Name: When AI Works for People: Transforming Engagement, Retention and Efficiency
Uploaded: 2026-01-21T17:29:14.712Z
Duration: 46 min 33 s
Description: When AI Works for People: Transforming Engagement, Retention and Efficiency

Transcript for "When AI Works for People: Transforming Engagement, Retention and Efficiency": Thank you for joining us today. We are here to talk about a topic that probably nobody ever heard about called AI. Well, but focus for us with our AI conversation today is to make things, to keep it real, to really figure out practically speaking what's working, where things are today versus the hype, What are things that are happening in retail stores versus in the lab? And we've got an amazing panel today that can actually share their insights, share their experiences and be very open about what's working and what's not. My name is Sanish Mondkar. I'm the founder of a company called Legion. We are reinventing workforce management and our goal for workforce management, I founded this company about ten years ago. Our goal for workforce management, the founding thesis was predicated in two important objectives. First was to make workforce management also about the workforce to add value to the workforce, not just optimize the cost of the workforce, because we strongly believe that is what leads to better store experiences, both for your customers and for your company, but also to automate workforce management. And, as a result of automation objective, we have always bet a lot on AI. We've always bet. We have looked at AI as a how for the objective of automation from day one. And so there is a lot of experiences that we bring to the table and we would love to today have this conversation with our panelists here and I would just proceed now to have them introduce themselves. So Steve, please go ahead. Hello everyone. My name is Steve Page. I'm the head of global technology transformations for Inka Group. Inka Group is the largest single franchisee for IKEA. We run 87% of all of the big blue boxes that you see and have a significant presence obviously globally, but very much in obviously Sweden from a digital point of view. My background is very much I came through the recruitment and HR's P and C perspective and have spent a significant amount of my time looking at technology transformation and underlying technology enablement for organizations around the people area. So, that's me. Hey everyone. My name is Ryan Holm. I'm the DVP of Retail Innovation and Operations at Helzberg. We're a full service jeweler. We operate 161 stores across The United States. Been in business for one hundred and eleven years. And since we're in New York, I feel like I have to share this little anecdote. Our founder's grandson actually found himself on the streets of New York City in 1995 and saw someone across the street that he really looked up to, who's Warren Buffett, and went running across through traffic to introduce himself and pitch selling his business to him. And we have been proudly owned by Berkshire Hathaway ever since. So, if there's a lesson in there, if you see someone you run into at NRF you look up to, introduce yourself. Maybe you'll, sell your company. Hi. I'm Karen Beebe. I'm with Bealls, Inc. Bealls, not "Beels". It's Bealls. Think of the Bell. And we all are also a 111 year old company. Both of us were founded in 1915, and we have, six hundred and fifty plus stores across 23 states and of course, online as well. And we outfit the family for less under our banners of Bealls, Bealls Florida and HomeCentric. We have, a unique family led and when you think about 111 year old company, you think probably there's a lot of legacy things going on, or you may have some things where you may not be as innovative. Our kind of fun fact of the company is we take crypto in our stores. So you can go and you can buy and purchase whatever you want leveraging crypto. Pretty, right? And soon to be online. That's great. I was quite proud of the fact that we were 83 years old thinking that we were gonna be good, now we're the youngsters. Wow, that's cool. Well, so talking about crypto, let's talk about the technology that's powering your frontline workforce. The question on everybody's mind is how has AI changed the day to day experiences of frontline workers? So I know we are going to go through a lot of questions here. The goal is to keep it as practical and as real world as possible and staying away from the hype . So let's start with what aspects of workforce management were pain points before AI and how have they improved? And I'll start with you, Steve. Yeah, I think many retailers with many companies have issues around manual scheduling. The more manual it is, the more prone it is to have mistakes and errors coming aside as well. This then leads to other issues as well. So you end up with inaccurate demand forecasting. You're in a situation where you leave yourself exposed potential bias or at least perceived bias by management teams or by your coworkers. And you're just not, you're not quick to market to change how things move. So I mean, everything is moving so quickly nowadays. If you're manually enabled, you can't really move at speed. So this is quite a problem that goes across. I think shifting it to AI, a lot of the administrative elements across the AI sort of frees up your workforce to do a lot of things, but it also reduces your potential issues around compliance because this is a big thing for us at the moment is that many of us have GDPR and data protection rules that we need to stick to. We have unionized agreements that we need to be sticking to. We have our own values and we have our own guidelines within our companies that we need to enforce. But if we're