The Trick to Aligning Tech, People & Process for Operational Success

Speaker A: I feel like our future in five years is going to be a lot more predictive in the first few months.

Speaker B: That's when we saw that, uh, 26% improvement in technician response time. We're looking into how can we use AI to leverage that data and essentially predict when a machine is going to need service.

Speaker A: Most engineers are playing the game of making things and actually doing the hands on work. And so that transition to leadership can actually be really difficult.

Speaker B: If your customer is calling you, it's already negative and this is your opportunity to make it a positive.

Speaker A: The double entry problem is something that I think every company suffers from that.

Speaker B: When we actually went live, it was a very successful launch. Since we went live with a new system, we've seen a first time fix rate above 99%. So it's been a huge win. The system is running through billions if not trillions of different permutations when it's working. Wow.

Speaker A: Wow.

Speaker B: Uh, this wasn't about replacing people. This is about helping everybody do their job. Everybody wants to do a good job.

Speaker A: Welcome back to Experts of Experience. I'm your host, Lacey. Peace. I just got off mike with Phil Parbury, who is the core services manager at Tamra Collection Australia. That's right. We had an Aussie on the pod, which was really cool. I am currently recording at 9:35pm My time to account for the time zone difference. This episode was a really amazing under the hood look at how operations affect customer experience. So for those who aren't familiar, Tomra Collection in Australia has a bunch of reverse vending machines, which I actually didn't even know was a term before talking to Phil, where you return items, uh, through the vending machines and then of course they take them m and distribute them. So a good example of this might be like plastic bottles that you turn into the vending machine. You get some money back for turning them in and then, uh, Tom Ra will take it and dispose of those plastic bottles. Right. Um, so it's a really unique and very niche business model. Um, but what I loved about this episode was that we got to talk to Phil about how his field service team is using technology in new ways. So that way they can increase what he calls uptime, which basically means that all these machines are operating. Um, and obviously if your machine is working and there are no problems, then you are not getting those customer support phone calls that say, hey, this machine is not working. I'm frustrated, I'm annoyed. Um, so it's really important that these operations are very buttoned up that the Response time is super high that these machines are operating to their fullest capacity as often as possible to avoid the negative customer experience. So it's a great masterclass in navigating the digital transformation. But we also talked a lot about what's in store for the future with these new technologies. So we talked about predictive maintenance, predictive analytics, personalization, when personalization is too creepy or when it's really helpful. And how to navigate this new world where your human employees and human agents are going to be actually working with AI tools and AI agents. The right way to do it, the wrong way to do it. You won't want to miss this episode with Phil. Plus, did I mention he's Australian? I mean, that just makes it even more cool. So tune in to this episode. But before we do, please like and subscribe to our YouTube channel, to Apple podcast, to Spotify, to wherever you're listening. And please drop me a line DM me on LinkedIn and let me know what you who you want to hear from, what company should come on the show, what questions you have that I have not yet asked. Um, and I would be more than happy to answer those questions and invite those folks onto the show. So without further ado, here is Phil Parbury of Tamra Collection Australia. Phil, welcome to Experts of Experience. I'm so excited for you to be here.

Speaker B: Thank you for having me.

Speaker A: Yeah. And you have to humor me. You have to humor our audience. Most of our audience is in the US So I would love for you to explain where you're calling in from.

Speaker B: So I'm based in Sydney, Australia, so other side of the Pacific. It's actually a day ahead for us as well.

Speaker A: Yeah. So what time is it for you? So for everyone listening, I'm currently recording at 8:30pm this is my latest interview I've ever done. So what time is it for you, Phil?

Speaker B: It's 11, uh, 30am the next day.

Speaker A: Yeah. I don't know why that always blows my mind, but every time I'm like, man, it's so cool and wild and interesting. Um, before we get into what you're doing at Tomra, I do want to kind of set the table there since you are in a different location. Most of our audience is US based. Is there any differences that you've seen with Australian culture from like a buyer or customer perspective that you think is interesting and worth kind of flagging for our audience? Before we dive in to what you do today?

Speaker B: We get a lot of influence in the American culture over in Australia as well. I think generally, I think a lot of the patterns can be very similar. Uh, from my experience, I haven't spent a huge amount of time in the U.S. but people like to have options. They like to have uh, high level of service and feel like that they're a valued customer and not just a number essentially.

Speaker A: Yeah, absolutely. And doing it all while there's just insane amounts of different animals and insects trying to get you

Speaker B: well. It's a beautiful country but full of lots of wonderful wildlife.

Speaker A: Yes, I have to visit. It is on the top of my list. We will be going there at some point. Um, cool. Well with that said, I would love to do an intro to TOMRA and what you guys are doing since I don't know if it's a company that many people have heard of. So if you could explain what it is and kind of what you do today day to day, would that be great?

Speaker B: Sure, yeah. So the TOMRA is a global company. Um, we have several divisions. I work for Tomra Collection Australia and what uh, we do in Australia is we make, operate and maintain reverse vending machines. So reverse vending, that is you return the empty container and get some money out as opposed to a traditional vending machine. We operate these as part of container deposit schemes across the country. We have about 2,600 machines across maybe 750 sites nationwide. And most of those are automated on staff machines. So we haven't got staff on site, the customers just working with automated machine. And my role is head of what we call the core service team. Uh, it's a cross functional team. We support our customers in different states across the country and my responsibilities include leading our supply chain, uh, our technical team and our customer service team. So it's quite a, quite a big role. It's one.

