CXChronicles Podcast 220 with Simon Kriss, Leading Voice On AI In CX

Hey CX Nation,

In this week's episode of The CXChronicles Podcast #220 we  welcomed Simon Kriss, Leading Voice On AI In CX, Author, Futurist, Board & C-Suite AI Mentor based in Melbourne, Australia. 

Simon is a customer experience futurologist and thought leader who works with Company Boards and C-Suite Executives on innovation in their customer experience.

He's also the author of the book The AI Empowered Customer Experience & the host of The CXII Podcast, please see the links below. 

In this episode, Simon and Adrian chat through how he has tackled The Four CX Pillars: Team,  Tools, Process & Feedback and shares tips & best practices that have worked across his own customer focused business leader journey.

**Episode #220 Highlight Reel:**

1. Lessons and learnings from being in the CX & contact center space for 35+ years
2. Leveraging AI to transform your customer support & success teams into superheros
3. How the largest contact centers in the world leverage technology, AI & SaaS tools
4. Improving your tech-stack adoption & utilization as you scale your business 
5. Training AI to help you master the design, facilitation & delivery of your CX/EX
 
Huge thanks to Simon for coming on The CXChronicles Podcast and featuring his work and efforts in pushing the customer experience & customer success space into the future.

Click here to learn more about Simon Kriss

Click here to learn more about Simon's Book

Click here to checkout The CXII Podcast

If you enjoy The CXChronicles Podcast, stop by your favorite podcast player and leave us a review today.

You know what would be even better?

Go tell one of your friends or teammates about CXC's content,  our strategic partners (Hubspot, Intercom, Zendesk, Forethought AI, Freshworks & Ascendr) + they can learn more about our CX/CS/RevOps services & please invite them to join the CX Nation!

Are you looking to learn more about the world of Customer Experience, Customer Success & Revenue Operations?

Click here to grab a copy of my book "The Four CX Pillars To Grow Your Business Now" available on Amazon or the CXC website.

For you non-readers, go check out the CXChronicles Youtube channel to see our customer & employee focused video content & short-reel CTAs to improve your CX/CS/RevOps performance today (politely go smash that subscribe button).

Contact us anytime to learn more about CXC at INFO@cxchronicles.com and ask us about how we can help your business & team make customer happiness a habit now!

Transcript

Full Video The CXChronicles Podcast 220 with Simon Kriss.mp4

Adrian (00:00:00) - All right, guys, thanks so much for listening to another episode of CXChronicles podcast. I'm your host, Adrian Brady-Cesana, super excited for today's show. Guys, we have an awesome individual joining us, Simon, Chris. Simon, say, say hello to the CX nation, my friend. 

Simon (00:00:17) - Hey, everybody glad to be part of the CXChronicles. 

Adrian (00:00:23) - So, guys, Simon, I've really really enjoyed my last couple chats with Simon. Simon is Australia's leading voice in AI in CX. He is a mentor on AI, all things AI, international keynote speaker, author, futurist. We've got a guy here that's got a lot of good things to talk about. So, Simon, I'm super pumped that you're joining the show and you're sharing your, your customer focus, business leadership journey with all of us. 

Simon (00:00:45) - Man, yeah, and look, thanks for overselling me right up front. Now I've got to live up to that. 

Adrian (00:00:53) - And the other thing, too: for folks watching, for folks watching this on YouTube today, simon is a super, super dope piece of artwork behind him. Simon, I know our audio listeners want to see it, but tell our video watchers what is behind you again, really quick. I just I have to call it out. 

Simon (00:01:07) - Yeah, so this painting behind me is a piece of work by an Australian indigenous artist and as much as possible. We try to support their work. We have a few pieces of art around like this and it's just a bit of fun as a background, bit of color it's. 

Adrian (00:01:22) - It's awesome in a super, super dope looking. Um, Simon, look, before we jump into the pillars, I'd love for you to just start, man, lay that, lay the ground, like what we do for all these episodes. She spent a couple minutes been talking about before AI, before contact centers, before all the speaking across world. What were some of those early stepping stones? 

