Name: AI-Powered Schedule Generation: Solving Workforce Management at Scale
Uploaded: 2026-02-20T20:01:07.537Z
Duration: 44 min 48 s
Description: AI-Powered Schedule Generation: Solving Workforce Management at Scale

Transcript for "AI-Powered Schedule Generation: Solving Workforce Management at Scale": Hey, everybody. How's it going? Oh, I see there's at least one person from outside Denver. That's awesome. I'm actually here in the Capitol Hill neighborhood of Denver for folks that are familiar. Oh, wait. Hearing audio. One moment here. Can you guys not hear me? Okay. Great. Most folks can hear me. I'm so glad to hear that. You guys made me sweat for a minute. Alright. Well, thank you guys all for joining us today. I'm really excited to be talking with you guys about the next wave of scheduling here at Assembled. It really is the wave of the future. Today, we're gonna be talking all about how AI and Assembled are coming together specifically for scheduling. So really excited to chat through everything related to this super exciting new launch. And with that, let me just go ahead and get started with my screen share. I do have a little deck to walk us through today, but we are mostly going to be in tool, which is, hopefully good news for folks that wanna see all the action. Alright. So I already asked you if you can hear me. Can you see my screen sound off in the comments? But anyway oh, excellent. Thanks, guys. So let's go ahead and get into the meat and potatoes then. Hello. Hello. For folks that don't know me, I'm Sam, and I'm a product community manager here at Assemble. And really what that role means to me is it's my job to make sure everyone here has all the resources that they need to be successful and learn about Assembled as a product. So really excited to be here. And as a fun fact for you all, I actually used to be the support operations lead at a company called Imperfect Foods back in the day. That's actually how I met Assembled and came over. Back when I came over, we didn't have any of the automated bells and whistles that you guys are gonna see today. And I have to say when our product team released this stuff, I was like, oh my gosh. This is everything I would have wanted for Christmas a few years ago, so I'm really excited to show you guys, all the new bells and whistles. Alright. Well, continuing on down the road here. Do just quickly wanna point out a few features of our Boeing seven thirty seven today where the exits are and all that. Just kidding. This is a webinar. But please do make sure to use the right hand console for asking any questions, messages. Oh, yes, Erin. I am a roadrunner fun fact, metro state for life. And then do be sure to check out the docs tab for recommended content after this. Definitely lots of good stuff in there too. And then this session is recorded and will be going out after the event. So, yeah, these are some nice housekeeping things to keep in mind. Honestly, if you guys do drop into the comments, I will definitely see you guys there anyway. I tend to very quickly peruse over there. So even if you guys forget to use the questions tab or something, I'll probably see it, but it does help us stay organized. Alright. Continuing down the road here. So a little bit about what we're gonna be talking about today just in case I haven't already hammered this home enough. We already kinda went through introductions. I feel like most folks here know me. But today, we're going to be going over a quick overview of all the latest and greatest of schedule generation. We're also going to be talking through a live demo so you guys can actually see what this looks like in real life in tool. And it wouldn't be a good session together if we didn't wrap up with some live q and a. So do feel free to drop in questions as we go. I'll try to get to them in the moment if we do have a second of kind of airspace there. But if not, I'll be looping back around to everything at the end. Alright. So continuing on forward. So I think as workforce managers in the room, we all know scheduling is kind of where the cluster usually starts. And what I mean by that, I actually think this, this graphic does a really good representation of this. You're not only kind of handling all the forecasting things, like, you know, the demand spikes and, like, what's gonna come in where. And maybe you have this, like, perfectly well oiled plan and it's awesome. But then kind of inevitably something comes in like a wrecking ball. And now all of a sudden, like, perfectly laid plan just kind of really quickly just becomes imperfect, not to make a pun about my former employer, really, really quickly. And in addition to, like, having to keep up with things like those, like, same day spikes, same day callouts, all of those just kinda crazy things that can happen in the moment, we also have a lot to keep track of that gets complex really quickly, like labor laws and compliance. And for for those of us that operate out of multiple locations, we all know this can really quickly become a nightmare. Shout out to my folks that are, managing call centers out of Arizona. We know that there's going to be lots of things with, like, labor laws, daylight savings time, all of those things together. I say Arizona because that's where my call center was based, and I was managing folks out of California and Arizona. And, man, those places have really different labor laws, really different ways of approaching daylight savings time. And if you're