Speaker A: I was going to say it's quite enjoy. Yeah, yeah. And how long have you been there?

Speaker B: I've been at TOMRA a bit over seven years now and uh, taken on several different responsibilities in that time.

Speaker A: Yeah. That's amazing. What were you doing before you started there?

Speaker B: I originally trained as an engineer. So I was, I was a mechanical research engineer at a manufacturing company in Australia. I've also worked in aerospace in the uh, UK as well, so.

Speaker A: Wow. Okay, wow. So what was the drive to kind of transition out of that for you?

Speaker B: Well, I joined TOMRA initially as an engineer as well. So it's more of a transition to operational and service management over time within the company.

Speaker A: Yeah. Oh, I love that, I love that. So I used to go to school for engineering before I went completely different direction towards creative work. So I always find it interesting whenever I get to hear like people have different backgrounds before they ended up in these operational roles that you know, kind of result in managing people. Most engineers are playing the game of like making things and actually doing the hands on work. And so like that transition to leadership can actually be really difficult. Um, so it's very cool whenever someone has been able to do that well. Um, and with that said, I know when we last chatted we had talked about this transition that you saw, was it three to four years ago, um, whenever you were kind of stepping into this new role, um, could you just kind of set the table of what challenge and challenges and pain points you were initially seeing the first couple years that you were at Tamra and what you guys were kind of thinking about three to four years ago, um, in terms of what you needed to change or to tackle.

Speaker B: When I joined Tomra Collection Australia, it was very much at the beginning. The company had only been operating for maybe four months at the time. Uh, and we were in that startup phase, that rapid growth phase. So uh, over those first couple of years or so we were launching lots of new sites, moving into new markets and uh, we didn't really have the time to set up uh, our uh, support and service and operations in the best possible way. So uh, it's very much, let's just grab whatever kind of works and fit it in. Think like square peg into round hole sort of thing. We'll make it work and we'll sort it out as we go. We'll build the plane as we're flying it, as it were. Uh, quite a favorite expression we have around here.

Speaker A: No, I mean that's like every startup though, it's just like band aiding things together as we start working. So it's always interesting too even when you have a global company that um, it is still like a startup. Whenever you're starting out in a new country, you still have to think really scrappy, you have to get really creative with what you're doing and that can be a great way to propel you initially. But yeah, as you guys I'm sure saw there, there were gaps as you did that.

Speaker B: Yeah, definitely. And uh, yeah, and then that's, I said a few years ago we kind of took a bit of a step back. We had a little bit of breathing space like right, okay, let's, we've got a lot more growth coming.

Speaker A: Mhm.

Speaker B: Let's pretend we were starting again. How would we do this? Properly this time. So I initiated a project where we just did a full operations wide systems assessment. What are all the tools we have, how are we using them and um, what do we not need and what are our pain points our teams are having?

Speaker A: What kind of came from that? As you started to uncover that we

Speaker B: had lots of these different systems, I think we worked out about 25 or 26 different tools in the end that we were using to manage the network. And uh, there's a lot of double entry, a lot of information being duplicated in different places but then contradicting itself. So it was quite a challenge for our frontline teams to deliver that uh, high quality service that we wanted to deliver when they're having to search five or six different tools to find the information they need and to create a work order to get something out to site and fixed. So yeah, they really just said taking that step back and, and sign with a blank piece of paper.

Speaker A: Yep.

Speaker B: To draw out the kind of Golden State we wanted to get to drawing that Golden State was a really effective way of setting up. This is the destination, this is where we are. And that helped us map a journey to get to that point.

Speaker A: Yeah. And when you were putting together this map, were you guys looking just like a year in advance, years later, in advance, like how far are you sort of planning as you were, you guys were kind of tearing this apart and rebuilding?

Speaker B: We were tying it into our overall organizational strategy as well. You know we have like many companies, we have a five year plan, uh, that we review every year. And, and what, what are we going to do in those next five years and how are we going to keep our uh, operations working in the best way to support that continued growth? Uh, so in terms of rebuilding the systems, we had it, you know, it was a two year process and it's a process I don't think we'll ever finish. It's, it's something one of my managers is. We're continuously looking to do this better, continuously improve and we're always learning new things as we go. And it's not like we can just set this up and then forget it again. We have to continuously evolve and continuously refine what we're doing to again deliver uh, that really high quality service.

Speaker A: Did your leadership team, they were completely on board with this and they were like, yep, I've noticed these challenges as well. We're willing to bet with you and work with you on this or was there any kind of, I don't know, advice you might have for other people who are in your position, who are like, hey, really got to work on this thing, um, maybe you have some advice or lessons learned as you were working with both your leadership team, but also the team that you're managing and trying to convince them and get buy in from them on, hey, this is actually going to work now, but this will actually result in something that's better for us in the long term.

Speaker B: It was very much, let's get everybody involved in the journey from the start. It wasn't just people I managed, it was people in other teams, other frontline teams, leadership team, as you said, getting everyone involved. Even from the beginning of assessing what we have and what we're doing now, giving them an opportunity to voice their uh, concerns, their pain points, but also what they've used in previous companies, you know, what do they think we can do differently?

Speaker A: That's great.

Speaker B: And kind of collecting all that, uh, into one essentially scoping document and starting the vendor selection based off that central document as to, huh, where we're going to go. And our leadership team well informed what we're doing and uh, when, when the time came to us for the money, they were already 80% of the way there.