Adrian (00:01:39) - Man, give our listeners an idea for sort of how you got into this space, how you got into the world of CX, you got into the world of AI. Give us a sense for sort of how well some of the early upbringings were. 

Simon (00:01:51) - Yeah, well, I mean, you know I, I left high school and decided I was going to be a CX superstar. No, I didn't like everybody else, everybody else I. I had a different path first. So I actually joined the military and I did some years in the military and when I, when I got out of that, I went to work with a friend and I was, we were selling curtains and blinds. I, you know, I was a like used to tell people. 

Simon (00:02:21) - I'm a blind salesman and I've started making phone calls to people who were building homes to, you know, and it was about the time that the word telemarketing was emerging in Australia- and I'm going back a lot of years now- and so I started kind of jumping on the telemarketing bandwagon and joined our big telco here in Australia- they were called telecom at the time, they're now Telstra- and I was on the phones as an agent- outbound calls, inbound, you know, disturbing people at dinnertime inbound calls, and then a team leader and then a manager and then workforce men, and literally made my way up through the contact centers, worked for a number of them and then struck out on my own and I was doing some consulting work and I moved to Hong Kong for two years and 15 years later I came back from Hong Kong having worked all around the world with contact centers on, you know, just about every continent, and so you know I've been in and around the CX space for for 35 years and and kind of was tracking AI for about the last 10 years, but it never really delivered on the promise. 

Simon (00:03:43) - You know there was always these promise of real-time agent tools and you know magic, this and but I they never kind of eventuated. And it was only as generative AI started to pop its head up about four years ago, five years ago, that there actually looked like there might be some promise. And then, of course, once it democratized in like 2022, that just opened the floodgates for a whole lot of work, not just text generative AI engines, but just in general. 

Adrian (00:04:15) - Yep, I love it. So I mean number one: thank you for sharing that. Number two: I- before we can get any further. So like 15 years in Hong Kong and working all over the world, that's awesome. What was 15 years like in Hong Kong? I- even if they're you just give me like a short- I absolutely loved living in Hong Kong and would move back there tomorrow. 

Simon (00:04:36) - It's great because it's right in the middle of the world. You know, it's 12 hours to LA, 12 hours to London, nine hours down to Melbourne. It's literally in the middle of the world and it's an incredibly vibrant city. So, yeah, it was just a great place to base myself from. But I did wind up working. You know, I built a call center in Poland and I helped revamp a call center in Ireland, and then I was in the US and then, you know, it was just kind of all over the world, and of course you're only two hours, three hours from Makati. 

Simon (00:05:14) - You know Manila, the home of so many US call centers, and not too far from India. So you, you're kind of right in the middle of that big BPO space to well. What was then the big BPO space? 

Adrian (00:05:26) - I love it. And I just, I find it fascinating. So many of the people that we have the fortune of having on this podcast, you know, one thing that I hear time and time and time again, these customer focused business leaders have an opportunity or have the fortune to be able to work in multiple countries, multiple cities, different cultures, different places. 

Adrian (00:05:47) - There's no way that that doesn't have something to do directly with why they're able to eventually think about how to work with customers, how to serve customers, how to build teams, how to lead teams, how to motivate teams, how to motivate people in general, how to talk, how to kind of storytell. 

Adrian (00:06:01) - You know, I say it all the time, but I think one of the, one of the most incredible forms of education that any human can go through, it's living, it's traveling number one, but if you have the fortune to be able to live in multiple different places, nothing beats that. You just get, it's just a game changer. It's a game changer as a human, it's a game changer as a business person. So I love that, man. That's fantastic. 

Simon (00:06:20) - Yeah, I've got to say way back, way back when I was just a team leader in a contact center, I remember my group manager of the time getting us all together and she said, you know, if you get hold of this call center thing, it'll take you anywhere you want to go. And I remember at the time thinking, well, that's a load of management BS. Some years later, in fact, about eight years ago, I went out of my way to find her. 

Simon (00:06:50) - She was in retirement living in Western Australia and I went out of my way to find her, call her up and thank her for that advice because the industry literally has taken me around the world. So she was right and I was wrong. 