not thinking about all those things all the time, again, it becomes a cluster. And then even working past that, there's also that kind of constant drip, drip, drip of changes that are going to be impacting you day to day. So I'm sure everyone here has probably received, like, a last minute message from a team lead saying, like, hey. Can we schedule a training at noon today to go over this really urgent thing? And now you have to find a way to fit that into the schedule. And, you know, again, perfectly laid plans fall apart pretty quickly. And then, of course, we're not scheduling for robots. We're scheduling for actual humans that also have their own needs around all of this stuff that we need to be keeping in mind. So that's my really long winded way of saying that scheduling, maybe even more than, like, forecasting, which I think sometimes tends to get more attention, is actually, like, maybe one of the most complex things we handle as workforce managers. Certainly, one of the things that has the most inputs coming from the most places. And if you're trying to manage this manually in an old school way, there's just so much room for not just, like, pulling your hair out with this being a labor intensive process, but also things getting lost in the cracks. Like, I'm sure everyone here has probably been in the moment where you accidentally violate a labor law, or maybe you schedule over an important meeting, or somebody had PTO on the calendar and now it's gone. And, really, I think that's where, as workforce folks, can lean in to automation, just take some of that manual burden off of us. Because, again, just like how we're not scheduling robots, we're not robots either that are going to just algorithmically keep track of every single thing we have to look out for. And that's actually a really good cue for where Assembled comes in and can be that robot that's actually gonna be keeping track of all of those things for you. So let's go ahead and just continue down the road once again. Quick poll question for folks. How are you doing scheduling today? Really curious how many folks we have doing it old school manual style, some kind of automation, a blend. You guys can check out that in the polls tab. If you guys want to let me know, please do. But, anyway, continuing down the road here, let's talk a little bit about why we decided to build out schedule generation. I think based on what I've already said, it's kind of a little bit of a no brainer, But we definitely found that it's one of the areas of business where there's the most need and the fewest good options. Because even in spaces where, like, supposedly some of this can be generated or maintained by kind of more robotic or automatic features. A lot of the time, those things are falling short or failing to scale with complexity. And, of course, it's not tooling that's available to anyone everyone to begin with anyway for folks who are doing this manually. So that's why we decided to build out schedule generation to really help out with especially, it's kind of next generation of complexity where things are often being held by internal teams, BPO, AI, who's doing what, where. Gets to be a lot more of a complicated question, and we wanna help you guys answer that at scale. Oh, and I actually just speaking to the poll that just went out, I do see we have quite a few folks that are manually using spreadsheets today or doing some kind of hybrid but still ending up in spreadsheets. It actually looks like that's the majority of the folks that we have here. So I imagine a lot of this probably feels pretty familiar for just the the day to day chaos, what can happen if you're not automating some of the stuff. So next, let's go ahead and talk a little bit about how schedule generation actually works. And I swear we're about to have the fun part where we're actually gonna get in tool and program in this end live together. But, generally, how it works is basically you set your business rules manually or using our AI features, and then we can generate thousands of schedules in minutes. Now to get a little bit more specific about what's going on here, there's a few different inputs that help us arrive at an actually optimized schedule. And for a lot of folks that have used other tools out there that claim to be able to do some of this generation stuff, I actually think this is where things often fall apart is really what are the inputs and how are they stacking together. And I'll speak more to that in a minute, but the assembled approach to how we handle this is the first thing that we're looking at is going to be what business rules do you have set us set up for us to think about. So that's things like what's agent availability looking like, what labor laws are we holding ourselves accountable to, Are there any special considerations for things like everyone has to have exactly half and half chat and phone time because I don't want an agent knocking on my door saying this agent got less phone time than they did, and that's not fair. Those are the kinds of things that we wanna be thinking about. And then in addition to those more business specific rules that you can tell us about, we're also gonna be looking at things like, okay. What agents do you actually have on deck for us to be able to schedule with? What needs can they meet from your forecast? And then, of course, globally, what does that forecast look like, and how are you what are the different combinations, I guess, you could