the ACD side, we’re reading call queue stats, data about the agents, agent state information, think someone who is on a aux code for x amount of time, how long were you at the call on a particular customer, and the big one is queue statistics. How many calls are in queue? What’s the longest call waiting? How many agents are unavailable? All of these data points become, inputs into the Intrademe rules engine so that we can do something like delivering a training session when there’s a lull in the queue or identifying this voice queue is really quiet. Let me move them over to a chat queue because I know that they’re quiet as well. Or sorry, they’re busy as well. There’s some volume waiting there. Balancing that that that channel between, one queue where you have that kind of capacity sitting and one queue where the capacity is really needed, taking those automations in real time, reducing a lot of that lag time of communicating from one to another. There’s a lot of power there. On the WFM side, we’re reading and writing schedules as well as information about agents, hierarchy data, staff group information, work, anniversaries, birthdays, even all of those things become inputs into the rules engine. So as an example, if you wanted to do a recognition for somebody’s birthday, give them an extra anniversary break, it’s really easy to do in the platform, and you can build the rule where I’m measuring the queue health to make sure I’m not gonna deliver something like that unless there’s really availability. So you can kinda balance that engagement benefit with the needs of the business to make sure that you’re not sort of, harming the queue or harming your service level. And the big benefit to your workforce folks is it’s not landing in a list that you have to go through and manually update segments. Anyone who’s ever gone and done kinda a to b data entry type stuff, couple hours in and your eyes start going cross eyed. So taking that workload off of those folks, letting them use their brains to actually do some forecasting, deal with collision events, talk with the operations teams. Instead of firefighting, we can prevent some fires from tomorrow. It’s a really big game changer for hiring into the workforce team because we’re getting rid of kind of the same idea that that chat bots and automations are supposed to on the call center. Get rid of those simple works that doesn’t really need to be handled by a human. Take anything that’s complicated and get the information to the right person and let them take it over the over the finish line to actually fix whatever the problem is. From a technology perspective, we we connect with anything from an ACD and WFM side. We really go where our customers go. We’re on prem as well on premise as well as in the cloud depending on kinda which revision and which version people are on. And we also have some other products, which I’ll talk on touch on just really quickly at the end. The rules engine itself is really, really simple. The logic is is Boolean logic, so it’s if, then, and then impact. As an example, this rule that you see here is measuring queue health. The frequency is how often we’re checking it. Maybe we wanna check the queue every five minutes to see if there’s availability. Those blue boxes are the conditions. What am I measuring to deliver something like a training session? When am I happy or sad based on the queue measures so that I can do something like a training session or a coaching session or channel balancing? Maybe it’s fifteen minutes to go read your emails or do a a performance evaluation or something about career development. At the end of the day, we like to think of ourselves as a time machine. We’re really harvesting the time out of the queue. So instead of relying on that army of people or prescheduling it so much that you have to have everything right, otherwise, the plan kinda falls apart, we fit in between those two worlds to simplify the day of your real time folks, Again, letting them focus on kind of preventing fires versus dealing with everything that that’s burning right now. And then as well within Intradiem, you can always limit down to the impact side of things if you wanna restrict it to a particular staff group or some of those conditions I mentioned earlier. Tenure, if you’ve been with the company for six months, maybe you get different training rules or different, outlier management rules, whatever the case is. We have a lot of flexibility on building the logic to make sure that we’re matching your environment. And whatever automations we’re bringing in, they’re really reliably exactly what you want to happen in that moment. And, again, to Ted’s point earlier about how structured WFM data is, we’re really in a a a a good place to be able to make that happen considering everything that we put into the platform. From an output perspective, we can output to a desktop alert, so that’s where you get your training prompts or kind of a a an alert if someone is stuck in an aux code or on a a long call or something like that if we wanna send out some support for them. If we send a prompt over to an agent, we can send them a start button to launch something like a training session or send them a yes no question to ask if they’re okay and they need some help. As an example, if you’re stuck in ACW and I send you a prompt that says, I see that you’ve been in here for five minutes. Do you need some support? And they say yes. We can route that to a Slack or a Teams channel or to an email box where someone’s managing it. Essentially, what we wanna do is if