Adrian (00:07:04) - Right, man. She was absolutely right. I love it. Simon, I'd love to, I'd love to dive into the first CXPillar team, man. You and I have had some awesome conversations these last couple of weeks, but a couple of things. 

Adrian (00:07:15) - Number one, I want you to answer however you see best fit, but there's a couple of things going on because you've got loads and loads and loads of just general team building, team construction, team leadership experience, but you kept giving me, our last couple of chats, all these different ideas around some of the ways the teams across the world are leveraging AI. I'd love for you just to spend a couple of minutes talking about from your own experiences and some of the big things that you've really kind of honed in on after you've worked at multiple teams. 

Adrian (00:07:41) - And then I'd love you to kind of sprinkle some of the things that you're seeing with the AI and the CX space around, and then some of the things you put in your book, man, too. I didn't even call this out. Some of the things you put in your book, the AI and Power Customer Experience. I'd like you to spend a couple of minutes just talking about the first pillar of team. 

Simon (00:07:57) - Yeah, look, teams, it's an interesting thing in context that so many of them say, oh, we have teams. And I kind of say, no, you have groups of people that are forced to work together. You know, a true team, if you think of the Seahawks or, you know, pick any team in any sporting code, Detroit Red Wings or whatever, they have a few things in common. One is that they have an identity of themselves. They have a group of people that are almost fanatical about supporting and will give themselves up for the betterment of the team. 

Simon (00:08:41) - And I think that's where we need to get to in contact centers in particular. You know, your CX groups need to feel like they are a team. I love that banner that's behind your head. You know, I could imagine each team in a contact center having a banner like that, having an identity, being a part of a team would be super, super strong. And the beauty of what I do so much work around AI is that AI can't build a team. It's a human connection. It's humans dealing with humans because, you know, humans are so weird, right? 

Simon (00:09:19) - We're full of empathy and sympathy, but at the same time, we're full of bias. And so, you know, we're really interesting characters. And for me, what we're getting even more and more and more into now is less about IQ and more about EQ, that emotional quotient, and LQ, a learning quotient. So for me, the leaders and the teams that are really going to win as we move forward are emotionally connected to each other. They're emotionally aware. 

Simon (00:09:54) - And, you know, they've got a great desire to learn and change and do rather than, well, this is the way it's always been done. So this is the way we're going to do it.

Simon (00:10:06) - So it's going to be a very different focus. And look, I think the role, sprinkling in AI, I think the role of a contact center agent is going to dramatically change in the next couple of years, which I think will be accompanied by a fairly significant salary bump. But we will just be expecting a lot more of our staff. We might have slightly fewer of them, but we're gonna be expecting a lot more of them. 

Adrian (00:10:33) - Yeah, I think that's extremely well said, Simon. What you just made me think about there is just like, well, there's someone I worked for years ago in New York City. I remember her telling me that the world was going, the future of the world had been traditionally been ran by left brain thinkers, but that we were entering a time, this was about 10 years ago, Simon. And she said, you know, Adrian was funny. She said, you know, we're entering a world which I truly believe. And she was an awesome leader, awesome lady, like just super smart. 

Adrian (00:11:02) - She's like, but I think that we're entering the world of the right brain thinker leading the world. You just said, humans are funny characters and we are, but you said empathy and you said EQ. And when I think about right brain and I think about, you know, a human's ability for emotions and focusing on all these abstract things and really kind of more creative and stories and people, humans, humans are like that, right? Left brain, logical, focused on facts, all that stuff. You gotta have that too. 

Adrian (00:11:27) - You gotta understand sort of what's black, what's white. You gotta know how to do your math or else you're not gonna get very far. But when she said that, I always, I always would think back on this sometimes as I kind of get deeper into my CX leadership journey. And it's funny because like, man, I've worked with all sorts of different types of leaders. I've worked with some incredible left brain thinkers that were just super duper duper high EQ smart, just smart, right? 

Adrian (00:11:49) - But you had a hundred customers on the phone with that person, or you had a hundred customers, a person with that person, maybe not the same as an individual who literally was able to have that empathy, do the connectivity, be creative, storytelling, listening. Man, you can't learn anything if you're not listening, right? A lot of humans don't get that part. But I love that. I love that thought. One last piece. 