say, of schedules that could land us at SLA? And like I promised I would get into a minute ago, I think where often these kind of generators fall apart in my experience is usually the order in which they're trying to layer these things. And, basically, it'll be like, well, this is scheduled that'll meet SLA, but it's not meeting my compliance laws now, or it's not actually considering when people are available. Because most generators aren't actually considering all of these things at once. They're basically considering them one line followed by another line followed by another line and changing the schedule kind of as they go through the funnel. And then you end up with something that is kind of the worst of all worlds that doesn't really meet any of the needs that you put in cohesively. So one of the special things about assembled is we actually have a special generation logic that uses AI to basically incorporate everything that you have entered all at once simultaneously and then actually be able to spit out something that's cohesive across all of your needs instead of only considering one at a time and ending up with something that looks a little gibberishy. Alright. So with that, as promised, we're gonna get into my favorite part, which is actually getting into the tool. So give me just one moment here to flip over my screen share, and we will be getting into the fun part, folks. Hopefully, other folks think time and tools as fun as I do, or else maybe I'm overselling it. But I think it's really fun. So, anyway, I'm screen sharing into Assembled here. And the first thing I like to call out is kind of how we know what we know from the jump. So to begin with, if you guys go over to configure, assembled sits on top of places where you already do work. And for folks in the room that are already assembled customers, a lot of this is probably gonna sound pretty familiar. But the first thing we're doing is we're basically looking to see how much contact volume's coming in, and we're actually using this as our basis for predicting into the future. So there's definitely lots of other webinars where we get way more into the actual math and foundational features here. We just wanna do a quick flyby of how we know what we know. We basically take that contact volume. And the first thing we do is we break it out into different channels or also different work streams and queues to understand exactly how much contact volume. And sorry. This is actually taking a minute to load for me. Exactly how much contact volume you have and how much we're predicting is going to arrive for you in the future. And, again, lots of things that we could get into here. Oh, I see a good question. Contact volume is based on tickets as compared to order volume. That is true. So out of the box, our configuration is we're looking at tickets that hit your CRM. There are definitely a lot of things we can do to consider other types of inputs. I'm just speaking to the default here. When I used to work at Imperfect Foods, how we would actually and not to give away too much proprietary information, but how we would forecast was actually based on how many subscribers we have inside of our system, and we had a general sense of our contact per order ratio that would basically tell us, you know, for every thousand subscribers we have, here's how many of them are likely to reach out to support. And that was actually our base model for forecasting that we hybridized with data from assembled for historic ticket volume. So, yes, it is historic ticket volume based, but, no, you can also add lots of other inputs too if you wanted to, and happy to talk about that more later. But, anyway, so the first thing we do is we're predicting into the future how much contact volume's actually gonna happen. And, again, you can break this down further by work stream if you wanted to. So if we wanted to say how many tickets in French are coming in, we can do that too. But for now, I'm gonna keep things at a high level, and assembled will also ingest things like what SLA are we actually targeting for every bit of work that we do. And this really is the foundational layer that feeds into everything else and lets us know basically, hey, assembled. I want you to build forecasts for me that are designed around meeting this SLA knowing that I in the background that we have this amount of contact volume. And I won't be touching on this as much today in our live demo, but we'll also be thinking about things like, hey. If you're predicting on French speaking volume and these are your agents that are available to speak French, what should their schedules look like versus, your folks that are more generalist in order to make sure you also hit that French speaker SLA? Thanks, Jennifer. Alright. So continuing onward down the track here. So once we are incorporating your specific forecasted values, your specific SLA needs, all of those kinds of things that are important to you as a business, the next layer down is we're taking into consideration what rules, basically, you would like us to keep in mind for how we're going to generate a schedule into the future. And the way I like to characterize it for folks that are already assembled customers is if you didn't put anything in here at all, assembled robot is basically just maximized to hit your SLA by whatever means necessary. But the things that the robot doesn't usually know are things that matter to us humans, like when are people going to take a lunch