we’re ever reaching out to someone, the dispatch guidelines need to be extremely clear or have some sort of path to get them to a solution. If they’re stuck and they need some support, get them that support. If they don’t know what they’re supposed to be doing, they’re in the wrong aux code. Let’s tell them which aux aux code to go to. Let’s deliver the training they should be doing. Whatever it is, we wanna remove friction from everybody’s day, both on the workforce side and at the end users or the agent side of the day too. Alrighty. So a day in the life. From a logic perspective, again, what we’re trying to do is remove the need for people to be managing a lot of these problems. I’ve touched on quite a few of these, and and Ted’s kinda talked about a couple of them as well. Ted, do you have any kind of high level examples you wanna share before I jump into the the the walk throughs? For me, the the the bread and butter was was training. I mean, we I don’t know who has schedulers who say, hey. Can I do more work trying to pretrain or preschedule training itself? And and at the end, that was our our kind of our cash cow, the one that we we cashed in all the time. So I would start there, and I’ll give you a little color after you explain on that one. To open that up, I’d actually love to ask the audience, from a training perspective, how often are you guys rescheduling it? We hear anywhere from fifteen to twenty five to fifty percent of the time. We always used to call it training whack a mole in my world where I schedule training for everyone, and then for whatever reason, huge portion of that population just doesn’t get it done, ask for an extension, has to do it again, didn’t see the segment, whatever the situation is. How much of that is a problem still in in everyone’s world just kind of percentage wise? Any any ballpark would be nice. Cancel because over the shady percent. My heart is pretty good. Yeah. I I have, fell over too. And twenty five percent? Yeah. I mean, we are always having to cancel because of of service level challenges. At the end of the day, it’s like you’ve got performance guarantees and, hey, it’s the twentieth of the month, and right now we’re hitting a hitting a ditch. And so, basically, all that work that we went to line up all those schedules just gets backed off the the, the the the schedule itself, knocked off that schedule because service level was king. So Yeah. Yeah. And I like your point, Lisa. Basically, there’s a policy that says if it’s stuck, they have to move it over. The other side of it is now you have an army of people dealing with those inputs. I have all of these segments that I need to change. So just on the workforce side, we’re adding workload there too. It’s It’s a much better world to automate it away where if we have reliable measures, I can see in the ACD exactly what people are doing. I can see in the WFM what they were supposed to be doing. And I have a list of all the training that needs to be completed or coaching sessions, things like that. All of that data lets us orchestrate when this should happen in those peaks and valleys that naturally happen in any big contact center. And then the due dates is a good call out as well. So so, from a logic perspective, I’ll kinda jump into it from a example. For off phone engagement, which is what we call our training delivery tool, what we’re doing is we’re measuring queue health like I talked about earlier. We’re looking at the ACD and measuring longest call waiting, calls in queue. We’re trying to see whether or not I’m happy or sad right now. Am I hitting my budget targets? Am I meeting my service level? Am I doing right for the business? While balancing that, I have work that does need to get done. I have training that needs to be delivered. I have coaching sessions that absolutely have to happen because people at the end of the day, if they’re feeling unsupported or if they’re having challenges kinda getting work done, all that’s gonna do is bleed into your handle time or your CSAT scores and at the end of the day, cause a bigger problem than than it would have just dealing with it in the moment. Being able to measure the queue and balance your budget versus the efficiency that you’re trying to get, It’s a superpower that you don’t need an army of people to manage. We look at the agent states and the the queue statistics. We check everything on an agent schedule to make sure that we’re not adding friction to their day. As an example, if I’m gonna give you a twenty minute training session, but you have a break coming up in fifteen, I’m not gonna offer you the training. Your break will come first. We can adjust your break and all that kind of fun stuff. But after the fact, if you’re unavailable in the future, the logic would be that I’m in a queue. I’m wrapping up my call. In this case, I have availability in the queue, and I also have training that’s assigned. So I don’t have any kind of collision. There’s no break coming up. There’s no team meeting that I need to be be present for. I’m gonna get the pop up on my screen that says start to offer me a piece of training. So it’d be the same thing from a coaching perspective, which I will show you guys as well. But, essentially, we’re saying, hey. Your queue is quiet right now. I see that you don’t have anything assigned. Let’s pop up something to give them a YouTube video or a LinkedIn learning training or link directly into your LMS system. We can also do integration to kinda automatically load tasks and take quite a bit of the workload away from administering all of