Adrian (00:12:09) - When everybody gets all scared and worried and hot and bothered about AI and machine learning and Terminator finally entering, you know, current day, you know, I love that you and so many other forward thinking AI focused thought leaders, y'all say the same thing, which is, look, it's really, it's here to be able to help humans do things that, number one, computers can just simply do way faster, way cleaner, way clearer. 

Adrian (00:12:37) - But number two, it's there to be able to allow humans to be able to make the decisions that they still know other humans are going to have to be able to agree with or to have some type of connectivity with. So I just think it's funny. Everybody keeps kind of talking about AI in such a negative bit, as it relates to the team and what we're talking about right now here, in this first pillar. 

Adrian (00:12:56) - Dude, there's so many different things that it can do to help with your employee experience and with actually better service, better understanding your team, better how to motivate your team. So I just, I love some of the ideas right there. 

Simon (00:13:10) - Think about it this way, right? In the movie Iron Man, which, you know, whether you like Marvel movies or not, and I hope that's the right franchise, you've probably seen bits, if not all of the Iron Man movie. Tony Stark, the character, is still a human and he is the superhero. It's just what allows him to be a superhero, is Jarvis, this incredible AI system that sits around him, does all the evaluations, does all the calculations, does all the grunt work, and then presents him with options and says, here you go, you could do this or this or this. 

Simon (00:13:46) - Now, often, and it's, you know, a theme of the movie is he goes rogue and says, no, I don't want to do any of those. I want to do this illogical thing over here. And AI says, that's not a smart thing to do. And he says, I don't care, I'm going to do it. That's very much how I see AI around people's lives is it's there, it's not designed to take over your job. It's not designed to do what you do because we're not in the land of artificial general intelligence yet. We're still in the land of artificial narrow intelligence. 

Simon (00:14:21) - So it's quite normal and quite okay for a human to go, you know what, no, I don't like that output. I want it to be like this. You know, and we, you just mentioned empathy before. You know, some of the AI systems today can feign empathy really well. You can teach it that if somebody says, oh, my kid got a broken leg, it can go, oh, I'm so sorry to hear that. Now, it doesn't understand that a broken leg causes pain nor does it understand that a parent feels pain when their child is injured. 

Simon (00:14:54) - What it understands is broken leg equals bad, say something nice. So it doesn't understand empathy, it just feigns it. And that's why the human element is still going to have to be there for a very long time. Even as we move into general intelligence, it is still gonna have to be there because we, you know, we can cry and laugh at the same thing at the same time. You know, you can be at a funeral and everybody's crying and somebody retells the story and everybody starts laughing, even though it's a sad event. 

Simon (00:15:30) - That's highly illogical for a computer, but completely normal for a human just because of how we are and how we've been brought up and the things we've been conditioned with. And you know, a five-year-old has probably had 20 times the input of any large language model.

Simon (00:15:52) - let alone by the time you get to 30, 40, the inputs that your neural network in your head is dealing with is a million times bigger than what a large language model does. So there is definitively a point for humans to still be involved. 

Adrian (00:16:12) - Yep, I think that is spot on. And I also just think your comment about the funeral, it's so true. There's still so many things that a human would be able to detect, understand, comprehend, and then be able to respond to in a way where AI, sure, it might understand some of the hints, it might be able to get relatively close from just a response-based perspective, but it's still going to, to your point, it's gonna take a long time if humans continue to manage. Now, one question though, open source, right? 

Adrian (00:16:44) - And this open source idea, and then this idea that theoretically, as all of us continue to essentially add to some of the different language models or add to some of the different ways that some of these models are actually speaking, that's a different thing, right? That's where the intelligence or the robustness of the comprehension of some of these different AI solutions, that's where they're going to, the velocity is gonna be extreme, yeah? 

Simon (00:17:09) - Yes, yeah, the velocity is going to spin up. But once again, you've got to understand, it sees examples and tries to copy those, right? And tries to predict, every AI model everywhere in the world is about predicting the next best whatever. So even your chat GPTs, because everyone loves to use that as the example, all it's doing is guessing what is the next most appropriate word, and then the next most appropriate word, and then the next, and then the next, and then it's predictive models. 