break or when is somebody available to actually do something in the sense of, like, what's their preference for when they should be scheduled? Because, you know, if you schedule people twenty four hours a day without any lunch breaks, yeah, you'll probably hit SLA, but you're going to have burnt out agents and something that's pretty illogical for real life. So this is where you get to bake in all of those things that we consider as humans and workforce managers. So let me go ahead, and I'll show off the first layer here. So the first thing you can do is you plug in the specific working hours for when people are meant to work. And you guys just saw me kind of whip up one pretty quick there. That's a standard nine to five. But this could also just as easily look like, hey. Maybe every other Friday, we give folks a half day. And oh, sorry. I misclicked there. One second here. There we go. So maybe it's the every other Friday, I'm going to give people a half day. So I can actually get in here and very precisely say that this is a schedule that's changing every other week to account for that. And this can really look like whatever I want. So this could be part timers, full timers, overnight people. Whatever makes sense, we can cover it here. And then, basically, you just drop in the schedule that somebody's working, and then you're gonna apply it to people. So unlike in kind of the the days of your when you might have to, like, have a template that you apply or manually schedule people, and then it becomes cumbersome when people change. Now when there's a change of schedule that's long term, it's as easy as just clicking this plus or minus button for who's assigned to what. Yes, Erin. I do think that for folks that are familiar with our old school shift patterns, that's a good way of thinking about it. This is basically shift patterns and rotations built in together into one really great feature. So that way you don't have to manage any of that separately anymore. So now you can just layer in all of the right people to be involved in this set of working hours. And you can have as many different working hours as you like. Typically, what we see people do is they would usually have kind of one working hour set for every kind of set of availability on the team. But, again, multiple people definitely work the same schedule for most companies, so you can also group people together. You can basically kind of slice and dice this however you want. Now the next layer is events. And I basically think about this as anything you wanna have added to people's schedules that isn't that productive ticket time. So this could be things like lunches and breaks, or this could be things like, meetings or one on ones, trainings, things like that. And before I go deeper, since it sounds like there are some folks in the house that have used prior versions of assembled or maybe existing customers that haven't seen this as much, our vision here is that this will replace our optimizations feature and basically do all of that and more with much more granularity. Oh, thank you, Judd. Yes. We love the UI too. So getting into the specifics here, I can actually schedule something like a lunch break. And let's go ahead and say that a lunch break is always thirty minutes long, and we want that to happen exactly one time every shift. Now this is where I think the, flexibility of this feature is really fun because there's a lot of different ways that you can actually configure this. So this can goes exactly where you want it and never where you don't. So out of the box, initially, this will say any time. And that basically means, hey, scheduling robot. Plot this lunch at any point during the day when it's maximized to preserve my SLA. But in real life, we often have constraints we wanna keep in mind that are beyond that. Like I mentioned earlier, I used to work in Arizona managing a call center out of California. And folks that are familiar with California labor laws will know oh, one second here. I'm I'm misclicking. Folks that are familiar with California labor laws will know that you always have to have a lunch before the fifth hour of someone's shift or else you could get in trouble. So we definitely wanna bake that in. And at the same time, we also know that we probably don't want people to take their lunch break in the first hour of their shift because it would be weird and uncomfortable for our agents. So I'm gonna go ahead and say it has to be between the second hour and the fifth hour of their shift for when this is gonna start. One other thing that's really cool here is I can also add additional conditions for who this is going to apply to based on their specific schedule for the day, even if it's a rule that usually would apply to them as a person. So for example, what that means is if agent Sam has this rule applied to her profile in general, but agent Sam is working a half day one day and we don't wanna plot that lunch break for her when she's only working a half day, assembled can account for that with just this conditionality clause. So we can say we only wanna apply this, rule here when the shift length is greater than, let's say, six hours because that means per my labor laws, this person should be getting a break. So next up, let's go ahead and talk about how we would actually apply this to a person. And that just follows the exact same rules as the working hours pieces we've already talked about. So I'm just getting in here and plusing them. For