Harnessing Intraday Performance Analysis for Real-Time Workforce Optimization

Hi, everyone. This is Vicky Harrell, executive director of QATC and SWPP, and I’m so excited to be with you today for our webinar with IntraDEM. We have Ted Lango and Jim Simmons from QLESS, and they’re gonna be talking about how to harness your intraday performance analysis for real time workforce optimization. And we’re looking forward to hearing from them today. They’re gonna be interacting with you through the chat today. So make sure you have your chat pulled up, and make sure you are set to send to everyone. Sometimes it’ll say send to just the panelists, but make sure you click that little drop down and make sure it says send to everyone. And, let’s go ahead and put in there where are you listening from and maybe what your weather’s like this week. We’ve got cold weather right now in Nashville, but it’s gonna warm up and rain over the weekend. Ted, how about how’s it doing in Atlanta this week? Actually, it’s Fort Lauderdale, Vicky. And it’s it’s just Well, then you’re in good weather. Yeah. And it’s beautiful down here. Well, it is sunny here but cold today. So I’m cold in South Mississippi. New York in cold. Saint Mary’s, Georgia, mild weather. Cold in Atlanta. So windy in New York. Lots of rain in Hartford, but last night, but sunny today. Very cold in Omaha. Sunny, but cool now cool in Calgary, Alberta to me is gonna be probably a little, colder than I expect. Day sunny in forty one in Long Island. Alright. Sunny and mild in Northern Cal Colorado. Alright. It looks like everybody has found the chat. And I’m gonna go ahead and turn it over to Ted and Jim to get started and, look forward to a great session, guys. Thanks for being with us. Thanks, Vicky. Welcome everybody to harnessing intraday performance analysis. Today, we’re gonna focus on, challenges associated with what I call fragile staffing plans and how we’re all sitting on mountains of data that can be used to support new ways of staffing, leveraging some new tools, that can really unleash the power of the data that we’re all sitting on. Why? We think that, this data can help you make a case for staffing differently than you do today. Let me just make sure I’m sharing. I think I am. So I’m Ted Lango. I’ve spent, over two decades helping contact centers operate more effectively. For the past two and a half years, I’ve spent a lot of time looking at new approaches to solving some nagging old problems that many of us in workforce management still face, and we wanna share some of those with you. Jim, you wanna introduce yourself? Sure. Jim Simmons, cofounder of QLess and, formerly with Synchrony Financial. I think I figured the other day, Ted, between the two of us, we’re, over fifty years in the industry. So, fortunately, we’ve got a little bit of knowledge, and I’m very excited about the topic. Having spent that amount of time, in the industry, you know, a few plus decades, it’s it’s a very unique time, and I feel very fortunate to see some of the changes that we have access to and, you know, how passionate I am about this topic. So very excited for today. Thanks, Jim. You’re making me feel a little bit old, but that’s okay. So let’s let’s get started. Actually, I just wanna ask everybody a quick poll question. I’m not gonna display this question. If you can just type in the approximate size of the contact center that you support. Is it less than a hundred agents? Is it something between a hundred and five hundred? Is it five hundred to a thousand? Is it more than a thousand? This this topic that we’re talking about is a little different for different size organizations, and we just wanna keep in mind that some of the stuff we present here may strike, you know, strike our strong cords for some of you. Some of you may be like, well, I’m, you know, I’m only supporting, you know, twenty agents. I don’t see too many of those. We see a lot of large organizations coming up here, and this is really an important topic if you are running a contact center that starts to get up into the hundreds of of agents. The first thing that we wanna cover is really the challenge. You know, and then one other thing before I get into number one is if you’ve got a question, today throughout this, if you can also type that in the chat as we go, we’re not gonna take questions as we’re, presenting because we’ve got a lot of material. But whatever time we have left at the end, we’ll try to answer as many questions as time will will permit us. So let’s jump into the the topic itself. I talk about fragile staffing models, but first, I wanna just align on some terminology. I call it net zero, staffing, but maybe other people call it something else. So I I just wanna first say to illustrate what I mean when I say net zero staffing, we’ve got a little balance here. And if we put demand on the left side, our volume and our handle time at an interval level and our supply on the right side, how many agents did we schedule to answer those calls? We’ve got three different scenarios that we have every interval as we go along. We’re either over demand. We have more calls, more handle time than supply to handle it. I call that a negative net staff number. We can be oversupplied. We can have less demand than the number of agents that we have. I call that a positive net staff. And then when I say net zero, what I’m referring to is an interval when we’re at equilibrium, when we’re able to match the demand with the supply. That’s what I’m calling net zero. If we look at it as we, calculate out the day, by interval, You know, we all have either software or some type of mechanism to say. It’s fifty calls at nine o’clock, five hundred and seventy seconds. We push that through our tool set. We could be using Erlang c. We could be using just our software to calculate, and we end up with a number that says, hey. We need twenty head count, you know, for that interval. At the end of the day, the twenty head count is only half of the equation. We have then have to go and schedule the heads. In this particular case, you know, I may not be able to schedule all twenty that I think I need, and I end up with a net staff that is negative. That could continue on into the nine thirty hour where I have a few more calls. Twenty two is now what I need. Twenty is what I’ve got. I’ve got a negative two. Or, and then I could go on into the ten o’clock interval where I could end up positive at some point, more, staff than what I required. And then at ten thirty, I could finally hit the nail on the head where I have twenty four that I forecast I require, twenty four scheduled. That’s what we’re referring to when we say net zero itself. As we forecast and schedule all these intervals, we see