Simon (00:17:42) - So when it's spinning up velocity and appears to be getting smarter and smarter, what it's doing is it's just seeing more and more examples of things and copying those examples. But imagine an AI trying to understand the concept of political correctness based upon scenario. So I'm together with three of my buddies, we're out fishing, the language is probably more colorful than I would normally use, we're four beers in each. What comes out as politically correct between guys picking on each other is one thing. 

Simon (00:18:22) - If I then move into sitting in a boardroom meeting, what is politically correct there? So it's situational, and humans learn how to adapt to situational. AI still struggles with that. All it knows is I've seen this example, this example, this example, so that's what I will mimic. 

Adrian (00:18:44) - Yep, okay, super well said. And then again, it's just another reason to remind folks that we are still very much in a position of control and that we're very much in a position where we still need to be able to simply leverage it as a tool that we need to keep an eye on and we need to continue to monitor, change, make iterations on, all of that. Simon, I'd love to kind of move into the second CX pillar of tools. 

Adrian (00:19:05) - Number one, from our last couple of conversations, when you've worked with all of some of the biggest contact center tools on planet Earth, you've seen all sorts of different types of technology, you've worked at all these different businesses. I'd love to kind of hear you talk about tools for a couple of minutes. 

Adrian (00:19:20) - And again, it can be high-level ideas around sort of what you've seen some of the best companies or what you've seen some of your best companies, sort of how they've leveraged specifically the tools that they need for their business and the tools they need for the customers. I'd really like to know some ideas or hear you kind of speak a little bit about what are some of the ways you've seen some incredible organizations help their employees with tool utilization? 

Adrian (00:19:45) - And then again, I'd love to, if there's ideas around how our listeners can think about how AI can help them with that, please, I'm all ears. 

Simon (00:19:54) - Yeah, it's interesting. People, you know, I have to peer back a little and talk, AI adoption is a little bit like teenage love. Everybody's talking about it. Everybody thinks everybody else is doing it. Actually, very few people are doing it and those that are probably doing it wrong. And it's very much that same way. So the adoption of AI tools is way slower than what people think it is. But there are some great examples of how people are picking up things.

Simon (00:20:30) - small chunks at a time. So if you're going to step into AI, don't think about trying to launch the Starship Enterprise. Just think about getting a seat on the next Bezos rocket that's going up to have a look at the Earth. It's that kind of thing, right? Small chunks. So things like, you know, a great use case that I saw a center using recently is for the chat agents, they just type in the words, no refund. Generative AI takes that and turns it into, dear sir, on this occasion, due to a lapse in your policy and blah, blah, blah. 

Simon (00:21:07) - And the fact that the car is 28 years old, unfortunately we cannot process a refund at this point. You know, something lovely and flowery, but the agent's not having to write all of that. They're, you know, they're being used for higher level things such as decision-making and things like that. You know, they're still human in the loop. I've seen a similar one to your point around EX of, you know, what annoys contact center workers? You know, at their core, what annoys them? 

Simon (00:21:39) - Yes, pay, permanency, parking, the three big Ps, sure, but not being able to control their roster. Well, there's an AI-powered tool that can help with that now. Having to do donkey work that they just hate, like writing a post-call summary and things like that. Well, there's AI for that. But we've moved beyond that too. There are solutions now, for instance, that listen to the calls and their whole job is to look for calls that have a higher motive load. 

Simon (00:22:13) - So think about a government worker that is in the pensions department and they get a call from somebody who says, if I don't get my pension payment tonight, I'm going to step out into the traffic or my husband's going to come home and beat me or whatever that terrible situation might be. They're highly emotive loaded calls. So we can now use AI to track that. And at the end of the call, hold up before the next call just comes through and say to the agent, that was a pretty heavy call. 

Speaker 3 (00:22:47) - Are you okay? 

Simon (00:22:48) - Do you need a break? Do you want to talk about it? Or flag to the team leader, hey, this person's having a really high emotive load call. You need to go over and tap them on the shoulder. It's those wellbeing care type applications of AI that I'm most excited about. I love things that really support humans and help humans do their jobs better. And those things that pick up people on the fringe that might otherwise not be serviced by mainstream tools and enable them to do that. 