something like lunch break labor laws, usually, what you would be doing is you'd probably be doing this geographically. So, for example, maybe instead of applying it person by person, I might actually apply it to everyone that lives in Colorado or everyone that lives in New York. Of course, every team's different, but, basically, I'm saying you can split this out in a location based way if you like. Or if you're a smaller team or maybe everyone's working out of the same location, you could just say add everybody, and now we're we'll be keeping track of that labor law for everyone. And the only exception would be if they worked a shift that didn't meet that requirement. So next up, let's go ahead and talk about one more kind of rule. So the other kind of rule is kind of for things that happen on a less regular basis. So perhaps a a show of hands in the chat. Who here has been asked to suddenly schedule a team training or team meeting for somebody? I know I have. Feel free to sound off if you have, or you can not. It's up to you. I realize I'm I'm asking for a lot. But anyway so oh, yep. I see a yep. Thank you, Bart. So, anyway, this is our answer to that. So inside of scheduling rules, very similarly here for event, I can say, let's say, training. And we know that this training is gonna be sixty minutes, because, the trainer isn't thinking about our utilization scores, they say they need sixty minutes. Sorry. I'm kidding. That's a shady joke. But, anyway, we have sixty minutes plugged in here. And instead of repeating this every shift, maybe we're saying that this is a weekly training. And by the way, you don't actually have to apply this weekly if you don't want to. This is just an example. And if you leave it up to assembled choose days automatically, once again, this would just be saying, hey, assembled. Just put this wherever you need to to make us hit SLA. But some teams do wanna be a little more specific about this. Like, maybe our team trainings always happen on Friday, and that's what we wanna do. Up to us. But I'm gonna go ahead and let the robot do its work and say Tuesdays automatically. And now just like the lunch break rule, we can once again be really specific here. So something like a training, we're often just gonna say anytime because, really, it can happen anytime as long as it meets SLA. But maybe your team doesn't operate like that, and maybe we wanna say that instead of doing it that way, it should always happen at the end of people shifts, or maybe it should happen halfway through. Or maybe we have a unique scenario where we always wanna have a live training immediately after our company all hands. Like, this can really look like whatever you want it to look like. Again, a lot of different ways to slice and dice here. And down below, this is where I would say this gets kind of the most interesting compared to the lunch rule we just set up. Something that's often really tricky about scheduling something like a training is this I I guess you could say the synchronicity. That's a hard word to say. Synchronicity of a training. And what I mean by that is, like, there's some trainings where we need everyone in the room on the same time. Like, all of us folks are here in the room together at the same time, having the same conversation. It's important that we're all here. The other side of this, though, might be an individual training. Like, let's say it's more of a Loom video or maybe HR wants to give folks time to fill out their w twos or, like, watch sexual harassment training, that kind of thing. In which case, you know, we don't really care when they take it in relation to each other as long as it doesn't impact SLA. They're gonna be alone in the room anyway. And then the last one would be what if this is involving another specific person? Like, this would be most common for things like a one on one with your manager or a coaching session where, basically, like, we need to make sure that that coach is also available for the person. And that's where these special settings down below come in. So shared is what it sounds like. That's saying that this particular event needs to be shared across everyone I'm applying it to, so it needs to go on at the same time. Separate would be saying this always has to be staggered. So perhaps there's, like, one trainer on the team that's having a one on one session. And if there's 20 different agents on the team, this needs to happen at 20 different times because the trainer is only able to be in one spot at once or individual. And once again, that's more of the we don't care where it happens in relation to other people as long as everyone gets an hour. So you can actually get in here and just plug in any trainings, meetings, all of those things following this very similar rule to how you might set this up for lunches or breaks too. And the last one I'm gonna talk about in today's session is productive work. And if you don't fill this in, Assembled will make inferences about what's the best thing to do with who's working chat, email, who's working this queue, who's working that queue. But if your agents are anything like my agents, I kinda made a little bit of a joke about it earlier, but we know oftentimes splitting time between channels is a really delicate balance because you probably don't wanna get a knock from an agent on your door that's like, hey, agent Melissa is actually working two hours less phone time than me this week. What's going on? Why do I have more