different types of risk. If we have a negative net staff, that normally translates into a service level risk. If we have a positive net staff, maybe we’re not gonna have problems with service level. But at the end of the day, we may have a boss that says, hey. We’re overstaffed. That’s an expense risk potentially. And then we come to the ten thirty interval, which is where we land on zero. You know? Is that just the right amount of staff to have? Are we right sized itself? And while staffing to net zero seems like the right thing to do at the end of the day, we’re perfectly balancing our our heads on the phone against the demand. At the end, this approach leaves us really with a mixed set of results as variance enters the picture itself. Let’s examine this in a graphical way. When we build out our net staff and if we’re staffing to net zero, we basically have a line that we illustrate for the day where we’ve got our forecast required staff, and then we ultimately schedule against that. For simplicity, I’m just gonna assume that I can schedule a hundred and twenty people, and I need a hundred and twenty people at eleven thirty. But then at the after the day ends, we always end up with variance to the the story. On the demand side, we could have demand variance and say, hey. We needed actually a hundred and twenty three people. And regardless of how good our forecasting is, you can’t get away from minimal interval variance. There’s always gonna be some degree of demand variance, but demand variance is only half of it. We’ve got supply variance. You know, when we have people that we say we need a hundred and twenty scheduled, some may not be adhering to schedule. Some may have some more may have called out sick than we were planning for. So in this case, we could have scheduled a hundred and twenty, needed a hundred and twenty three because of the demand variance, and then ultimately, we had a supply shortage on that interval where we only had a hundred and seventeen. It’s each of these variances that really introduces two types of risk, service level risk and then expense risk. Service level risk when supply is less than demand and expense risk when, our supply is over our demand. And it’s this area that we really wanna look at here further and show you how to evaluate. How big of a problem is this for your organization itself? Are there lots of issues where supply and demand are misaligned and it’s costing you minimal. Again, no single answer, but that’s part of what we’re, gonna show next, part two. Jim, your thoughts on on technology and how you’ve seen the industry, address this over over the years? Yeah. For sure. I you know, one of the things that strikes me about our industry is, to some extent, how little has changed over time. So I mentioned when we kicked off that I feel very fortunate to be working in this period because so much of what we’ve seen in the past as a hindrance to success and workforce management is now being addressed through things like AI and machine learning and technology in general when implemented correctly. You know, if you think about it, I can speak from firsthand knowledge. Again, having been in the industry for, you know, twenty six years now, Not a whole lot has changed in probably twenty four of those years. We use the same KPIs. We chase the perfect forecast. Right? And then we scramble the day of. And the reality is it’s we always kinda look at the past to predict the future, and then our reaction time is really what dictates how successful we are. And I would I don’t have any data to support this, but I would venture a guess as an industry, we’re very good at forecasting. I’ve talked to people who have forecasts that are well within plus or minus five percent, especially if you’re in a larger organization. That’s huge. When you go out and just look at forecasting in general, we’re doing a great job. But it’s that reaction time that really makes the difference. How do you react when reality occurs? And that’s where I see the industry changing. And I think as a group of professionals, and one of the things I really love about SWPP is that it brings a group of professionals like this together. It’s incumbent upon us to understand the ramifications of this technology and then really speak to that with our executive and senior leadership teams. So it’s an age old problem that I think we have a very new perspective on. Great. Thanks, Jim. So how big is the problem? And there is really no single answer. And and a few of you are asking about shrinkage and and adherence, and, one of the things as well is, you know, are we averaging across intervals or both? And all of that comes into play here. That’s the interesting thing about these questions, and we’ll get to some of this as we go forward here. But let’s move on to the second section, which is really getting your arms around, this answer. There is no single answer for, like, how big the issue is for you. As a matter of fact, there the the it’s a it’s a ever changing answer based on the health of your queues themselves. You could have queues that are fairly healthy this month that all of a sudden are coming into a period of time where there’ll be a lot more variability. And this is where the PDP analysis can really reveal the opportunity for your individual organization itself. Before I I go and start to share some of that again, let’s align just on some terminology here. I use the term PDP or prior day performance report. I think as an industry, most of us are pretty familiar with the term IDP, intraday performance report. We’ve got WFM software that tells us how we’re doing throughout the day, what the staffing looks like, what the forecast looks like. When I use the term PDP, I’m speaking to prior day performance for the full day itself. Essentially, a detailed analysis of that prior day’s workforce and operational metrics that are comparing our forecasted values against the actual values and then our staffing situation and then ultimately the end results, the service level, the abandoned, the ASA. It’s all those types of things. At the end of the day, PDPs will look different from one organization to another depending on, you know, the software that you’re using itself. Most of them, you know, by nature are looking at interval level data, and then whether or not you have all these variables or not depends on how you’re accessing it. Forecast calls against the actual calls and then the variance itself. Forecast handle time against actual handle time and the variance. You may or may not be looking at your interval level workload deviation, taking volume and handle time and saying what are the those two things combined. And then the next section really looks at our staffing. All of us