that. But from a logic perspective, once we finish the training side, we’ve offered them a training session. As soon as we offer that training session, we update your IDPs, which kinda ties back into the level three requirements that Ted was talking about earlier. If I have an IDP and I don’t have attendance automated, I don’t have training delivery throughout the day automated, Any of those changes become variants that bleeds into your plan, which means if I’m trying to make decisions off of an IDP that’s a little bit old, I’m making decisions off of day old donuts. And at the end of the day, I don’t know if it’s the right one or not. I think it is. It probably is close to. But instead of probably, we can make sure that it’s absolutely what’s happening. So while we measure the queue, I know right now that it’s quiet. I know that you don’t have training or another meeting or segment that’s gonna cause collision. All of those things I can check-in the moment automatically. Instead of having a poor real time analyst go through and do all these checks manually in three different platforms, we use the integrations to do it all right in the moment before the pop up would even happen on the agent’s desk. Once they finish the training, we actually will truncate it down to whatever they actually completed it for. So if it was a twenty minute training, they took seventeen minutes. We’re gonna truncate it down to the seventeen or ten minutes here in this case. But, essentially, we offer them the training. Don’t say how long it is, and we kinda let the amount natural amount of time that they take to complete run through the system, and then we’ll update your WFM side of things. There’s reporting all the levels down. So anything that we offer over to an employee, we’ll have reporting on what was offered. There’s a lot of really great assignment type tracking, which we can talk more if anyone wants to kinda follow-up. Love to go into more detail on it, but there’s a, a big benefit to be able to corral the environment or all the executives, everyone from a leadership perspective around what are we trying to do here? We’re delivering training. We’re trying to make sure that all of these different activities are getting done in a reliable way. So at what point do we cancel training? Having Intradiem, being able to have that conversation before everything’s on fire, before kind of emotions are running hot and executives start trying to make decisions in the moment, it’s really, really powerful to have everyone running on the same page, marching towards the same beat of the same drum so that we can do all of these kind of things in the moment. We so my testimonial on this is prior to Intradiem at my previous environment, we’d reschedule everything, reschedule the training, the certifications, the relicensing, you know, all of that. After one full year of Intradiem, I mean, we looked back, and we totaled everything up. We ended up delivering three times the amount of training and support to agents with no change in in shrinkage just by harvesting real time availability, three x the amount of training and support that we deliver. And, you know, that for us was just mind boggling that we could support agents. There’s downstream impacts to that too. On top of that, you know, agents who are supported and trained, guess what? They don’t attrit nearly as much. So this for us was just such a place to to start and and, you know, the data doesn’t lie. We don’t change any shrinkage. We just start delivering training every day, every interval based on the conditions themselves. Thank you, Ted. Just to close on this topic from a an off phone perspective, it isn’t just training. I’m gonna walk through coaching in just a second, but coaching is a great example. One of my favorites is, like, leadership videos from the c suite. The call center tends to be the first place that gets dropped for something like a two minute engagement reel or a marketing buzz, something kind of fun and simple. You expect people to do it in their downtime with during ACW or while they’re on break or in available time, things like that. Being able to deliver something engaging to the entire company with no touch to your service levels, not affecting your queue at all, and your poor workforce team is not updating five thousand segments for a two minute update or something like that. It’s really, really powerful to get everyone kind of running on the the same beat to the same drum. And and, Kevin, I do see your question. I don’t have any data offhand, but one of my favorite examples, we had an in person event where a leader shared that an agent had moved from a competitor over to them, and the competitor had Intradiem kinda automating brake adjustments. And their question was basically, why are we audit why are we doing these things ourselves? Like, kind of why don’t we have this automated? I like to think of it like moving from switching calls manually with with kinda cables to having the ACD do that work for you. We’re trying to do the same thing on the workforce side so people don’t have to focus on it. And you’re absolutely right. There is a huge engagement win, but we are are very conservative on that. We kinda focus on the dollars and the efficiency side of things. The the I I add data, I won’t share direct numbers, but I will say to some degree, it depends because we worked across different segments of business. And different segments of business have different leaders and have different policies, processes, call types. I I was able to draw definitive connections, though, to