phone time than Melissa? So if anyone has agents that are similar to mine, this productive split will help you out. And, basically, what this does is it kind of puts in artificial guardrails for how much time people should be spending in each sorry to call you out there, Melissa. How much time people are spending in each channel. So inside of here, I can specify, hey. This group of people that I'm talking about, they can be scheduled for chat, email, and let's go ahead and say phone. We'll keep it general. But when it comes to time limits, my agents are really sensitive about how much time they spend on phones without burning out. So I'm gonna say we want them to have a maximum of four hours every week they're gonna be or not every week, every event that they're going to be doing that. So we won't have an event that's longer than four hours on phones. But at the same time, I know if people get scheduled too much lower than four hours, I'm probably gonna have a bunch of people knocking on my door asking why they can't be the one that works one hour on phones. So if that's the case, I can also put in a minimum threshold here. And this minimum threshold can also be really helpful, by the way, if it's more just a setup where it's like, hey. Everyone across the team has to take at least four hours of phone time a shift. Like, we can definitely plug that in here. So this is basically your control room for burnout, fairness distribution, that kind of thing. Again, if you don't put any settings in here, assembled will just automatically infer kind of based on who's on the schedule and what would be most efficient. But this is just optional guardrails you can put in, and you can do this either by individual chunk of event or per shift. So during one shift, how many hours can somebody spend focused on one thing or another? Another thing that can be really helpful to set up is what is the maximum amount of time people can context switch? So a lot of the times, feedback I've heard from agents is like, hey. I don't wanna be switching between live channels every hour on the hour because it just takes a lot of mental capacity to transition to that new kind of work. And if you guys get that kind of feedback, you can also plug in guardrails here like, hey. We only want people to be switching between channels up to three times a day. Oh, sorry. I put minimum. Three times a day maximum. And then maybe we want them to do it twice a day minimum, just as an example. So, anyway, I hope now you guys have a good sense for some of the different guardrails you can plug in using scheduling rules. And the idea here is basically once you have your rules built out, you then land in that click a button, get a schedule world. Oh, Mark asked, does this training configuration support exceptions such as excluding specific employee pairings or restricting matches based on-site only? Yes. It does. So the way that that would work would be things like whatever you're setting up the rule, you still pick and choose who it applies to. And, honestly, I probably went a little bit too fast through this application piece on the last one since it is the same UI as working hours. But, basically, you would just get in here and pick and choose the specific people. If you prefer to do this by filter, you can too as well. Like, if maybe your BPO partner has a training that your internal people don't, it's a pretty easy thing to plug and play with. Or another common setup I've seen is like, oh, hey. Well, maybe it's Christian's team has this training, but Frederick's doesn't, and we don't want them together. We can absolutely run it that way. Oh, getting some really good questions, guys. I actually think I'm gonna loop back on these at the very end, but I will be back. I promise. So the next phase here, once we do a generation, is over here in this dashboard. And this is basically once again where you do that motion of click a button, get a schedule. So you can just go ahead and click this button. And then I'm not actually gonna click it here just because we have a pretty limited window of time together, and it does take a few minutes to load. I will call out for folks that have used other providers in the past. Another thing that makes this special is this moves usually much, much, much faster than it does for other solutions out there just because it's built on newer tech. So usually, a generation can be done in less than five minutes even for really large teams. But anyway, getting inside of here, what you would first see is we would pump a schedule into your test schedule environment that would look ideally, like, 90% of the way to where you want it to be. And it should be compliant with all the different rules you entered, all the different trainings that you entered, everyone's availability and working hours, and maximized as much as possible to hit SLA with the head count that you actually have. But, of course, we wanna pump this into a test schedule first just that way you have a little bit of wiggle room for getting in and editing things when you need to before it's actually live for agents. So this is kind of your step usually for just getting in and seeing things with your human eyeballs that maybe our robot eyeballs missed. Another metaphor I often use here is, let's say there's, like, a one off event you're trying to plot for that wouldn't typically be something that's part of a rule that's gonna happen on a regular basis. Like, maybe this is something like this