basically forecast at an interval level, how many people we think we’re going to need. And then at the end of the day, we’ve got a different number. How many people did we really need based on how volume and handle time came into play? Then we scheduled a certain number of people based on our forecast required. And at the end of the day, we actually had a different number or maybe it’s the same. Sometimes it lines up. And then we get to two really important numbers, the forecasted net staff. How many people did we think we need versus how many people we scheduled? This is before the day unfolds, and here, we look like we’re fairly healthy. We were set up, scheduled thirty people. We thought we’d need twenty five. But then the second number, the actual net staff tells us the end story. How many did we actually require based on, how volume and handle time came in and how many people did we actually have? In this particular case, I go from what I think is a fairly healthy interval, a plus five at the beginning of the day to a negative five. And then that shows in my end result of service level being thirty three percent and abandoned rate being, you know, five point eight percent. So these numbers themselves, at the end of the day can tell a a lot about what’s going on, you know, in your individual queues, at a very detailed level. The problem is there’s a lot of data here, and that’s what we’re gonna get to next. But before we do, I wanna interject a poll question. I’d love to know from you guys, like, how many of folks are currently using those prior day performance reports? Do you have those readily available to them, and do you analyze them daily? Do you have them, but the analysis is only done, you know, when things are on fire and you really wanna dig into why? Do you not have them? But, you know, maybe you’re planning to implement some type of report like this, or do you not have them and don’t really have any plans to look at at things, like this? And if you’ve got them and you’re able to analyze them, that’s fantastic, because they’re full of of rich datasets themselves. Second poll question, as as you guys are thinking about that is for those of you who just said that you do use them, how easy is it for your team to perform the analysis on it? Is it super easy where you can basically come out and say, this is why, you know, I ended up the way I did because I’m already doing something with this PDP analysis? Is it somewhat easy, or is it difficult? You know, is analysis itself time consuming? Is it unclear? You know? Or maybe you’ve got the the data in a way where it’s like, hey. You know, we don’t analyze PDP reports because at the end of the day, it’s just too cumbersome itself. And I’m getting a mix of answers, which is interesting. Myself, I I think there’s gonna be, again, a mix of answers based on, one, the size of the queue, and and call center that you’re working at, Number two, the technology that you’re working with. And number three, the folks that you have that are able to actually analyze data. You know, if I just had one queue itself, wouldn’t be so difficult for me to write some Excel scripts to go through and do some pretty deep analysis here. But many of us are working with not just one queue, but dozens of queues or hundreds of queues. The previous place that I I worked with workforce management, we had fifty five queue segments. Each of these PDP reports has about five hundred different data elements on it, and that meant for a single month, if I wanted to analyze what was going on, I’m looking at over a half million data points to analyze. This is, something I could do in Excel, but it’d be fairly labor intensive. It’d be prone to errors. And, honestly, it would really limit me to the equations that I could write within Excel to do that type of analysis itself. And that’s where, today, I feel like there’s just a whole new way of of approaching analysis like this from a data standpoint. For the past two years, you know, since the release of of chat GPT and large language models, I’ve tested out at least a hundred and fifty, probably two hundred different AI applications. And I’ve sort of honed in on a suite of tools that I use myself, to keep going back to. It depends on what I’m doing. If I’m doing research, there’s a certain set of tools. If I am trying to build something new or produce videos, there’s a different set of tools. But this area in data analysis, I found particularly interesting, especially with platforms like Deepnote, which is what I’m gonna show you today. I’m not here endorsing or selling Deepnote. It’s just a platform that I’ve tried out that I find is is pretty friendly around this. And what this does is it allows you to take today’s newer technologies with LLMs and really harness the power of some older tools that have been around for years and years. At the end of the day, data notebooks are really as simple as creating a notebook the same way we create a tab in Excel, connecting or importing our data, and then really prompting, the data itself. Before LLMs existed and before ChatGPT has caught fire, these things existed, these data notebooks, but they required skills in using Python programming, data manipulation, and and visualization, really fairly deep domain expertise and knowledge on statistics. You also had to manage the environments themselves. If if anyone’s worked with Python notebooks prior to ChatGPT and some of these new platforms, you know, you had to, manage the environment. You had to debug code. You had to be able to write clear documentation, and it was really quite a steep learning curve. But with these tools that we have access today, we essentially can, prompt the the tools themselves, generate Python code itself, and very quickly analyze datasets that I spoke to before. I can go through and just calculate and visualize in the matter of moments how fragile my staffing plan might be. What are the results of me attempting to staff to net zero? And that same type of line, this is using real data, that same type of line, we can go and visualize with our data after the fact. How many people did I schedule? How many people did I require the red line, that’s there after the fact? And then how many people did I actually have staffed after the fact? And I told the tool itself, let’s let’s help me with some visualization in looking at, you know, in red shade in red, the areas where I’ve got service level risk, shade in green, the areas where I’ve got excel or excel risk, where I’ve got, expense risk. And, again, you could do this in Excel, and go and visualize it. But then if you wanna start to take it to scale, if you wanna analyze every interval across every queue, across months or even a full year, it quickly