Speaker A: So it was, this wasn't a surprise. They were like, we knew this is what we've been working on. Yeah, I love that. That's great. Yeah. And when you talked about like kind of scoping outlet systems, you guys were looking at or looking at vendors, was there key things you were looking for or how did you kind of make that decision? Because I don't know, when I look at the tech space now, there are so many options that it's very overwhelming. Were you just kind of listening to your peers and hearing what they were using and what maybe people in your company had already had experience with? Or was there a different way that you were sort of looking at and evaluating who or what you wanted to work with?

Speaker B: It was a combination of things. It started out with, I said, you know what, what do we want? What is our, what are our goals? I set three goals that we're trying to achieve with this transformation. Eliminate double entry or a single source of truth. So all the information in the right place and some a system that's going to be scalable and sustainable to grow with us. We, as I said, involving different people. What was used in the past, uh, and also looking at external publications. What are other people using? Yeah, we went through quite a, quite a significantly long list of different vendors before we narrowed it down to a short list that we worked with, um, a lot closer.

Speaker A: Those three goals, you know, you're able to communicate them really well and like articulate them. Pretty short, but they're massive goals to tackle. Like, the double entry problem is something that I think every company suffers from single source of truth. Like, oh my gosh, how difficult that is. I mean, just those three goals, it's funny because they just sound kind of simple, but they're so hard underneath to actually tackle that and solve that. Um, so, yeah, I just think that is interesting that you're just kind of nonchalantly sharing those. But those are very, very difficult solutions to solve. Yeah, I mean, so I kind of want to dive into lessons from that transformation journey. I want to hear how it's playing out now. But are there any key lessons or advice that you might have for other people that are sort of in that, that transformation phase currently? Because you guys were doing it a few years ago, but there are plenty of companies that are making that investment and decision now. What do you wish you knew when you were doing that?

Speaker B: I think, uh, keeping it grounded to the, to the core goals. What are the core goals that we're trying to achieve? Uh, because it can be said there's lots and lots of different options out the tech space and it can be quite easy to get distracted by, uh, you know, the shiny new thing here over here and the, and the really awesome feature over there. But it's like, okay, but bring it back to what are we trying to achieve, right? And we don't have to do everything all at once. It's like by mapping out that journey, you know, we don't have to go from 0 to 100 immediately. We can, we can step along the way and realize those gains at each step rather than trying to jump straight to the finish line. I think that was, yeah, it was a really, uh, a really key learning for us.

Speaker A: And so what did that rollout look like then? So you've gone through the process of blank piece of paper. We're just going to lose all of our assumptions. Anything that we built, we're just going to pretend like it doesn't exist. What does it look like in a perfect scenario? Okay, what tools can help us accomplish that? Maybe it's something we already have. Maybe it's something we need to invest in. Okay, now we've agreed on this plan. We've gotten buy in from leadership. We got the check, we've got the money ready to go to do this investment. Right. Then what you said it took two years to roll out the technology so what's happening in those two years?

Speaker B: It took about six months to assess where we were to build that proper scoping document, about six months vendor selection and then another six months to get all the approvals before we actually started the implementation. So there's a lot of whilst, ah, we had local, local approval relatively quickly, the level of investment we were trying to make required high level approvals to get global company. There's always different levels of authority what we were proposing, very different to what a lot of our sister companies were doing in other parts of the world. So there was a lot of work involved in getting that buy in. We got there and uh, so a lot of work during that kind of approval time. A lot of work was spent planning implementation, making sure uh, everyone knew what their role and responsibility was during the implementation process. And then there was only four months of actual implementation before we went live with the new platform.

Speaker A: Oh wow, that's actually really, really fast.

Speaker B: It was uh, quite a quick implementation. We were against a hard deadline ourselves as we were preparing to launch in another state. And it wasn't like we can't do this at the same time as the launch. We have to be ready for that makes sense. So it was a case of this is the deadline, there is no pushing this back. Allowing lots of time for users to test and to learn and get involved in the development of the system meant that when we actually went live it was a very successful launch.

Speaker A: So I want to paint a picture for our listeners of kind of, we're talking about, you know, you've developed and launched the system but like what does the system look like? You know, what are you talking about when you're saying there's this new system that has been launched? So I'm wondering if maybe you could paint a picture of um, like before we had 26 different systems and now we have, you know, this one system or two systems and uh, the day to day customer rep experience looks like this or the field service rep looks like this. Uh, do you mind just kind of sharing this compare and contrast for us? It used to look at this and now with what we've implemented we're seeing this.

Speaker B: Sure, yeah. I want to break it down more or less into two general groups. So we have our kind of office based customer service team and our frontline field teams. The office based customer service team, they previously had to use six different systems to create one work order because that involved checking is there already someone going to the site? Is there already something else happening? Now? It's just One, it's, it's our Salesforce platform that we, that we're using now for the field teams. It's like I actually have the information on my mobile device or tablet and the information on my device as to, you know, what, what has, what has the other technician done at this site last week? You know, it's that sort of level of information that they, they have at uh, their fingertips to, to then better diagnose the problem on site and do a longer lasting solution.

Speaker A: And so previously were they having to like call up a technician, be like, hey, I saw you were working on this, what did you do? Or now it's just all kind of in this iPad system that they can look up on their own.