doing all manually, these get lost in many ways. So I think transferring administration over to AI is a big thing. I've built my entire career on eradicating admin in organizations and trying to make things easier to manage in that way. So I think AI is making that a lot quicker to do and making it easier for us to comply. And I think that makes a lot of sense. Efficiency and things like compliance management are really big pain points, right? One big question on everybody's mind always is, and I get this all the time myself too, is AI replacing the people actually doing some of the things that you talked about, like around compliance, maybe around productivity, around efficiency. What's your perspective? And maybe Ryan, I'll start with you. Yeah, mean, my role, I'm one of those rare people that came up through the stores. It was not that long ago I was managing territories for Helzberg. And through that, firm belief and what I'm very mission driven about is retail frontline associates and store managers are super humans. They are incredible. They multitask more than anybody. They care about their customers, about their teams. And if you have the right people in place, it's exactly what Steve said, right? It's eradicating that administrative burden on them. They're human connectors. They want to be multitasking. They do not want to be locked in an office writing a very complicated Sudoku puzzle, which is a retail store schedule. And it's a Sudoku puzzle that if you don't do it right, the company could be sued, speaking of compliance, or you're gonna have unhappy employees turnover, you're not gonna be able to serve your customers. There are huge ramifications to getting it wrong. So I'm of the belief that if you can give a store manager back thirty minutes in their week, they're gonna do one of two things. They are either going to go home and rest and recharge with their family and come back better able to serve customers the next day. Or they're gonna spend that time training and coaching their team and doing things that I don't think AI can soon replace. I mean, I couldn't agree more. I mean, this is one of the things that we need to keep a firm eye on is the fact that it doesn't really matter how you improve their lives or how, what they do with that extra time getting back, as long as it improves their mindset, their feeling of well-being and they're actually being more productive with that. So I totally agree. We just need to create the environment where they're feeling better about the job that they're doing. Exactly. And that's, I mean, that's super important, right? You would need your managers to be feeling motivated, feeling productive, feeling like they have the store under control versus the other way around. Karen, at Bealls, how has AI reduced stress? Like Ryan made the point of managers having superpowers and they have a lot of stressful things to deal with. So how's that working for you all? Yes, so we recently had implemented Legion and had put in from the automated scheduling process. And we have, it's really changed night and day compared to where they were and what they were doing and how much time they were just spending on scheduling and the craziness that you talked about. And now they're, I think the exciting part is they're actually, they quickly adopted when they saw how, how great the schedules actually come out and it aligns with how they're thinking and the process that they were going through. They were energized by it, saw the time savings and could see where they could put energy into the guests and to their associates. And so it was, it's not like we had to say, you need to ensure that you do this. They wanted it. And then when some additional stores were piloting it, they saw the other store managers of what they were doing. It just actually got contagious in a good way. I think that's a really great way to position things because oftentimes the question about AI is, is it replacing people? Is it replacing things? But it's really about what we are talking about here is how it's improved the lives of managers, how they are embracing it because it's been, it's helped them be better managers and be more productive. Steve, let me ask you a question, a follow-up question on that. In your world, can you share some examples of how AI has contributed to those types of outcomes? Yeah, I think one of the things that we're trying to do at IKEA and I think it's being done at many companies is we're trying to get our coworkers more involved in setting their own standard, given their own preferences, getting involved in the pre scheduling stage. So telling us when they want to work, how they want to work, where they want to work, all of these sorts of things around, whether they have a school run that is really important to them, whether they have childcare that they need to consider. All of these sorts of things are really important to work life balance. But if we don't know about them, if we don't have a place to deposit them, we can't actually action them. So I think getting that input very early on and then combining that with an AI engine that can take the demands or the requirements of the business along with the preferences of our employees and then actually optimizing the schedules to combine both, you suddenly get something which energize your employees because they know that their thought process is getting taken into consideration. You get more schedule predictability because you know that people are gonna turn up because they've already specified that those are the hours that they want to work if they can. They also know that if it doesn't quite meet what they want, they've still been considered and there's still opportunities to swap shifts or