becomes a problem that’s fairly difficult to manage. Yet leveraging these tools, I can analyze my intervals at scale. A full year I did for this single queue, forty eight head count. I could then build in a business model. At the end of the day, there’s a hard cost to overstaffing, and there’s an opportunity cost to understaffing. Potentially, a financial cost as well to understaffing if you have performance guarantees and you have to pay for those intervals. So this is just a little preview into something that we’re gonna offer at the very end here. But if you’re not familiar with these tools and tapping into them to really start to dig into the mountains of data, that’s a part of what we wanna share with everyone here. Jim, let me just take a breath and and allow you to chime in at this point as well. Sure. So, you know, one of the things that you you put into the title and that you call out and what I really like about the approach is we talk about the fragility of the plan itself. And I would venture a guess that most people on this call have experienced some version of a senior leader calling and saying, what happened yesterday? Right? Why do we suffer? Why were we over? And I could tell you my personal experience and with the clients I’ve worked with and and, again, having been in the industry for many years, we spent spend an inordinate amount of time looking behind us trying to fix something for a future state. And that time looking behind us is money as well. And some organizations that that maybe are a little bit smaller, one or two WFM folks, and they’re trying to manage forward, right, the day of. They’re trying to manage everything real time and go back and look at what happened yesterday and last week and analyze it and then leapfrog over today to make things better for tomorrow, and that’s problematic. In other organizations, the larger ones, there’s whole teams of people just divided up between forecasting and then they the day of the real time management, and then there’s a whole analytics team. So it’s not a small amount of effort that goes to looking at this information. And the reality is, as a business, as an industry, we’re better served to spend less time looking behind us and more time honing our reaction skills, our capabilities, leveraging technology in a thoughtful way that makes sense. Right? So if you’re one of those organizations who spend a lot of time scratching your head, looking back, pointing at a couple intervals to go. I think these intervals are what were problematic yesterday or have been for the past few weeks or months. Going back to my original statement here, if you really talk about the plan itself and its capabilities being fragile or not, right, I think that’s a whole different conversation, which obviously and I, you know, I like the fact that you’ve included, some of the cost factors in here because that’s what it’s all about. Right? At the end of the day, our industry and the professionals on the this webcast right now are tasked with help managing that cost very effectively. So I I think it you know, if we start to shift our thinking a little bit away from that, what happened yesterday, and we reframe that in the context of how fragile is my plan and how effective am I at reacting in real time because we all need to. It changes the conversation, and it not only helps us do our jobs better, it allows us to get better insight to our seniors and executive leaders. Yeah. No. I I agree hundred percent. At the end of the day, this analysis comes and brings us to a point where we’re sort of questioning the way we’ve been doing things for years and years, because we think there’s a better way at the end. So let’s move on to our next session, which next section, which talks about a a different approach to staffing models themselves. But first, I wanna jump into one or two, really, two last poll questions I’m gonna ask the team. The first one is, how does your contact center handle training and coaching sessions for agents today? Do you preschedule the training and coaching sessions in advance? Do you use intraday automation to deliver this this coaching in real time? If you’re answering too, you probably already know the a part of what we’re gonna share here. Or number three, you don’t currently schedule regular training and coaching, sessions at all. You just wing it whenever whenever the time seems right. And I’m seeing a lot of ones, a few twos mixed in here. I’m glad to not see too many or any well, one and three. I see one number three. I think that’d be a pretty bad, environment to be in themselves. So, let’s talk about or let me ask one last question as you guys all are looking at that. The traditional approach, and and Jim touched on some of this, has really been to balance and supply a balanced supply and demand down to net zero. I’m curious as well, and you can just type yes or no. Does your contact center aim to staff to net zero? No wrong answers, here. Some people will have a real reason why they need to staff to net zero. Some people are like, yes. We do staff to net zero. And I’m curious as to if people are saying no, because there’s a number of people who don’t staff to net zero. Anything you wanna share about what you’re doing opposed to staffing to net zero? Do you basically build in overhead, and and how do you build in overhead? If you wanna share any comments about that, it is interesting to to hear some people will say, yep. I’m building in some overhead for x y z reasons. And even though I said there’s no right or wrong answer, our position is that you should stay away from from staffing to net zero. However, you may have a leader that says, hey. I want you to push that thing as close to net zero as you can because it’s wasteful just to leave additional staff on the table itself. So net zero staffing, I see a a number of folks who said no to it. Very good if you’re not. I also be curious, inside conversations here as to how you explain that you’re truly not overstaffing, and there’s ways around that. Our position has been that, you know, you shouldn’t staff to net zero, that you should build overhead into your staffing plan. At WFM Labs earlier this year, we developed an approach that we called Erlang o, which is half the solution that helps you quantify how positive you can bring your net staff line to. It doesn’t replace your your algorithms that you have today even though it’s called Erlang o. It doesn’t mean that Erlang c is no longer valid. At its simplest level, Erlang o takes your core algorithm, whether you’re using Erlang c, Erlang a, or your own calculation, and it incorporates things like shrinkage, alternative channel work, and then a degree of variance and volatility into your staffing