Speaker B: It was a case of yep. They might call up uh, somebody else say what happened here? Or they have to scroll back through other jobs and struggling to find it. Or they just ended up doing the same thing the previous technician had done a week ago. Yeah, and the problem wasn't really solved. It looked like it was solved and it probably did hold for a couple days but it wasn't really getting to the root cause. So there's always little things like that really. And um, since we went live with the new system as well, we've seen a first time fixed rate above 99% so it's been a huge win.

Speaker A: What is kind of driving that? Because on the face of it I hear like, okay, a new system has been developed if someone needs to go back out and fix something. Now we know what was going on previously. But you're saying even initially that first time fix is happening at a higher rate. So what do you think is driving that? Is it just better communication between, you know, the dispatch and the person on the ground?

Speaker B: There's definitely an element of that. There's also element of um, having access to the more information at the fingertips as well for the field team.

Speaker A: Yeah. Okay.

Speaker B: In um, terms of. Yeah, we've got things like user guides in the app. So if I send you for a job about a printer, some information about the printers, that sort of thing. So that's all automatic. Uh, and yeah, that helps especially for the technicians might be doing a job that they haven't done very often or of less experience with.

Speaker A: Has that helped with training new techs as well?

Speaker B: Uh, I think so. We haven't adjusted our technician training approach with the new system, but it does help with the follow ups. Um, and technician, when they're on site, if they need help, they can call up the technical support team. But Then again, they also have those user guides available to them as well, and some of them use it more than others. That's always the way it is with dlt, for sure.

Speaker A: Yeah. Oh, yeah. Um, that's actually kind of a good segue into a question that I was thinking about with this is as much as we love new tech, especially from a leadership perspective, we're like, look, you only need one system now. You don't need these six other ones. People can be really like, oh, but those six other systems. Systems are my babies, and I don't want to give them up. And, like, I'm really used to using them. So was there sort of an element of getting the team comfortable with this new technology, um, as you guys kind of rolled it out, or was it just. It sounds like it was a. Maybe a long enough process that people had time to get used to the idea before it was in their hands.

Speaker B: During the implementation phase, we made sure there were representatives from all the major teams impacted in the development phase, even before the testing phase. So it was like they kind of do, like, mini testing as they went. So our implementation partner, uh, they'd build something and say, hey, what do you think of this? Try this out. And then there might be some feedback, make some modifications. And then, then we did a formal user acceptance testing phase again, where they had a chance to use it as if they were doing it for real. And, and again, provide any feedback or input. Uh, but even after we went live, there was. There was always going to be more lessons learned. There's always things you missed during testing. And, uh, we've made updates as we've gone along. Yeah, we've overhauled the field service app again since we went live with it to, again, better streamline that user experience for the field teams.

Speaker A: Yeah. Um, I think what's funny is that we actually haven't. I don't think we mentioned the word or the phrase AI yet. I don't think we've touched that. Even though I'm automatically already assuming part of the system update is AI, uh, do you mind sharing if and if any of the system that you guys have invented is, you know, writing off of some AI tech and how you're using AI with your teams now?

Speaker B: Definitely the main M use case we have is with dispatching. Uh, you know, the system is running through, you know, billions, if not trillions of different permutations when it's working out what's the most optimum allocation of jobs to different technicians in the field. If you had 10 work orders and one technician. There are over 3 million different ways you could arrange those work orders.

Speaker A: Wow.

Speaker B: Yeah. So our dispatchers are very good, but there's always. And they'll probably get close to the best one, but there can be a better way. And then when you add in multiple technicians, those different numbers grow significantly. And then we have new work orders coming in during the day. Again makes it, uh, you got to almost restart the process. So having that AI involved in the dispatching and help present, uh, this is an optimum way of dispatching your technicians to get your best results. And our best results are what's going to maximize uptime. And since we've fired that up. Yeah, we've seen a 26% improvement in technician response time, which means machines are getting attended to quicker.

Speaker A: Uh, you said the term uptime, which I'm sure some people who are listening know what that means, but do you just mind defining what you. How you guys would define uptime?

Speaker B: Uptime is the time a machine's available for customer. Use it. For us, that's our most important KPI as an operations team. Because if the machine's not working, it can't generate any revenue. And also it will just annoy the customer when they turn up to use it and find it's not working.

Speaker A: Yeah. Yep. Sometimes scaling means giving your people the backup they need to crush the big stuff. That's where AgentForce comes in. AgentForce gives your team a smart digital workforce that actually gets work done. These AI agents answer customers, route requests, and hand off complex issues to humans seamlessly, all with the full context of your Salesforce CRM. They're not guessing, they're not hallucinating. They're acting on your data and the rules that you set. The result? Faster service, less chaos. And your team gets to focus on work that truly moves the needle. All momentum. No burnout. Visit salesforce.com agentforce Tell me a little bit more about that. So how are you guys getting notified that, hey, these machines are offline? Is there a system inside the machine that lets you know and you guys can get dispatched that way automatically? Or is it customers like, hey, I can't use this. Please come fix it?

Speaker B: Short answer is both all the machines are online and part, uh, of our, uh, customer service team's role is to monitor those machines in real time throughout the day. They're all connected to our Central system called TomraConnect. And they can see all, uh, right. This machine is not working. We can try and fix it remotely, or if not, we'll get Someone out there to sort it out. But then of course, the customer may also call in and say, hey, this machine's not working. What's happening? And then that feedback to them that we've got someone on the way, or give us five minutes, see, we can sort it out for you. Or there's another machine around the corner. Uh, you can try that one. We know that one's working. I can see it's working right now, sort of thing.