use dynamic elements that are within the technology to be able to enhance it as well. So all of this really brings sort of a more heartfelt, more warming sort of mentality for our coworkers and their view of the company. The other side is that you can reclaim hours that are wasted with people not knowing necessarily where they should be or what they should be doing and it could make people more efficient. One of the things that came from our coworkers is they wanted to know they were adding value. They wanna spend time with the customer, but if they're doing administration work or if they don't know where they're meant to be or they don't know where something is meant to, they're meant to find something, they're wasting time. And this again, eradicates that from the place because they feel that they're getting more out of their job as well. That's really important as well. Following up on that, if, is a way for AI to be used to provide employees with their preferred hours or preferred timing and things like that. How do you ensure transparency and fairness and how important is that? I think the transparency thing is getting a better understanding is getting a better understanding of exactly exactly what what the, they want and to have that you have to have the data flow. You have to have the understanding of exactly what it is that's being requested. You need to be able to use it in a more intelligent way. And I think doing that and having AI drive that rather than actually it being driven by a manager, you then get rid of that bias. You get rid of the thought process of I would prefer, or this works for me. And actually you get this optimized brain actually making the best call on your behalf. And that actually means that it's much more transparent to the coworker. They've got no one to complain to because they can say they don't like it, but they've got no one to blame or to point at. And again, they've been considered in the entire process. Managers also then get the situation where they're provided with insights. It's not just data points. They're being provided with insights that tell them why this is happening, why they need that particular person over here rather than over here, why that skill set is required in that particular area. So it's given them more understanding of why it's set up as well, which then means you can implement standardized rules. You can implement guardrails that apply to everybody. And again, then you get the transparency, then you get the consistency, then you get something that people understand and can refer to. Managers still need to manage. That's not what we're trying to take away. AI is just enhancing their role. It gives them the ability to have the information at their fingertips that are gonna help them make the right choices and also give them the ability to be able to justify any choices that they make if they're required to do so. So that's really important as well. And it becomes much more equitable. It becomes much predictable. And then you, again, it's a happy coworkers, happy customers. And that's what we're looking for. Just when he was going through that, it made me think. And when Ryan said superhuman of the store manager, just think about how many data points they had to remember or keep in their heads or write down and then pull it all together. And by leveraging AI with all of those data points, you get that consistency, you can get that recommended method that should be consistent across all of your stores, right? And it's just, they just wow, superhuman managers and how do we now enable them to go do what they are really great at is be with their store associates and the guests. And it also creates some trust as well, because I think if got 95% of a particular task done and all you're now doing is checking, validating, making sure you're comfortable with the outputs, making changes, making sure you have input, maybe additional data that you've been given, you're there can input into, you're still in control. You've not lost it. What you've done is a significant amount of the workload to get to that point. You then still carry on managing. I think that's the thing that builds that relationship a little bit more as well. And if I could, I think you're both spot on, but talking about the transparency in AI, right? I think I've certainly got partners in info security who really wanna have some detailed questions about the model being used, right? But when it comes to trust and transparency with the retail associates, the analogy I think fits here is, it's like going to a restaurant, right? When you show up, you might have questions about the ingredients. You might wanna know is this vegetarian? Is it organic? And you have opinions about the outcome, right? You may want your steak to come out well done and you care about that outcome. At no point is anyone likely asking what model of stove they used in the kitchen to cook the food. And I think that's the same with retail employees, right? They want input. They want to talk about the ingredients that went into that schedule and be able to have a voice in that. And they wanna make sure it comes out as a quality schedule that works for them. That's an excellent point. I think the, we have seen over the years as we have built Legion, there's been a very specific input from the employees and it's all around transparency, understanding what the ingredients that were, like whether their inputs were taken into account and they're okay with sometimes not getting the schedules they're looking for or they wish they had, but they want to know that the inputs were actually taken into account and there was a fair process of evaluation of the inputs before the schedule was created. The