Thursday, there's a birthday party for Melissa happening in the break room. We have cake, and we wanna make sure everyone shows up for an hour in the break room. Those would probably be the kinds of things you would wanna put in manually just because it's not gonna be Melissa's birthday every Thursday. But, anyway, happy to talk more about exceptions to that in a minute. But, basically, you would just edit in here. And then whenever you're ready and you you dubbed that this is a good looking schedule, you can go ahead and just publish it out to your default schedule. And once it's published, that oh my god. Is it actually Melissa? So, anyway, once you have this set up, you would publish it out into your default schedule. This works very similar to a template for folks that are familiar with assembled UI. So you can also do things that you're probably accustomed to, like keeping any customized events versus replacing all events and allowing conflict. But, basically, this just pushes this out of the test schedule and makes it real, moves it into the default schedule. And then this is then what all of your reporting, what your agency, your adherence, what all of that is gonna be based on. So with that and, yes, happy birthday, Melissa. That pretty much takes us to the end of what I had planned as our demonstration for today. I'm gonna quickly post one more poll. What part of schedule generation are you most excited to automate away? And while you guys are answering that question for me, I'm gonna go back up in our chat and answer a few more questions for you guys. Hannah asks, could Assembled handle a multi week training program? E g, week one, they have three hours of content. Week two is a two hour mentor session, etcetera. So the thing maybe a little harder to capture there, and I'll I'll be really transparent with you, Hannah, is the staggering of how long the event is going to be because assembled rules do want you to be a bit prescriptive about how long an event is. And, typically, if it's going across multiple weeks, basically, system's expecting it's the same time or it's flexible and can be anytime between kind of two goalposts. But what I would usually recommend if it's something like week one's reliably three hours of content, week two is reliably two hours of content, etcetera, You could potentially set up multiple rules for that that apply to different weeks at an alternating cadence, and that's probably how I would handle it. It would just be through multiple rules. It just wouldn't be through one. Jeremy asks, I think this question goes back to working hours. Can you manage hours of ops for a group of employees and employee preferences separately? For example, they could possibly be scheduled between 6AM and 11PM, but their preference is to start between 6AM and 8AM. Attempt to match preferences first, but if it can't, put them as close as possible within the broader hours. In general, we kind of view this as like a single layer. So, basically, whatever working hours you input are the working hours we assume for that agent. And it's kind of usually a mix of, like, what hours you guys actually need from them and what hours they're specifically available to work just depending on your use case. But most commonly for a lot of the teams I work with that are using this already, it's kind of more like every agent has a specific shift that they usually work. And that could be a rotation that could change later, all of those things. But, usually, it's the shift that they would work that you would want them to be on that you would input in as working hours. And, the shifts people arrive on once again are kind of a combination of business need and agent preference, but typically that's entered in as the working hour. And that shift becomes our implied default. And, again, can be rotation, can change later, all of those things. The last thing I'll touch on here is the next iteration of schedule generation, which I can't show off live here yet just because it's not ready to go, is we are making a version of this where it will be more assembled, tells you what people's shifts should be as well. Right now, we are going based off the working hours you enter, but in a near version of this, it will be more you give us a sense for when people can work, and then we'll tell you what shifts make sense inside of that. So I would say that use case would probably be more covered in an automated way by a future release. If I have a person and I don't care if they work a 9AM shift or a 6AM shift, do I have to include them in two separate working hours roles? Usually, you would just include them in one and just whichever one makes the most sense for you. And then, of course, you know, we do have agent flexibility features separate to this where they could request a change later if they wanted to, if that's part of your workflow. But generally speaking, you would assign them a specific shift. And then once again, just like we were talking about a moment ago with Jeremy, a near version of schedule generation is going to release that would basically account for that use case more by having a kind of wider variety of possibilities and then more just generating a schedule within that range. And we would the way I describe it is less you tell us when people work and more we tell you when people work, and that's coming out at some point very soon. Alright. Oh, one more question from Mark. Is this system fully standalone, or does it support integrations