Speaker A: Yeah, it's honestly really impressive technology to just be able to remotely restart that. I mean, I'm just thinking even a couple years ago you would have to physically go do that with all these machines. So to have the tech in place that you can do this all remotely is very impressive to me. Um, and I can only imagine how much better it's all going to get as AI continues to improve what it can predict and um, what it can do with dispatching. I'm really excited for what the future holds there. Um, but before we get too far into future trends, which I will definitely be asking you about, um, I wanted to talk more about results. You said 26% faster resolve of issue. Right. If there is a machine that's broken, um, is there any other sort of metrics or results that you guys have seen? It's been, what, about two years now since this has been implemented, or a little bit less than that?

Speaker B: Uh, just under two years since we went live. Um, continuously made, uh, upgrades as we've gone along as well. So for example, the dispatching, uh, assistant is not something we turned on straight away. Uh, we had to. With any system, it's only going to work as well as the information it's got as the data is being fed. And uh, before we decided to fire that part up, we wanted to make sure that everyone was comfortable with, uh, essentially inputting the data into the system and using the system. Once we were at that stage where we felt comfortable, we essentially activated the AI dispatching assistant, or optimizer as we call it. And the first few months, that's when we saw that 26% improvement in technician response time, which is really powerful to us.

Speaker A: Have you seen any other results or like success metrics with what you guys have rolled out so far?

Speaker B: Yeah, so we're definitely seeing improvement in uptime to go along with the response time, uh, which is, which is really good. So linking those two together, we've seen a, uh, reduction in cleaner work orders as part of, part, uh, of operation. We use contract cleaners to go out and um, clean the Machines? Yeah, they're collecting empty uh, drinks containers. They can be dirty and they can be a mess. It's important that the machines are cleaned regularly. But then you know the machine can go down if it gets too dirty and uh, if we have to send a cleaner on an extra work order we have to pay a premium for that. Ah and as part of um, part of what we've seen since we launched uh, new systems is we've seen a reduction in those um, essentially additional premium work orders. So yeah, obviously very good for the bottom line and goes well with the increased uptime.

Speaker A: Yeah, yeah. And with all of this have you seen any interest from other countries or other parts of your guys organization like actually seeing the results from what you're doing and saying hey we kind of want that because it sounded like when you were initially um, going through this process of rewriting the script there was a little bit of like needing to convince. So I'm wondering if now it's like oh we actually really like what you guys are doing there.

Speaker B: In Australia we have had some interest uh, uh, shared what we're doing with some of our other markets. Uh, the Australian business model in uh Tomra is quite different to many other countries business models. So it doesn't necessarily copy across in the same way and they've got different objectives. But yeah we are seeing some of the other uh, markets have ah, expressed interest although they haven't ah, gone as far down the same journey as we have.

Speaker A: So besides what you guys have already been investing in and doing with AI, um and the systems and technology that you guys have implemented, what are you currently maybe in the process of working to roll out or what are you planning to be rolling out in the near future?

Speaker B: Yeah, of course. I think the next step for us again with the focus on uptime is predictive uh maintenance. Yeah, we can be quite reactive. So we wait for a machine to be down before we action it. Uh, we do have preventative maintenance runs and so forth but it's an area for improvement um, from our end that I've identified. So yeah, we get a huge amount of data from the machines and we're looking into how can we use AI to leverage that data and essentially predict when a machine is going to need service. And then again we can better streamline our field teams and our contractors to deliver that and further improve uptime, which is again better for our customers as well.

Speaker A: Yeah, or predict before it needs maintenance. Right. So just understanding kind of all that. So when you talk about sort of the data that you guys are getting from these machines, Is that just like, how many times someone's been using it or how many pieces of, I don't know, cans or glasses we've sorted? Um, is that the kind of data or is there something else that maybe I'm not hitting on that would help predict if it needs maintenance?

Speaker B: Yeah. So we get, uh, every single container that's, uh, put through the machines as a data point that's fed through, uh, all the different, uh. Yeah, it's a customer session. A had 26 cans and customization, B had 35 or whatever it is and so forth. Uh, but then the machines also report in their status. Uh, there's lots of sensors in the machine. So it's all the raw sensor data as to this sensor seeing this, the sensor seeing that, and then converting those technical data points into actionable insights.

Speaker A: Yeah. And so you guys are actively working on something like that right now, or you're sort of planning on it in the next year?

Speaker B: We're planning on it. We haven't, uh, actively started working on that just yet. We're essentially. We're scoping it out at this stage.

Speaker A: Yeah. Awesome. No, that's really cool. That's exciting. And I think this like, predictive maintenance piece is going to become a lot more common across industries. I know we're just talking about vending machines or I guess reverse vending machines. Um, but, you know, you mentioned printers or, um, like ac. I know here in Texas we all love our AC units. Um, so there's a lot of stuff that I'm like, if you can apply more sensors and can accumulate that data over time, I feel like our future in five years is going to be a lot more predictive and maintenance will be a lot smarter. Like, even with cars, like the ability to really start to see, oh, you need, you actually need your oil change now. And like this, not just roughly around this time frame, you probably need it done. Um, so, yeah, I'm really excited by that for sure. Are there any other industries that you sort of like, look at that you have been impressed by or followed?