other thing I wanna go back to is what Karen said about the number of data points. So I remember a few years ago we did this analysis at Legion, where for a roster of 30 people, which is not a very large roster, if you take into account four inputs from the employees, so their availability, the number of hours they're looking for, the days of the week they're looking to work at and the preferred shift lengths. Then there are over 250,000 possibilities of creating the schedule and it is absolutely impossible for a manager to do this and really actually look at the employee and say, yes, all your inputs were considered and we followed a fair process without something like AI or really advanced technologies being deployed. So that's kind of the bottom line of fairness and explainability where you can actually honor sort of the commitment to the employee saying, yes, all your inputs are taken into account and this is the best possible schedule that works for the business and for you all that is generated for this particular week. That's great to hear that explainability and fairness is a big point for everyone here. I wanna add a little bit to that as well. We've all been using AI in retail for many, many years, right? Forecasting, in machine language, like we have got that embedded in a lot of our inventory systems and in scheduling systems actually, but they were usually either a black box and you didn't know what was going on. So you had so many trust issues and really, partnering with you and what we're seeing within Legion, when you're literally showing us all the detail and that transparency, and that is building that trust for the store manager. And with AI, it's continuing to learn and get smarter. It's not just a very specific algorithm that doesn't change. And so it's having that really transparency, it's making a difference. And that's a great point. And I'm glad you added to that answer because one of the things about AI, which is very different than even smart algorithms in the past is the ability to learn and the ability to learn continuously. So in other words, if there is real well built AI in your workforce management or other enterprise products, that product should get better without changes to configurations, without changes, without doing additional implementation or configuration or integration work, just getting more data and learning from the usage of the products. In this case, as Karen mentioned, when managers edit the schedules, when managers are looking to modify certain things, all those things are inputs and those things will continuously be evaluated to make a determination of whether there are better opportunities to optimize the schedule. So that's very important. That is something that you can only expect from an AI powered platform. Moving on to, so we talked about efficiency. We talked about a good use case of AI that's very practical today is helping with efficiency productivity. We talked about how transparency is important and so on. So assuming a lot of these things are in place, I'm curious how you are directing this. There's a time that is saved for managers, for example. Where is that time going in the new world? So maybe Karen, I'll start with you. How is shifting administrative work from AI to other things? Actually, what are these other things and how is that changing outcomes for the stores? That shift of time is really going back to that store manager to work with their store associates, like really making sure the store is ready for the guests and being open for the guests and really reflecting that, Hey, I'm not having to work so much on this administrative stuff, and I actually can look at how something is set on the floor. Did we really set that up correctly? Is it welcoming for the guests to now be able to shop? So you're getting more focus where we really want to be. And recent in our rollout, so we're continuing to look at how are we getting that true adoption and where do we start to see even those savings of time and how to measure that. And then being able to make sure that we're, again, focused back on the guests and focused on our store associates, the coaching, the setting of the capability of shopping in the store. And by the way, the great thing about all the things you mentioned, these are the things that AI cannot do. It is the things that you need people to focus on and spend their attention and really look at the details with the guest experience and the, the shopping and all those things. So that's awesome. I know for a lot of you, it comes down to examples of specific things. Where is AI being deployed today? So let's talk about that. And Ryan, I'm gonna start with you. Can you share a specific example where an AI driven automation has created a measurable result for your organization? Yeah, actually it's a pretty easy one. We were curious about exactly this going into our Legion rollout a number of years ago. And so working during implementation, we actually just sent out a survey to our store managers. We had well over a 100 responses and we simply asked them as part of the survey, how many minutes per week on average are you spending creating a schedule? The average answer we got was forty seven minutes. Certainly some variation in there, somewhere a little bit better at that Sudoku puzzle than others. And then after just two months after launching Legion's AI generated schedules, we asked the same survey and the average dropped to 17 a week, which for anyone doing quick math, 66% reduction in time spent on scheduling every week. Amazing, that's awesome. Great to hear. Seventeen minutes seems like a lot to me, but we can talk about that. We'll get there. Let's focus on the, with AI, with automation comes