which groups leveraged Intradiem heaviest, what their what their attrition looked like beforehand, and what it looked like after. And the most dramatic case that I studied in the past, it cut it in half. Now I can’t promise that, but at the end of the day, you do it is something you wanna track as you go along, and you’ll see a correlation too. It’s just a logical correlation to we’re investing in our people. We’re delivering training. We’re delivering coaching, which we probably won’t have time, but it’s the same concept. We’re delivering coaching in real time. Agents who are supported, who are coached to get these surprise breaks, they’re happier. They stick around. So it’s something I when I talk to people, I encourage them to track because it’s a side benefit that can add up to huge dollars as well, especially when you first introduce it. Once it’s there, you’ll have a new baseline for attrition. You take it away. You’ll have agents who will be like, well, Intradiem isn’t here anymore. I’m going to another organization, just like Josh said, that actually uses this tool because agents love it. Awesome. Thank you, Ted. I do think I have a few minutes to kinda quickly go through coaching. From a logic perspective, it’s the same measure on the queue to say, hey. The queue is idle. There’s something to do here. Let’s harvest that time. Let’s do something productive with it. The big shift is on the coaches side of the equation. That coaching console that you see, that would be a tool that your coaches, your team managers, team leads, whoever it is that’s doing your agent level coaching. They would go in there and manage their teams essentially. The check mark means that I’ve done a prep work. I’m ready to have a conversation with them. Maybe it’s a five minute connection. Maybe I just wanna say hi. Whatever it is, I’m ready to talk to that person now. The time is how long you want that segment to be for. That’s what we’ll offer to the agent. We’ll pop it up on their screen, let them know how long it is, and we’ll update the schedule if it’s less than that time. So you’re trying to kind of manage efficiency within the process as well. So logistically, you can have your kind of your prescheduled coaching. You look at your time throughout the the week where Tuesdays between two and five is a really good time to do coaching. Instead of prescheduling it, we challenge everyone to kind of paradigm shift into the new world where we can automatically deliver that where it’s the best time versus where it was supposed to be the best time based on the plan based on the IDP. So I’ve gone through my logic. I’ve built the queue measures. I’ve seen that the queue is green. I’m the coach. I put myself into available to coach, and I have people who I’ve listed and I’m ready to coach to. On the other side, the available to coach means that Intradium now can see them as a resource, and I’m gonna pair them with a coaching an offer for coaching. So the agent isn’t available. They see that they don’t have a break coming up, all that collision checks, all those things would still happen as well. We don’t wanna add friction to anybody’s day. From there, the pop up would happen on the agent side of things. So sorry. So from the the manager side, I’m putting myself into available to coach. The agent’s gonna wrap up their calls just as before, so they’re hitting the available queue just like they did in the last example for training. In this case, though, it’s gonna pop up and inform them it’s time for a coaching session. The big difference here is there’s no vehicle. There’s no content, nothing like that. We are just finding that time. So this is just a prompt for them to go have a conversation with, their coach or or go into their their journal, whatever it is the coach wants to use this time for. And then once they finish it, they’ll get an offer on the the manager side of things so that they know that they’ve they have a coaching session ready. We automatically pull them out of available to coach so you don’t have kind of collision or any confusion there. Their coaching session is conducted, and after the fact, we’ll go in and we’ll update the schedule. So we’ll put it in for however many minutes that coaching session was done for, and we can truncate it if necessary. Really, the idea here is kinda moving everything into one channel, balancing your noise of communicating to people during the day. If I’m gonna send someone a team huddle to read, I don’t want them to do it while they’re on a call with a customer. I wanna know that they’re unavailable. I wanna know that they don’t have another event coming up so I’m not confusing their day. And I wanna give them that time to measure from a manager’s perspective. If I know that you got the training because I see it in your sched segment, I know that you’ve gotten the coaching sessions because I can see that in your schedule. I know that you got ten minutes in a week to go read emails. Instead of going to sit down and ask them, did you how what you did you see the content or did you have time to do it? The question is, what did you think about the content? It really focuses the conversation to the meat and potatoes of what you’re trying to accomplish, be it training or enablement or communication, and it kinda takes away some of that ambiguity where you have to dig through data. We don’t want your team leads and your managers to be analysts. We want them to focus on the people side of things. That’s what they’re good at, and that’s what they’re there for. And that’s what all of these