Speaker B: I think, uh, it's something I think airlines do very well. Obviously an aircraft can't have a failure mid, uh, use the lessons from that industry is something that I think you can apply to a lot of different industries in terms of actioning, uh, or fixing things before they go wrong. Also, in your car, the last thing you want is, uh, to conquer in the middle of the highway. Then you're stuck. Right. I completely agree with you. I Think that the ongoing Internet of things and connectivity of different systems and devices, uh, can have a lot of benefits in their space and making sure that when the end user is using them that it's reliable. It's always going to work in the way they expect.

Speaker A: I mean, yeah, from the customer experience standpoint, that is 100% like the dream, right? Where I am actually never calling your support team because I'm never having a problem because the machine is never broken and everything's always working. So it's like sort of the dream scenario of like, I don't. Our support team doesn't hear from anyone ever because it's always working. Uh, what about customer behavior in general? Um, are there sort of trends that you're seeing in the space as people start to rely on AI, they get more comfortable with AI, they're getting more comfortable with, um, predictive analytics and things like that. Are there any things around customer behavior that you're tracking closely, uh, that you would want to share with our audience?

Speaker B: Yeah, I think there's a couple of things. I think, uh, one of them is about the personalization. Um, people, uh, might touch on this earlier, that people want that more personal touch. Um, I, uh, think everyone's getting much more comfortable. There's a lot of data out there, uh, but at the same time, not excessive amounts in terms of, you don't want it to look creepy. So that's that personal touch, that personal level of service. Uh, and when it comes to AI, lots of companies are using various levels of autonomous agents as their frontline customer support. And I think that's, uh, how well they work can be a big factor in the customer satisfaction with that company. I think we've all used a, uh, uh, chat bot that it's been like, this is terrible. Just let me speak to somebody. Uh, oh my gosh. Whereas I had an issue, um, uh, with my bank a couple of weeks, uh, ago. And uh, their frontline service was autonomous. It was great. Solve a problem. And, uh, there was a potentially very frustrating, uh, scenario that they were able to sort out for me. And uh, that sort of thing where people trust the AI agent is only when it actually works for them and helps them and they feel like they're still getting the level of service they would get from a, from a human

Speaker A: agent or, or they have the option to get to a human agent if you needed to. Right. Like, it sounds like your bank was able to support you with this autonomous agent, but if you really were like, no, like, I need someone to talk to like, this is something that this can't help with, the ability to quickly get to a human. And they have the context of. I already know that this person has been on the phone for this many minutes. They've already asked these questions, they've already said these things. I have the account pulled up. Like, that level of service is amazing. Um, and I had that experience recently with I'll shout out FedEx, I'll shout out StubHub, because both of these companies, I had, you know, random mishaps. The StubHub tickets I was at forgotten are day late, but I was traveling, so I needed them the day that I needed them. Um, but when I called, they tried to do the automated thing. I was like, no, this is actually a thing. I need to talk to a person. I was immediately able to get to a person they knew who my name, my tickets when I needed them, by what was going on, all that context. And the woman went so far as to say, hey, I know this isn't resolved yet, so I'm going to call you back at this number in a couple hours to check in on you. Is this time frame okay? Such good service. And that was all because I'm assuming they have, you know, autonomous agents kind of working on the back end for them to support the agent whenever a human did. Like, I did need a human in that instance. Um, so, you know, random rant there on, like, my own experience. But I think that there is, you know, this huge opportunity for these AI agents to be supportive when in certain circumstances where I get sort of upset is whenever people are pointing fingers and saying, oh, that was just an automated system. Sorry. So, like, I've had a recent call with a customer service agent where they're like, our automated system flagged it. I don't know why, I don't know how to fix it. But try again in a couple hours. Like that, to me, is okay. AI is not working properly for that company. Yes, like that. Yeah, so. So, yeah, I think people will get a lot more comfortable with it. And I think the companies that are doing it well are just going to continue to really stand out. And the ones that aren't and can't move quick enough and can't figure out how to get it to work with humans, they're going to really start to suffer, and people are going to start to not to choose to work with them. And that's my side rant, Phil.

Speaker B: No, I think I agree. And I think, uh, a big part of setting up, uh, your company to deliver that is You've got to train your human agents to work with your autonomous agents and make sure that they understand how it works and how it's supposed to work. Ah. And also empower them to take decisions when it doesn't work. What we see as well is if our customer's got a problem, uh, they're usually much happier if it's solved quickly. Uh, like I said, don't call me back in two hours. That's just not a good experience. And typically, if your customer is calling you, uh, it's already negative. And this is your opportunity to make it a positive. Uh, and you don't want to let that opportunity go to waste. And a big part of setting up your, uh, support center and your support teams around that is making sure that they can have, they do get all the information. You're not asking the customer to repeat themselves or even repeating what they already told the AI.

Speaker A: Oh, my gosh. I know.

Speaker B: Yeah, yeah, yeah. And as I said, just be able to say, look, we'll take a decision. We'll sort it out for you. Right now, it's setting up your people and your systems to work together to deliver the best outcome for your customers.