Adrian (00:23:22) - Yeah, I love that. I've constantly been blown away by how many of the companies we've had the honor of working with here at CXC that they might have loads and loads and loads and loads of call data, right? But then when you start to dig into some of the VSU reporting, you start to try to understand what's the customer level, what's the customer, hey, what are we doing? What do we need to do more of, less of? What do they think positive about our product? 

Adrian (00:23:46) - And my point is like some of these companies, they might have hundreds of either inside sales or calling agents a day and no one's ever actually done any type of deep dive on sentiment analysis, on keyword smithing and understanding top 10s, top 20s, positive, negative. So positive sentiment, negative sentiment. All of those things, once you start to understand them, they can be unpacked for both the CX side of like, okay, how do we make our product better or service better? You are spot on, same on the EX side too. You can do the same damn thing. 

Adrian (00:24:18) - You start to extrapolate that across different teams, different departments, different calling agents. 

Adrian (00:24:22) - You can begin to literally see who's had a really great month and lots of, seems like a lot of positive, positive activities this month versus sometimes like every one of us that's listening to the show has had the teammate that just gets the worst customer calls or they just get like, it's almost like their luck is just absolutely, even if it's around Robin or whatever, however you guys are doing some of your distribution on the call side, but like everyone has a teammate that just pulls every unlucky rabbit out of the hat, right? 

Adrian (00:24:50) - So I think you're spot on there. 

Simon (00:24:51) - And is that, it comes down to, is it a certain type of call that is frustrating the agents? And maybe that's because the process to handle that call is an absolute dog of a process. Or is it this one particular agent just seems to get nasty calls and actually it's not the callers, it's the agent. Is it the agent on the day? Do you have agents that are fantastic on Monday and absolutely terrible come Thursday or Friday? All of these things. 

Adrian (00:25:20) - Is that all of us or no? 

Simon (00:25:22) - Maybe. I can neither confirm nor deny. All of those types of things are being able to done, be able to be done now. Whereas before we struggled with enough compute power and enough tech to be able to do. So think about quality management in the contact center. What are you listening to? Half a percent of calls, maybe 1% of calls, a really common thing I hear is four calls per agent per month. Well, that's fantastic, except they're taking 5,000. 

Simon (00:25:55) - So as you get into, and even speech analytics, as good as it can, bad speech analytics is just word spotting, basically. It's just looking for certain words. Good speech analytics says, well, we heard this word within five or six words of this word, or we heard these two phrases. And so we can derive something. Generative AI takes the entire telephone conversation.

Simon (00:26:22) - and, you know, can analyze kind of what's going on in that entire string. So, not just little sentences or little phrases, but the entire conversation. Because there's a big difference between someone saying, I want to cancel my account now, and somebody saying, I want to cancel my account now. They're two very different phrases. And when you turn them to text, it's hard to tell them apart. 

Simon (00:26:49) - But generative AI uses the context of the entire call to derive that, okay, early on in the call, you were calling me, you know, an SOB, and, you know, you hated me and blah, blah. And this part of the call, you said you wanted to cancel the account now, we can kind of guess how you said that. So, and it can do that because of compute power, you can now do that across 100% of your calls. So, you can derive sentiment, you can derive to your thing, not only keywords, but key themes. What are people calling up about? 

Simon (00:27:24) - If you suddenly get a surge of calls over two hours, wouldn't it be great to just be able to type into your AI engine, what's going on that's driving this increased call volume? Yep. And let it come back and tell you in a couple of seconds what's going on. Yep. You know, there's some real power in this stuff. I love it. Simon, what have you seen across our industry, but also just from some of the different teams that you work with, some of the different clients that you're working with today? 

Simon (00:27:52) - You know, when you talk about process, right, and you talk about some of the playbooks, or the standard operating procedures, or even a lot of companies now, part of how you're going to train your AI, it's really going to be contingent upon how your knowledge base, or how your FAQs, or how your internal facts are actually, how they've been curated, who wrote them, when were they written, are they old, are they new, are they fresh, are they stale? 