plan itself. At the end of the day, that’s why we call it Erlang o is because we’re considering components of overhead to put into our net line itself. I see a lot of no’s, and I’d I’d I’d love to have conversations outside of this webinar as to how you go about calculating how much overhead to put in. If you’re saying, yes, I do staff to net zero. Even if you are staffing to net zero, you are still probably putting some overhead in. And the the one that everyone does or should be doing is you can calculate at eleven thirty, four hundred and sixty calls. I need at four hundred seconds, I need a hundred and ten people, but we all or all should be at least looking at day of absenteeism. Absenteeism itself happens every single day. So instead of putting in a hundred and ten people, we don’t schedule a hundred and ten people. We take that absenteeism and build it in as overhead into our staff line. So I’m gonna go and I’m going to attempt to schedule a hundred and twenty people. At the end of the day, Erlang o extends that model. It essentially says, let’s look at the rest of our overhead and put some of that in our staffing model as well. If we go back and we visualize again using this chart where I was showing demand and supply variance, At the end of the day, we have these risk areas. How big these risk areas can be determined by doing that type of PDP analysis. Once you understand how much risk you have, you can then go back and start to look at your base Erlang model or your base algorithm that calculates required, and then you can start to look at intraday shrinkage to continue to bring into your staff line itself. Now this is is different than just, hey. This is how many people I need to budget. At the intraday level, you bring shrinkage that you would normally try to preplan. You also can bring alternative channel work into your staff line. You can bring in variance and volatility factor. And that’s what Erlang o does is it attempts to look at how much overhead you can really bring into your staff line to ultimately rise up above this wave of positives and negatives that happen every day. That’s what we call Erlang o. It’s a new forecast required staff line at an interval level that basically rides above then mitigating those service level risks. So now we try to get variance, volatility, shrinkage, overhead into there. But what about the expense risk? If you are staffing on a positive net staff interval after interval, someone may come along and say, hey. At the end of the day, you’re wasting money. You know, Erlang o isn’t overhead, Ted. You’re just overstaffing. And if I don’t have a way to take the variance out to deal with variance that swings in a different direction, then I potentially could be over staffing. And while a part of your expense risk would be eliminated through natural variance, this is where automation really comes into play. Real time automation can allow you to take those prescheduled activities, training, coaching, communications, and really deliver them in real time based on how your line is going throughout the day. Those are productive activities that can be delivered all day long that can then tune your net line back down to zero. This allows you really to protect your service level to optimize expense and be able to say to the CFO, I’m not wasting money. I’m not just overstaffed. I’m not wasting money. I’m optimizing and to invest far more into your people than you could before. At WFM Labs, we call this, level three of the maturity model itself where we’re introducing real time automation. What we’re going after is we’re trying to attack these notions of having to preschedule all of your activity and then react to it when you’re you’re basically missing service level, having to cancel those coachings and cancel the training and reschedule. Instead, we’re delivering all that activity in real time, whether it’s coaching, training. There’s only roughly five percent of the industry that’s doing this on a full, automation basis today. The other eight ninety five percent is either too small where they’re still doing things in Excel or stuck kind of in the legacy process of trying to preschedule everything. I’ll just take a pause there, Jim, see if you wanna interject anything about this at this point, and then I’ll talk a little bit about automation. Sure. Yeah, I I go back to what I’ve said is that, you know, a lot of this is very standard in terms of what we’ve done, before, especially as it relates to sort of building up. I’ve looked at some of the answers coming in. And what I think is interesting is when we ask the question, do you staff to net zero, A lot of folks answered no. We plan for shrinkage and benefit time and things of that nature. But, ultimately, what we’re still trying to do is we’re adding on going, I know people are gonna be on vacation and holiday, and we’ll have attrition. So you plan up to those, but you still hope to, day of, end at that net zero line. So I would argue, in many circumstances, we’re still trying to plan and land on that. What I don’t think a lot of organizations are actively doing is saying, I’ve accounted for everything, and I’m adding a plus five percent factor or a plus ten percent factor, most likely. There probably are some out there that are actually doing that. What I think happens more often than not based off of my time and talking with folks is in that real time world or near real time, we’re shifting resources around. We’re doing things like bringing in temp workers. We’re adding a lot of, overtime for any given period. We’re shifting folks around. So we’re still using that real time reactive behavior to normally, what I’ve seen, is push things up. Right? So we kinda end up a little bit below the line into how do I shore it up to meet my service levels simply because it gets very scary to say, I’m gonna I’m gonna take into account shrinkage and all of the things that happened, and I’m gonna add ten percent to that number. Now let’s hire to that. That’s never an easy conversation and most likely will end with you hearing no. Yes. Which is why I like when you can talk about that in the context of here’s how fragile our plan is. And the one thing I always say when I talk to WFM, professionals is we’re really good at our job. And I’m not saying that to stroke people’s ego. I’m saying it because it’s true. But one of the problems with that is I think our executive and senior teams don’t realize how hard it is to do what you do day in and day out. So I I think something that is important for professionals in this industry to think about is how do you become more active and proactive? A lot of times, we get hired into WFM roles as BAU. Right? Just come in and do the things, make the schedules, create