kinda automations are designed to do is get them away from worrying about the fire that’s happening and get them to preventing fires for tomorrow. So Looking at time. Yeah. They’re looking at time. I think that’s a good time to close here. Any q and a in the last couple of minutes, folks? Really appreciate everybody’s time. Wanna drop down? I’ll I’ll just say a few more words on the last slide with the maturity model Mhmm. If you’ve got that available. And we can share more use cases with everyone. You know, another huge favor of mine was adherence. We automatically can adjust break some lunches as calls run-in. The use cases are endless. But this is a paradigm shift. It’s getting out of this mindset of I need to do all this stuff up ahead of time. I need to preplan coaching, preplan training. Honestly, automation, at the end of the day, as people are saying, how are you adopting AI and workforce management? Start here. Just start automating things. You know, you’re you’re not necessarily leveraging machine learning to write these rules, although we do have a a side product as well that uses that to predict attrition. At the end of the day, that’s over on the right side of it. But it’s a great spot to get into to really shift how your workforce management organization can use real time data to deliver so much more at the end of the day. We believe this this is, like, critical as AI continues to come down. You’ll be in such a much better spot if you’re able to get away from all those manual activities that we still are stuck with today in WFM. I don’t know, Josh, if there’s any questions you wanna try to tackle here in the last couple of minutes. Yeah. We have a few minutes. I can tackle some of them. I I answered one of them on the multiscale. We essentially would kinda manage towards the the lowest common denominator. So if you have a specialty queue or something like that, we do have to consider making sure that we don’t take people out of that queue and cause a routing problem. For John, for your question from a training perspective, we do track which trainings are completed. In general, we use a custom segment on the the WFM side of things that’ll write back or match back to an Intradiem code so you can always track what work we’ve done versus if you did it somewhere else in the platform. And then we also do track completion within Intradiem. That pop up that offered an agent training when they press complete at the end, we have a record of that as well. It depends on if you wanna use your LMS as the source of record or if you wanna use our kind of tracking on delivering the training. And we also can kinda sync those things up depending on how complex your environment is to make sure that you’re working off of one source of truth. And, we will follow-up, Chad. Thank you very much. Any other questions? Yeah. I think you captured it, but we’ll just do a last call. Any other questions? Alright, Vicky. Thank you. I’ll hold everybody. Yep. Thank you so much. It has been a great morning. Lots of great information. I really was interested to see that there weren’t many people past the two the level twos on the maturity curve, but, hopefully, everybody is is moving that way. Right, we got a couple other things coming up, guys. If you can, Yeah. So the the flex offerings, I’m not I’m not following, Brian, if you can elaborate on that. For HR tool integrations, we do integrate with some LMS systems. In general, all we’re doing is building in tasks, or kinda loading in task side of things. From an HR tool, it’s usually your WFM system that’ll connect to payroll that connects to HR. So we’ll get the information that we need, hierarchy data, things like that, from your HR tool in some cases, but generally, it’s just pull it right out of your WFM side of things. Alright. Well, thanks, everybody. It’s a pleasure to share this with you. Feel free. I’m kind of all over the place on the Internet. WFM Labs, ping me. If you ever wanna chat about this from a end user perspective, I’m happy to. And, of course, Josh, you’ll offer the same. Likewise. Sorry. I was kinda hoping that may that maybe you were gonna get a follow-up to, to what Brian was asking. So that’s why I was I was it was a little silent there. But I I I think he might be thinking of kind of flex schedules or something like that, like Uber style style scheduling. And the answer is we look at the segments and the IDP coming from the WFM system. So if you plan in a, let’s call it new version, a sporadic way where you’re kind of doing flex scheduling, we would get the same inputs, what we would have to do is consider those thresholds a little more carefully to make sure that we’re not getting ahead of our skis, so to speak, and and taking too many people off where you do have a big lull in your staffing. We can take in net staffing from your WFM system to do quite a bit of that. Kinda ties right back in and puts a very nice bow in the whole presentation. The data in is the data out. So we need to make sure it’s really good quality on the way in, and we can make really good quality decisions on the way out. Awesome. Alright. Okay. And I will be sending out the recording so that you can share it with anybody or, I’ve even had an email from somebody already. He said he had to drop. He wanted to make sure he got a copy of the recording. So we will get that out to you in the next couple days, so be on the lookout for that. And thanks again to Ted and to Josh for all the great information. Thanks so much, everyone. Thanks, everyone. Alright. Thanks, everybody. Have a great day.