change and change is always hard for organizations, particularly where there's been years, sometimes decades certain prior practices or standard operating procedures that need to be changed. So I'm curious to understand the biggest barriers to adoption of automation and AI and what's your wisdom in trying to overcome these barriers? And Steve, I'll start with you. Yeah, I think we're still in the early stages of adoption across IKEA, but I think one of the things that we're trying to do is change the language that we use when we talk about AI. AI can be quite threatening. As a statement, it can be quite frightening for people to to think about that there's a lot of myths around it and we felt the need to really sort of like explain the benefits of rather than exactly the details behind the scenes. So we talk more about invisible AI. We changed the language to more about enhancing the tools. We talk about getting data in one place. We talk about platforms that then will give more accuracy, more information flow. This helps with adoption because no one is then really trying to work out whether what we're bringing in is gonna take over their job. And obviously when you're a frontline retailer, as you say, these are things that AI can't do but it doesn't stop your brain going in that direction thinking that it might. So we're trying to change the language to really give coworkers more control over the narrative that's being put out there. Make them part of the decision making. The transparency then gives them the ability to understand that there is no threat. There's just more enhanced availability of information and data and things that they can then quickly grab hold of to work with the customer much better and much faster. And I think the one thing that that does, we found that that does is create more trust. And yes, we can talk about AI centrally in our organization within our group digital environment. We need to discuss it but on the frontline we're changing the language to be more acceptable. And that also, I have my head of change management in the front row here and she will agree with me that it helps the coworkers be part of the change that's coming in. They see the benefit rather than the change itself. They understand the value that's coming with the new technology and they understand and want to be part of it. I think that's absolutely crucial for adoption across the board. So we're at the early stages, but we're seeing some really, really positive signs. I think the point you made, Steve, about AI being kind of intimidating as a term is something us technologists really need to take to heart because we love putting the label AI if it's AI because it is a technology that powers our applications. But from a user standpoint, maybe that is not the best way to promote something. So that's something for us to think about. Yeah, and there are some people who want the language and some people who want to talk about it and you can do that with them. When you get, if you have a 110 year old business or an 83 year old business, which is actually has a lot of culture and branding and values that have been there for many, many years, AI becomes a stamp on top of that, that says this is changing us to something we're not, but actually it shouldn't be that. It's just enhances that value and the ability to focus on the things that those values really create, which is that interaction with your customer, the experience that they have, and also how to develop that relationship with your coworker. And I just, I agree with you. It's like, how do you, it just amplifies, it just makes things better. And how do we get that kind of language out there while you're using it? And when you're putting it together, how to make sure that they understand I am not replacing you. I want to give you time back here. I got your schedules down to seventeen minutes. They're more accurate and you have more availability for people to be there with you from a store associate standpoint to service our guests and really not feeling like, oh, this is used with AI. So I agree with you. And would you like to add something to, especially you all have been live for a number of years now. And so how has adoption trended over the years and also how the implication of, is it even thought about as AI at this point or it's basically part of your, just the operating practice? Yeah. I mean, one answer to that could very much just be, it's such a useful tool. I think oftentimes that it being AI kind of fades into the background and it's just a really useful tool. At the same time though, I think there's a conversation to be had around AI adoption in retail holistically. And as much as we all here at NRF, two years ago, was just AI on everything and there was no practical use case. And I think the energizing part of this show, this year in particular is that no one's talking about AI without a practical use case. And that feels very natural to all of us. In fact, I'd ask the room, anyone in this room who does not use an LLM as part of your daily or weekly routine? No hands, one maybe. I think if this was a room full of retail frontline people, there would be a lot of hands up to that question. And I think there would be hands up for people willing to admit, I've never tried one. And I think it's up to us as technology leaders to make sure that the business, that operations is comfortable that the leaders have played with this tool and what it's capable of. Even just go to the free version of ChatGPT or Gemini. And that's where the leadership skills that these leaders are really good at. Those interpersonal skills that AI can't replace can come out. Because then they can go to individual store managers and associates in