Simon (00:28:13) - But I'd love to kind of hear you kind of talk about how process fits into some of this, or some of the things that you're seeing clients, or you're seeing different companies doing to stay ahead in the process, and to make sure that they're wrangling process, and to make sure that they're kind of tying it all together to appease the other pillars. So what I'm seeing so far is that they're not. What people are trying to do is strap AI onto an existing process. 

Simon (00:28:38) - And, you know, it's like strapping a rocket engine to a car with three wheels and hoping that it holds together. It's just not going to, right? You can't get a 1970s Pinto and put a jet engine on it. So what we're starting to see is that people are using the analytical side of AI to try and derive where the problems are in their process, and then go and start fixing them. The thing that's holding them back, though, and you hit on it perfectly, is the data, right? So, yes, they may have some structured data in a structured database. 

Simon (00:29:19) - And that might be a knowledge management system. That could be your call tracking system. It could be your call recordings, your CRM. All of those have structured databases to them. And aligning those is not overtly difficult for AI. What becomes a problem is when you get to your unstructured data. This is the stuff that's sitting in PDF documents, in Confluence, in SharePoint, in Teams messages. Very, very unstructured data. How do you bring that forward and get it in so that AI can use it? And then the last piece of that is tech debt. 

Simon (00:29:56) - And we have a lot of organizations that are carrying a lot of tech debt. I had someone just recently say to me, we've got three Lotus Notes databases that we'd love to put AI on. And I went, good luck. Like, that's just impossible to do. And they've never paid that tech debt of moving those databases, migrating them to something that a technology like AI could use. So, there's a few things kind of getting in the way of people being able to really change their process. But as they do, we see that the changes in process are not small. 

Simon (00:30:36) - The changes in process are massive. So, you know, I always encourage people to not think about an externally facing AI application first. Don't do your external chatbot first. Please, for the love of God, don't do that. Because when you make a mistake, not if, when you make a mistake with implementing AI, if it's internal, it's controllable. If it's external, your reputation's already done. And we've seen examples of that.

Simon (00:31:08) - So pick a simple, internal use case and go with that first. You know, one of the ones I love is build a little internal chat bot that takes all of your HR policies and procedures, put them into a conversational bot that people can just ask questions like a human of, hey, how do I do this? How do I apply for that? What's the policy around this? And have something just respond to them. It's quick. It's easy. Often that information is somewhat structured. 

Adrian (00:31:43) - Yep. 

Simon (00:31:43) - And so it's a, you know, it's a good place to start. And it's not a moving set of data as in, you know, unless you start folding in all of people's annual leave entitlements and blah, blah, blah, blah, blah, which you can do as a second step. But yeah, I'd say step small and look for wins in process that way. 

Adrian (00:32:04) - I love it. I love it. Simon, fourth pillar, feedback. Just spend a few minutes talking about some of the simple ways that you are seeing some of the leading companies in the world leverage AI for feedback. 

Adrian (00:32:19) - And then I'd love for you to kind of just share a few bits of wisdom or advice around some of the things for our listeners, a lot of founders, a lot of CXCS context and leadership, a little bit of advice from all the different companies you worked with and from all these different places that you've traveled all over the world speaking on this topic, meeting these different customer focus business leaders. Love for you just to kind of wrap us up and spend a few minutes kind of talking about feedback. 

Simon (00:32:44) - Yeah, sure. I think there's a couple of bits of feedback. One is obviously customer feedback, right? That's the big elephant in the room. And my prediction, I'm going to look into my crystal ball here and make a prediction, is that post-interaction surveys are dead. They are on the way out. For the most part, only a small percentage of people complete them. Usually when they complete them, they're either so pro the outcome, like they got exactly the outcome they wanted and more, or they are completely anti the outcome, right? 

Simon (00:33:25) - You tend to get a lot of the two extremes. And when you discount those, how much intelligence are you really getting? 

Adrian (00:33:32) - Yeah. 

Simon (00:33:33) - To even then put up a question that says, tell us in your own words how it went. Well, that's okay. You can analyze that. But why are we imposing that upon the customer? That's what I call having CX done to you, not CX for you, right? 