the forecast. I think, you know, we should be thinking, where am I at on this curve? How do I build a WFM strategy for my organization that meets our goals? And how do I have those conversations at the executive level effectively? So I go back to what I said at the beginning of this this webcast. That’s why I get excited about organizations like SWPP and what Vicky’s done and these kind of conversations that you and I get to have, Ted. Yeah. No. Great points, Jim. And and let me just clarify, as I’m watching a few questions come by, and I’m not gonna try to answer everyone because we’ve got so many people on and and a lot of comments. But I I just wanna take one second to clarify. When I talk about shrinkage, we’ve got kind of two worlds that we operate in. We’ve got our budgeting. We’ve got our, you know, three months, six month outlook, and we’ve got our capacity planning that we build that which says, as Jim says, I need to go out and hire, you know, hundred people. And you’re making assumptions around shrinkage, you know, in those models that says, well, we wanna spend, you know, three percent on training, two percent on coaching, this amount of percent on PTO, this percent on absenteeism. We all add up and come up with a number, like thirty percent shrinkage. You know, in between our call volume, handle time, our thirty percent shrinkage, and our occupancy, we calculate how many people we need to have on the roster. When I’m talking about net zero and interval level performance, the majority of software packages out there allow us to put in the intraday absenteeism factor, and it bakes that into your staff line. But at the end, if you’re prescheduling coaching and prescheduling training, you’re probably staffing to a point where it’s like all that shrinkage stuff is already handled for. And what we’re saying here with Erlang o is rather than try to preschedule training, preschedule coaching, rather than ignoring minimal interval variance and volatility, rather than trying to schedule back office or email activity. All those activities can be baked back into your net staff calculation. How many forecast required do I need at eleven thirty? And building some of that into the overhead is we feel the the right way of doing things. Now if you’re building five percent in, that’s fantastic. Then you’ve got a bit of a buffer. But you also may have a potential challenge if you don’t have a good way to take it out in real time and react to variance in real time. And that’s where you can get caught with intervals where you’re just overstaffed. I mean, at the end of the day, it’s like we all know in workforce management if volumes coming in, you know, low and there’s people sitting there idle, it can be very difficult to respond to those variances quick enough to ensure that you truly are optimizing your net line. So when I’m talking about shrinkage, I’m not saying, hey, you know, we need to account for PTO and VTO or PTO and and, absenteeism and training and coaching in our planning number. It’s when we get right down to the intraday performance and the net line that we’re looking at that I’m speaking to with some of this. And, really, there’s only one way of doing it, and that’s with automation. If you wanna ensure that you’ve got enough overhead, that you’re not getting knocked upside the head every time with service level dings because you’re staffing too close to net zero, and you wanna say to your CFO, I’m optimized. I’m making the most of every person I’ve got. The only way to do that quick enough is with intraday automation itself. That’s where Intradiem comes in. We are a critical component of the contact center. We’re different from your WFM system. We’re different from your ACD. We live between these two environments. Interdiem is the only company that reads real time AC data and writes back and reads your WFM data to allow you to write these real time automation rules themselves. Intra Diem itself integrates with all the standard ACD and WFM platforms, whether it’s cloud or premise, and it really empowers your workforce management team with tools that doesn’t require IT to write if then logic about your queue conditions. If the queue looks like this, then let’s do this. The final component of it is that it also communicates out. Yeah. If you’re gonna basically ask the agent to do something, you need to be able to send an alert. So Intradium has a an alert and a communication mechanism that allow pop ups to come on the desktop. You can write rules that ultimately send messaging to Slack or MS Teams to SMS to email to be able to communicate those actions that we need to take when we have variance throughout the day. The use cases are are endless, for automating your workforce management activities. I’m only gonna touch on two of them here today, the off phone engagement and the coaching time. I feel like these are the two that, are the ones that our customers get just a huge bang for the buck out of we all know, you know, when we go and we try to schedule, training, schedule off phone activity, is there ever a right time? And the answer often is is no. There really isn’t a right time. We do the best that we can to slot those activities in, but we have to cancel them quite often because, you know, we may be getting hit with service level. What Intra Diem does is allows you instead to write a rule. If the longest call is less than ten seconds, if there are less than three calls in queue, now would be a good time to do training. My own personal story, prior to joining Intra Diem was at my previous company, As as much as I tried to optimize my prescheduled training and coaching activities themselves, I always had to sit there and cancel and reschedule some percentage of it. But instead, by writing rules that respond to my real time queue conditions, the first year after I deployed Intradium, I was able to deliver three times the amount of coaching and training than I ever could in prior years to that. And it’s not just coaching and and training activities. Anything you can imagine in terms of off phone engagements. Training was a big one, but if it’s just communications, now would be a good time to catch up on this product update, surveys, huddles, tasks. Jim, things that, strike out for you and or stand out for you in terms of this use case, that you made use of in your past with Intra Diem? Oh, yeah. For sure. I mean, one of the most frustrating activities in our WFM group was prescheduling anything. Everyone knew you would schedule it, and there was a large chance that it was not gonna happen the way it went into the plan. Even if everything went okay with call volume, you’d put in for twenty people and four of them wouldn’t show. Right? So you’d have to go make