their charge and knowing the things they know about them, maybe they have an associate that's into health and fitness. Walk them through having an LLM, them a meal plan for the first time and have them adjust it and play with it and just get comfortable around what AI is and what it can do for them. That's excellent advice. And that was very much our design point of view when we created the Legion AI assistance, where we asked ourselves exactly the same question, how many of frontline employees are using or having an experience like ChatGPT and if they could, in what context would that be useful and how would they actually leverage that and that led to the creation of Allegiant AI systems. That's, I'm really glad you shared that perspective. For retailers who are still hesitant about AI, there is a number of reasons, good reasons why you would have some concerns about AI, like everything that type of things we hear and see about like this hallucinations where answers are not accurate or best practices in data management, right? So is your data being used in the right way or the wrong way? So there are definitely reasons why you would be concerned about AI, but for retailers still hesitant, what's your recommendations as it, how should they go about evaluating? How should they go about examining whether it is going to add a value or not? I would love to hear that. And I'm gonna start with Karen. Data. So you've heard this, you've heard it from everyone actually that's talking about it, it's data. And I think that's where from a standpoint of leveraging Legion, we were able to have the data all in one place and being able to access it and so that we could leverage that and feel actually getting verification of the information and making sure the data is accurate. And that's very important. And when you have 110 plus year old company, you've got data everywhere, right? And so we're continuing to work through how do we get our data set up in a way that can be leveraged in all the possibilities and feel comfortable and with the right compliance and the right controls in place that it's giving accurate information. And so I feel really great about how we've partnered with you and knowing that the data is contained in a place that provides those accurate and transparent answers. That's great to hear. Ryan, from governance standpoint, so Karen mentioned data, there's also the other side of the same coin as governance. How has your experience been and what's your recommendation for retailers? I think the biggest thing is find ways that are easily approachable and that data staging is really, really important. AI is only as good as the information it has. So I would say certainly pick one and go slow and choose the right partner. That's great. I heard the go slow part. So, just looking ahead, right? Mean, there's one of the exciting things, but also sometimes it makes me nervous how fast AI is moving and how fast things are changing, how fast new possibilities are emerging. Things that used to be on the horizon now is like next quarter. I mean, you see that and you hear about that all the time. When I ask each one of you to think about what are the new AI capabilities that you are most excited about or things that you think are going to be real soon or as a possibility is emerging, which excites you and especially something that is relevant to your frontline teams? And I'll start with you, Ryan, since we were just talking about this. Yeah, voice assistants. I think that they have improved so much and I'm so excited imagining the possibility of a store manager being able to talk to an AI voice assistant and simply say, I had a call out and I need you to fill the shift for me or make those recommendations on the fly very, very quickly. It feels like just yesterday, I think most AI voice assistants in my opinion were kinda garbage. Like if I breathed wrong, it thought I was interrupting it. And it's incredible how much it's improved in the last six months. Ditto on that one. I agree with you. It's an app you're carrying around. I don't wanna key in things anymore. I don't need to key it in. I just wanna talk about it and use the data I have and be able to take that action. Yeah, I'd like to come up with a different thing, but actually I think this is exactly the thing that we need to look at is voice activated interactions is to be able to be with a customer face to face, get question asked and then, you know, either maybe the future is AI glasses where the answer is coming up in real time when you're speaking to them, but at the moment to have, to be able to ask your earpiece for the information, where is the stock? What is our value? When can we do deliveries? Is this available? All of these sorts of things, which can take an inordinate amount of time if you don't know where that information is and it's time wasted with our customers who do not want to have their time wasted. So I think that is really the next stage for us definitely. Just following up on that, as you see voice assistants being more pervasive and firstly working effectively and then being more pervasive, do you see the role of the store manager evolving And maybe Karen, I'll get your perspective on that. I actually want, I wanna go back on the voice assistant. There's a place where the voice assistant, because then it starts making me go into the cyber thing of, okay, what about the deep fakes? And what if something's prompting in a way that potentially could cause more chaos? So there's definitely something that we need to control there as well. And you don't want it to where it lacks that control. So we do need to watch that as we're going down that path. Absolutely, in fact, I mean, that's one of the areas of, probably