and API connections with external systems such as payroll and Genesys? So there's a couple different things that people often mean when they say integration. Assembled is usually integrated to whatever CRM you guys are doing work in, and we do directly support a Genesys integration as well specifically. But that'd be things like Zendesk, Salesforce Service Cloud, Intercom, kind of all the usual suspects we can integrate with for call volume. And that also allows us to do the that forecasting magic I was showing off the beginning of the call as well as a bunch of different performance monitoring things that are probably pretty familiar for most folks in the room. So we can really do all that bread and butter workforce management stuff connected to really wherever your work is happening. We also have a really robust API situations. That way, even if you don't have something we integrate with with one click out of the box, we can usually custom build things with you as well for things that are a little bit more niche. But beyond that, as far as the payroll piece goes, I've seen people do HR integrations via API, but we don't have any payroll integrations specifically. Can HR information be sent back and forth with HR systems? So what we do support out of the box for HR integrations is time off imports. So for example, if a time off request is imported in one place, we can pull it into assembled automatically. Specifically, we support out of the box integrations for that with Workday and HiBob more on the horizon. And there are some people that automate with other vendors using our API. Transparently, for the pushing back of information from assembled into HR platforms, it's not so much of a technical limitation as it usually is a kind of culture, limitation because a lot of the times, HR teams will not let us write information into HR tools just because HR tools tend to be very sensitive and picky about giving vendors access. But if it's something that your team's open to, we can always talk about solutions there. But most of our teams that have an HR integration, it would be read only. Alright. Cool. Well, any last questions I can help with in the last three minutes here? Oh, can you export flat file with that info for HR to input? Yes. You can definitely export information out of assembled for HR to input more manually instead of the API if you prefer. Actually, while I'm while I'm here and I have a a couple minutes, I might as well say it. Assembled, one of our unique things that I think is really cool about us is somebody that's used a lot of the vendor systems out there. It's pretty much everything that you see in Assembled is really easy to export and potentially import to other places. We don't generally hold your data hostage or make you jump through a lot of hoops to get your data or contact support or anything like that. Pretty much anything you see in assembled is exportable, and you can pretty easily take that and move it to other places when you want to. Alright. Last two minutes here. Any last questions I can answer for folks? And I'll give us till we have exactly two minutes left. That'll be in about forty seconds. But if no new questions come in, I will go ahead and wind us down. Oh, is there a way to show the approved leave on the test schedule screen? That's a really great question. I'm actually not sure on that one, Mary. I think we would wanna loop back to you on that one. I know for sure whenever you approve time off, it appears in the default schedule, but I don't know about carrying that over into the test environment. Usually, how I see people handle that use case, though, is it would be approved and then appearing inside of the default schedule. And then whenever you do move to publish what's in your test schedule out into the default, you have the option to keep customized events. And usually PTO would be considered a customized event, so we kinda preserve it in the last push. Oh, interesting new integrations into other platforms in 2026. So Genesys is actually a newer integration for us that we're really excited about. Trying to think if there's anything else that I am especially excited about integration wise, hope, going on right now. But we are always thinking about new integration partners. We're always expanding there. That's definitely something where, like, every single year, the amount of things we integrate to really expands. Another newer one that I've also been really excited about, though it is already out, I wouldn't say it's a 2026 integration build, is we do integrate directly with Slack as well. So a lot of the information you've seen today can be built in the Slack workflows, and we can also do notifications inside of Slack for things as well, which is pretty cool. But we'll I'll definitely, we'll be sure to keep you in mind, Jennifer, for any, integrations that we're building. And if there's anything you wanna see us build, definitely let us know. That's a huge part of how we build. Alright, guys. We're we're now down to our last minute here. You guys have been a great crowd. Please just let us know if there's anything else we can do to support you or if you're curious about any of these other technologies we showed off today. We will be sending this recording and some related resources to you guys over email after this. Feel free to share it around to folks that couldn't make it. And, yeah, you guys have been great. Really hoping to see you guys in another one soon. Bye, guys.