adjustments. So when we looked at the product itself, it wasn’t just about the automation factor. It was about all of the back end rework that occurred to keep up with that. Right? And that was a big win, and that was something that not for nothing, you know, is a smaller group normally as compared to the contact center itself. But, you know, those are the folks, the WFM team that are in there every day pulling all those levers. So that was a a big win, for us when we implemented. Yeah. I mean, huge huge lift off of the workforce management team itself. But, it’s so for me, it’s a win win win is, like, at the end of the day, I could deliver three times the amount of coaching and training. I couldn’t I didn’t have to sit there and do all that reschedule work instead. You know, my real time team could focus on more value add things. It can. Yep. So I just saw a great comment in here that it’s all still a calculated crapshoot, which I think is a hundred percent true. What I would say is, how do we improve that calculation and our chances? Right? So it’s just like any gamble you take. Right? You want the best information, and you wanna use the best strategy possible. So I I like to comment, and I think that’s how we should approach it is how do I make sure my calculations are spot on. Yeah. And I can hear Rick saying that too. At at the end of the day, the same goes for coaching. And and, honestly, calculated crapshoot is what it is ahead of time, and that’s why I love this approach is, you know, same thing with coaching. If I attempted to set up times where the, you know, supervisor’s available and the agents available, I can’t tell you how often we had to reschedule those. This is a little different use case, because instead of just prompting the agent and saying now would be a good time, this actually takes coaching and puts it in the hands of the supervisor. With this part of our tool and our solution, you can basically prioritize which agents you wanna spend time with. First, you can prioritize or you can, indicate how much time you wanna spend with the agent. Of course, there’s a coach now thing. If you really do need to interrupt and jump in, you can do that. But this allows then the ACD and the workforce management platform to further inform what’s going on. You can set conditions to say, if my queue looks like this, then now would be a good time to coach. Message goes back to now both the supervisor and the agent saying, hey. Now is a great time because the queue conditions are saying it’s a good time as opposed to the calculated crapshoot of, hey. Nine thirty might be the right time. Well, it might not be the right time. At the end of the day, this eliminates that constant rescheduling. Again, helps you deliver just better relationships between the supervisors and the agents. It’s a better a better ending as well then for the customers because at the end, you’re getting just a ton more development time, you know, invested. As I said, we could spend a whole hour just going through use cases with IntraDiem. Those who’ve used Intradium recognize the the power behind it. But one of the areas that that may be tricky for organizations is really quantifying, you know, how big of a problem is it for me. And that’s kind of our offer here today. Jim and I are happy to talk to people. We’ll try to answer a few questions at the end. But if you wanna dig into this a little bit deeper, take a picture, of that link. It will take you to a link where we’re giving you a free offer to do some of this PDP analysis on your contact center itself. If, data notebooks and and Python scare you and you haven’t used it before or you just like to see kind of a new approach to how to really analyze what’s going on at an interval level in terms of the delivery of both service level and your expense objectives, this is an exercise that we think is super valuable. It can really help you put your finger on how big of a problem it is. And if it’s not problem, great. You know? But at the end of the day, we’ve all worked in this industry so long that we just know if if you’re staffing to net zero, you’ve got volatility baked in. If you’re staffing above net zero and you don’t have a good way to get people out of those queues when variance comes in, you’ve got some waste going on. And at the end, without that automation, you know, you’re just not gonna be able to tap into, the savings. The savings are real. We do studies, you know, every year with our customers where we go out and quantify how much money, automation is saving them. And we’re coming up on a billion, nine hundred and thirty million over the past four years for just our top ten customers. This has got hard ROI. And this this offer that we’re giving you here with the PDP analysis can help shed some light on it. Jim, anything else you wanna add before we, try to take a couple of questions here? Just the fact that, you know, the nice thing about this and one of the things that I really liked about it was it it is not a heavy lift. This isn’t, hey. We gotta get under the hood and spend three weeks or a month in there. This is, you know, hours, and maybe a few days. So it’s a very easy lift. And quite honestly, if you’re an organization and you’re not quite sure if you have all the data ready to go, that in and of itself has some value. So, it’s an easy one for folks to take advantage of. Very good. Yeah. And and, so we’re happy to kinda show you under the covers if you’d like to learn about that. It’s a it’s a a great way just to start a business case if you’re struggling with, chasing service level and chasing expense. Far too many organizations, you know, still wake up every day, feel like you’re doing the same thing, and there’s a better way to go at it. So I wanna thank everybody. We’ll try to jump in. I don’t know, Jim, if you grabbed any questions. There was a or if anyone has any questions, they can throw them in in chat here. We’ll try to cover one or two before we gotta wrap up here. There there were quite a few. I think you addressed the broader ones that I was seeing, so I’m open to somebody throwing something in the chat. Yeah. If anybody’s got something they still are are, are curious about, feel free to type it in the chat. We’ll stay on for another six, seven minutes or until we have hey. No. We don’t have any questions. We’re all set. Ted, one of the things that everybody I mean, kept coming up over and over and over again was the the net staffing having to do with shrinkage and, you know, and adherence and those kind of things. And then also the you know, what is what is VTO really, and what do people say VTO is? But Yeah. And and at the end, I I think and then this is we’re happy to have, deeper conversations with people