Speaker A: Yeah. And to me, you know, you said personalization. That's personalization. Like, I get on the phone, the AI agent knows who I am already because they know my number. Um, you know, then I need to be transferred to a human. The human already knows. Like, that to me, is not creepy personalization. That's helpful personalization now, like the text message from some random spam message that knows my name and all this other information about me. No, that's creepy. And I don't want that personalization. So, like, I definitely think personalization in the right way, to your point, is going to continue to be way more expected. I don't, I shouldn't have to tell you. My email, like, you know my email. I don't need to give it to you. I've worked with you guys for 10 years as a company. You have my email address. You don't need to ask me for it. Um, so, yeah, I think that that will be really interesting. We all expect fast service now, but inside most companies, speed is still a struggle. It's the approval chains, the handoffs, the who owns what debate. AgentForce cuts through the mess and actually takes action. It talks to your customers, crushes tasks, and keeps things flowing, all based on prompts and rules that you set so your customers aren't waiting and your team isn't stuck in the weeds. Speed isn't just a nice to have. It's a Competitive Edge and AgentForce helps you deliver it 24. 7. Learn more@salesforce.com AgentForce with bringing in new AI systems, bringing in systems in general, any kind of new system, new technology. There is a unique culture in a company that's willing to do that. Um, and I've talked to companies that are moving very slowly, and frankly, I'm surprised because these are large enterprises. And I'm like, I can't believe that, you know, you aren't accepting this, like, innovation and rapid movement. Um, so could you talk to me a little bit more about what makes the culture at Tomra unique, such that you guys are willing to make these bets and these investments and provide such an exceptional experience to your customers and to your employees?

Speaker B: Yeah, definitely. One of our core values as a company globally is innovation. It's one of our three core values. So it's something that we encourage up and down, uh, the business all over the world. When it came to this specific digital transformation, a common reaction sort of thing can be, all right, does this mean that you're replacing me with an AI agent? Uh, is this going to threaten my job? And I think it was what different companies can see. What worked really well for us is that by having that kind of openness and transparency all the way through the process, uh, is that everyone knew what we were trying to achieve. And this wasn't about replacing people. This is about helping, uh, everybody do their job. Everybody wants to do a good job. And given the tools, the information to do their job in the most effective way, we had the advantage of being quite a rapidly growing company as well. It was like, yes, this probably means we're not going to recruit as many people to support the growth. Um, it's about doing more with what we have already.

Speaker A: Yeah, yeah, yeah.

Speaker B: Setting ourselves up for success, essentially. And also reducing the admin burden, reducing the part of the job that no one likes. It's data entry. And, um, a technician doesn't want to spend an hour fiddling around with their tablet trying to enter the information to what they did.

Speaker A: Ah.

Speaker B: And more often than not, they might, uh, skip over it. Right. And by making it easy for them, um, they can focus on what they're good at and what they like doing, which is actually working with their hands and. And, um, fixing the machines. Getting the satisfaction out of fixing those machines.

Speaker A: Yeah. No, I love that. I think that's. It is something that I find actually really impressive that a company as large as Tamra is Like, I mean a lot of people say our core value is innovation. And you're like, cool, great, like, show me. And you know, you guys are actively playing that part, doing that. And it sounds like your team up, uh, like from leadership down to, you know, everyone that's working on fixing the machines, getting them placed, your operations team, making sure everything's connected, the customer support team, like everyone is on board with that philosophy and moving that way. And that's really special, especially in a fast growing company, because it's kind of can be easy for a fast growing company to bring out a bunch of people and keep doing that and use that as a means to continue to grow. So to decide to kind of be like, okay, we know we've got the core team, we've got it in place, we're ready to go. How can we make what we have work best as we continue to grow? I mean, I really commend you. I think it's very, very hard thing to do.

Speaker B: I appreciate that.

Speaker A: Of course. Phil, as a consumer or even as a business leader, are there any other trends or technologies that you're really interested in or betting on or you're just praying to all the companies out there that someone please invent this?

Speaker B: The trends we're seeing is that uh, people have very different ways they want to interact with companies now. So it used to be that you'd always just pick up the phone, uh, and call. Uh, but now there's so many different channels that people want to use and it could be some people still want to pick up the phone and call and speak to someone. Some people refer to email or web chat or WhatsApp, social media. There's so many different, uh, channels interact with the consumer and having those channels available to the customers and still setting up your support team in a way to handle that in an efficient and effective way, ah, is something I think is really crucial for businesses to do and something that we're continuously, uh, evolving as well.

Speaker A: Yeah, that's actually, that's a really good point. I mean there could be a future where instead of calling you and saying, hey, this machine's not working, I'm just like on X and I'm like, yo, hey, this machine isn't working. And like, how cool would it be that you have a system in place that can pull that, get it to the support agent and they can be like, hey, so and so I immediately can support you on that. Um, that is a really interesting future that I hadn't fully considered in that multi touched way. Um, but yeah, I'm thinking about even, like an issue I had with Venmo earlier today that I'm like, how cool would it have been if I could have just been like, instagram tag, hey, Venmo, this isn't working. Here's a screenshot. And they could call you and contact you and get it figured out for you. Um, so that's a really interesting feature that I would very much appreciate. All right, I am going to take you down a new path that we have never gone on our show. Okay. I'm going on Reddit. So for those listening, we've never done this before. This is a first time and I think we're going to start doing it unless someone LinkedIn DMs me and is like, hey, that was horrific, Lacy, don't do this again. Um, um, but I made a Reddit account. I've had several Reddit accounts over the years, but I made a special one that's only customer experience related. So my whole feed is just customer support, customer service, customer experience, marketing, sales, business, logistics, operations, all kinds of things. That's just a business feedback. So I went through and found one question that I would like for us to answer on air right now. Um, and see, yeah, if we can't help this lovely Redditor out by answering their question and sending it to them after we, uh, we're done here. So. Hello, uh, CRY Revolutionary 7536. They have asked Reddit what tools or strategies have actually improved your customer experience. And we obviously have discussed some here that you've talked about. Uh, so if you wouldn't mind just sharing quickly, uh, what your answer would be to this Reddit user. Um, what tools or strategies have actually improved your customer experience.