one of the biggest areas of concern with AI and how easy it is to impersonate people. So that definitely with the technology comes those pitfalls that you'll have to watch out for. I'm gonna end this with one last question for all of you. And this would be, there are many folks here in the room who would love to hear your experience. You've shared your experience. You've been very generous with your time here, But if you're to leave them with some advice, what separates retailers who have successfully leveraged AI from those who fall behind and you have examples and you have now proven that you are able to do that successfully. So what would be that advice? And I'll start with you, Steve. Yeah, I would say integrate it into your company strategically, both centrally and frontline, make sure it's part of your way of working, make sure it is a change in the way that you embedded into your organization. And I think the one thing we can't ignore, we've mentioned several times, invest in clean data, in validating that data because as Ryan said, it's only as good as the information you feed it and it is so essential that we feed the right information in. Prioritize the customer and the coworker experience as well. I think this is quite important is that sometimes we just focus on the customer, but actually it has to be a good experience for the coworker as well. Your employees need to feel that they're part of the process of selling, but also the service that they're giving you as well. And remain agile. You've got to remain agile. You've got to be able to adapt to changes in the market. This is moving so quickly. You just can't just sit back on that. And we've been mentioned before about remaining transparent. You need to have an ethical values driven mentality around AI. You need to keep your mindset on really what that means to your coworkers as well. That's great. Ryan? I'm gonna say it's about balance and guardrails. I think what's really interesting is thinking about retail three or four years from now. I think we're going to see some companies that are struggling and it's gonna be on two sides of a different coin, which is really interesting with AI. I think we're gonna see some companies that rely too heavily on AI to make, customer facing impacting decisions and could have reputation damage. On the flip side of that, I think there's gonna be companies that, don't get involved with AI and fall behind quickly. And so that balance and diligence and keeping guardrails in place in the process, think is really important right now. It's a very interesting perspective of the too much too soon with AI that could lead to reputation damages. Karen? Love both of your advice, all of that advice. And just going deeper on your business strategy should include your AI strategy. Like you said, it's not just an AI strategy. Have an AI approach, have a framework, have a governance, have the compliance on it. But the AI strategy should be embedded in your business strategy, which then hopefully can give you that balance that you're looking for and that you don't go too far too fast on just one tunnel of, oh, I need to get this piece of AI up, but I totally let my store associates not even have capabilities that I could have been giving them. So really trying to make sure that you are looking at that whole picture with it and bringing in the most impactful areas because those areas will then translate into efficiencies if you're leveraging it. Yeah. That's great. What I would say as well is that we've discussed all the things that you should do. One thing I would say is what you shouldn't do. Don't treat AI as a one off IT upgrade. That's not what it is because you'll suddenly start to see benefits and then you'll lose it very quickly because, and you need to build the infrastructure in your company around it. That's not just the technical infrastructure, but the organizational infrastructure. If you suddenly start to realize benefits and values from implementation of AI, That means your employees are getting efficient, your technology is getting efficient. What are you doing with that efficiency? What are you doing with what your gain is? You need to be organizing a structure around it. It could be that actually you want your people to have that time and you don't want to fill anything with it, but you also want to be in a situation where you want to leverage it if you can as well to make sure that you've got competitive advantage. So there's a lot of things that you can do, but there's also a few things that you definitely shouldn't do as well. That's wonderful advice and we are out of time here. I'm just gonna make a quick attempt to summarize a lot of the things that we talked about. So good luck to me on that one. The three points that really resonated with me, first is AI is not here to replace people, especially not their judgment. AI is the most effective usage of AI that we heard from our panelists today was around areas where productivity and efficiency can be gained and it gives back time to the managers, gives back time, makes them more productive and actually enhances their superpowers. The second is, think, don't chase AI for technology sake, the pitfalls, there is the balance part, the going too fast, not making it an IT project. It is all about being thoughtful with how AI is deployed, how AI is thought about as part of the strategy. I think that's super important. And last but not the least is AI works best when it's transparent, when it's explainable and when it keeps humans in the loop. And with that, I would really like to thank everybody here for spending time with us, for being amazing customers for Legion. I really appreciate your partnership. Thank you. Thank you.