if you wanna dig into that analysis, even if you wanna take fifteen, twenty minutes and go through it in your environment. But as I said before, the the the, the there’s there’s two levels of of looking at shrinkage and probably more levels. You know, are you even tracking shrinkage at a detailed level where it’s being used properly for your planning? Assuming that you are tracking shrinkage, that you understand, you know, your training, your coaching, your PTO, all of that stuff, you know, builds into your budget and how many head count you really need to have on your roster. But when it comes down to interval level, scheduling, how many people am I scheduling at eleven o’clock? The majority of people that I know are building in one component of of shrinkage into their staff line. It’s absenteeism. Now if you’re building in a buffer, great, then you probably are gonna end up set up with a slightly positive net staff. Shrinkage comes in a lot of flavors. VTO itself voluntary time off for me has always been a lever to say, I’m overstaffed, and we can automate this to say, hey. My forward looking, net staff line is a plus ten. You know, I’ve got ten more people than I theoretically need, so let’s let five of those ten VTO, most powerful lever in terms of optimizing. But if you can’t automate that, then you’re probably not taking full advantage of that. And while VTO is an actual use case or interdiem automating the offering of VTO. The bigger challenge that most people face right off the bat is taking their training shrinkage, taking their coaching shrinkage. Your budget may have had two percent for, for coaching, three percent for training. You built that into your budget. You hired for those number of people. You’ve got a hundred people in this queue and you supposedly have enough to deliver two or three percent shrinkage. Do people then go back after the fact and say, well, did I really deliver it? Because most often, you know, when the tide turns against you, those are the things that get sacrificed. They get canceled. All intention to reschedule, but do you really then deliver five percent or six percent of development time? Or is that five or six percent got, you know, got sacrificed? So that’s one of the big questions where, you know, shrinkage, you gotta separate it out into a few different views, a time view, a category view. What we leverage it is overhead that can be capitalized on where we then can deliver it in real time. You actually can reduce some of your shrink preplan numbers if you’re just going and harvesting the variance in real time. Hey, Ted. Yep. One of the questions. Any concerns with this approach for very small segmentations? Perhaps FTE needs less than ten. Yeah. And, Jim, I don’t know if you’ve got a view on it. I think it’s it’s even more important for small segmentations because of the volatility that’s associated just in the small numbers. You know, when you get up to a queue that’s got four hundred or five hundred people in a single queue or even a couple hundred, you’ve got numbers working with you. The smaller the queue gets, the more volatile it’s going to be to handle time, volume, and staffing variations. Even more important of a reason why you need some overhead in and you need the ability to go and grab it back out when things are working against you. And I don’t know if you have an answer for that as well, Jim. The the only thing I would add to that is, obviously, depending on the size of the organization is if that’s all you’ve got, certainly, it demonstrates to your point the fact that you really need to be on top of how fragile your staff plan is. If you’re a larger organization and you find yourself with those, it’s a great opportunity to go back and talk about your economies of scale. I think in a lot of the larger organizations, it gets very easy to build out these very specialized groups, And then all of a sudden, you lose track of it, and you don’t realize how much economy of scale you’ve lost. So I think it’s a great opportunity to have that discussion. Yeah. I and the the comment that just came up, I agree with as well, and it talks to smaller and larger numbers. You know, best practice to implement staffing minimums, especially on opening and closing hours where requirements are low. Yeah. That is a best practice because at the end, you’ve got a far more volatility with some intervals that may have five or ten people or even two or three people, than you do on the intervals where you may have a hundred or two hundred people. So best practice to put it in helps protect. But at the same time, if I’ve got an opening thirty minutes where I’ve got ten people and, hey, you know, the volume has only said I need, you know, six or seven, then why not let those other folks do something productive rather than sit and twiddle their their fingers? So that as well is where automation comes into play. My WFM team, your WFM team, we can’t respond quick enough in the first thirty minutes to say, well, there still seem to be some people who might have some idle. Let’s get them, you know, going and training. That’s where automation is looking every second all day long, plucking off those little opportunities of one rep here, one rep there to deliver productive activity, to shift them into alternative channels at the end of the day to VTO them out. So alright. Well, I don’t see any other questions coming up, and we’re coming right to the top of the hour. Vicky, I wanna thank you for for having a for having us and Jim for contributing and, for all the community out here. I wanna thank you all for, attending our our webinar. I wanna wish everybody a happy holiday as well, Jim. Same. I appreciate it. Always great talking to you. And, Vicky, thanks again for hosting and and, always bringing the the folks. It’s great to see, strong attendance, and I hope to see everybody at the conference, next year. Yeah. Thank you, I guess, so much for today. I love that y’all came out in a Thursday in December just to listen to Ted and Jim and get some great information. I do hope everybody comes to the conference in, in April. You’ll be able to find Ted with no problem. Ted’s always the tallest guy in the room, so you can’t miss him. And he’ll be at the booth, I’m sure. Wanna make sure that everybody knows we’re gonna be sending out a link to the recording for this webinar and a PDF of the slides so that you can share that with other people in your organization if somebody wasn’t able to come. But we sure appreciate you being here today, and we’ll let you get back to whatever else is happening on this Thursday in December. Thank you so much to Ted and Jim for all their great information, and you guys have a great rest of your day. Thanks, Vicky. Bye bye.