Speaker B: Uh, at the risk of, uh, repeating something I said earlier, uh, yeah, go ahead. M. The main thing is empowering the frontline agents to make decisions to support the customer. I think we've all had experiences with different companies where, okay, after contact my supervisor or have to come back to you on that one, it's like, yeah, be able to have that resolution on the spot. That decision on the spot, um, is really big for customer satisfaction. And, uh, I think that that's, that's probably my. I'd say that's my main, um, strategy I'd recommend in terms of, I said improving that customer experience is empowering those frontline agents to make. Take decisions.

Speaker A: Yeah. And empowering them with what you guys have done is use technology to empower them. Right. So they can. They have a great system that they can leverage, they can use, they feel comfortable in. You've trained them, you've prepared them. So there's a lot of different things that you've done to really make it so your team feels empowered to do that. And the reverse of that is what we have discussed. Yeah. Of like whenever you get on the phone and someone clearly doesn't know what's going on with the system and can't help you in that real time moment. Um, so, yeah, thank you. Phil and I will share this with our Reddit user and let you know how that goes. Okay, I will report back. Uh, okay. Our final question that we ask every single guest before we leave is one of my favorite questions. It's what is one experience you've had as a customer recently that you were impressed by? So this is your own experience as a customer. Maybe you bought something, maybe you went to a restaurant, maybe you got a haircut. Right. So what's something. It could be like a local business, could be a big business that you were like, oh, my gosh, I was so impressed with how they served me.

Speaker B: Uh, bought a new set of sneakers a couple of months ago. Yeah. So it's just the way I had been there before. Um, so they knew who I was. And, uh, uh, sneakers wear out about after a year or so, so I do quite a bit of running. So they tend to need some new ones, um, probably more often than some others. Uh, and it's just the way they were able to. Okay, we have, we realize that we, we did a really detailed measurement of your feet, uh, you know, last time you were here, so we don't need to do that again. We've got, uh, you know, here's, here's some quick options for you. I've got them out in your size and. Yeah. See what you think. And I was able to, you know, just try them on pretty much straight away. And the fit didn't feel quite right, but they were able to quick say, okay, well try it with this insert instead. And it's like, yeah, that feels spot on. And, um, yeah, the fact that they just knew so much about their product and knew, uh, exactly how to kind of, they said, solve my problem.

Speaker A: Ah.

Speaker B: And, uh, able to provide that solution on the spot. Ready to go. Uh, that was really, really impressive and

Speaker A: really exciting and actually care and say like, oh, actually, like, I want to help you get this shoe. So I'm going to figure out a solution that would work for you. I mean, if you go to like a, a large. I'm sure. This must have been a smaller store. If you go to, like, a really large store, they're like, oh, it's not in your size. Sorry. You know, good luck, you know, So I think that's really special. And it's. We've gotten shoes before. Recently, uh, my husband and I went to the store, toddler shoe store, and they measured. They did the same thing. They measured our kid's foot. And I was like, whoa, this is really intense for, like, toddler shoes. Normally I just spend 10 bucks at Walmart and get him stuff, but they were, like, showing me why it's so important that his feet have, like, a certain shape and da, da, da, da, da. And they, like, picked out the exact size shoe that he would need. And they documented in their system. So next time we need to go in, which is pretty frequent when you have a growing kid, they would know exactly which would be the next step that he would need. Um, and it felt like, so luxurious. It was like a, I don't know, 15 minute interaction. It was very quick, but it was just felt luxurious to be able to, you know, have his feet taken care of in such a way. So I think that's a great example. Do you want to shout out that. That store or do you want to keep it anonymous?

Speaker B: Yeah, that's fine. Yeah, I was thinking athletes, but, uh, it is actually a chain, um, in Australia. I'm not sure if they're in the US as well.

Speaker A: Uh, I think they are. I'm. I'm not sure. I don't. I guess I don't shopper shoes enough.

Speaker B: Yeah, it's a. It's a train Australia. But, yeah, there's. Yeah, the. The staff at the local store I was at, um, yeah, just said they knew exactly what they were doing and it was a great experience. And, uh, yeah, the sneakers have been fantastic. Of, um. Yeah, done a lot of. A lot of miles on them.

Speaker A: Oh, that's great. That's amazing. Awesome. Well, Phil, this has been so much fun. Thank you for sharing everything that you guys are doing at Tamra. I feel like if we were to check in, in a year or two years, like, there'll be so much more, like, interesting transitions you guys have done. So keep up the amazing work. I can't wait to hear how it all goes. Um, where would be a good place for our audience to either get in touch with you or follow you and keep up to date on what you are doing or just in general, maybe connect with Tamra if they're interested in learning more about what that business does.

Speaker B: Yeah, I think, uh, best thing is to reach out to me on LinkedIn.

Speaker A: Yeah. Okay. We'll drop your LinkedIn in our show notes, and people can go add you. So if you get an influx of in mails and dms, I'm not responsible for that. Okay.

Speaker B: No worries. Awesome.

Speaker A: Um, well, thank you, Phil.