Adrian (00:33:48) - Yeah, yeah, yeah. 

Simon (00:33:50) - We're forcing the customer to do these. Why don't we use the tools at our discretion, you know, the tools that are available to have AI derive all of that? Let it train it to listen to calls, talk about sentiment change, because I don't subscribe to the theory that each call has a sentiment. I think sentiment changes over the call, ideally from red to green, like the person starts off, oh, you guys, I hate you. You've done a terrible thing. And ends up with, wow, thank you. That was awesome. That's kind of the epitome of customer experience. 

Simon (00:34:28) - But letting AI do that, and then letting AI stitch together all of the various parts of the CX journey. You know, often as call center people, we get caught up in CX is the phone call. And CX isn't the phone call. CX is the product, how the person purchased it, which might be in-store or online. This call that we're talking about is just a part of the service process. Then there was, how did it get delivered? Was the driver well-dressed or was he stinky? And, you know, did he park the truck on the lawn? 

Simon (00:35:01) - You know, all of these things kind of fit into CX. And starting to stitch together data from right across your organization to truly get a picture of CX is where I think feedback is going. So I think feedback is going to become more a function of what we do as an organization rather than something, a burden that we place upon the customer. 

Adrian (00:35:26) - Yep. I love that, man. I love those ideas. I think you're spot on. I can't tell you how many times, and, you know, I'll just call it out because everybody already knows they go through the same damn thing. It's airline companies, it's telecom, cable companies here in America, man. These knuckleheads, man, like, first of all, no, I didn't enjoy your terrible Delta flight because you already know that we were two hours late. You already know that we sat on the tarmac for an hour. What do you mean you're asking me to answer 27 questions about Delta? 

Adrian (00:35:56) - Get out of here. Like, no, forget that. That actually just made me hate you more, Delta, frankly. That's number one. The other thing, too, is with telecom, the same type of thing. They can see all of our user data. They can literally see everything. They can see our internet usage. They can see what we're watching. They can see what we're consuming the most of, the least of, and then they'll send these, like, again, just pages and pages and pages. 

Adrian (00:36:18) - It's almost disrespectful, and it's crazy that some of the guys and the gals that are running CX at those companies who probably, Simon, similar to you, similar to me, they probably have decades of experience.

Adrian (00:36:29) - Aren't you changing that? It's crazy to me. That's number one. The second thing is just like I think that you're spot-on with the journey. It's funny when we do our journey maps at CXC with our clients, one of the things we really push our clients to understand is where are you guys already capturing or where do you already have historical feedback based data right? 

Adrian (00:36:50) - So as you're going through touch points and as you're going through all the, all of the intricacies of the different areas of you know, pre-selling, all the way through the, the conversion, the through post-sale, regardless of how you're building your mess, and it's crazy how often you literally just get a post-sale one. It's a. We literally already know that we need to go through a hundred prospects to maybe get three to five customers to convert. Why aren't we tracking how we theoretically lost the other 97 to 95 prospects? 

Adrian (00:37:19) - Because then how do we ever help our marketing and our sales team? So, like I think you're spot-on, we're just getting more thoughtful, a little bit more sophisticated with being able to leverage or have the goal to be able to go and ask feedback at different points you traditionally don't. And then the last thing: Tuesday place. 

Adrian (00:37:34) - I mean, I just think- and I know you broke this up, but like you're right, like customer feedback is one part of it, but then in the employee experience sides, so many companies undervalue what the guys and gals that they literally pay to show up every damn day, speak to those customers every damn day. They don't take any of that EX based feedback and try to correlate it back into the CX side and that's just. It's just mind-blowing to me. 

Adrian (00:37:58) - You know, I know it might just be the CX nerd with some of these companies that it's like how could you not leverage that and just try to tie it together and then find your priorities from there? 

Simon (00:38:06) - So just some and this, and there's some, there's some stuff. You see, you know, how long did someone take in the sales process online? And compare that to propensity to call, if you really want to. You know- and I'm not an advocate of call deflection, I think that's just a stupid, short-sighted CFO response. But if you're really trying to just to make it so that people don't have to call, which is, you know, the best type of service- is there a link between the time taken to purchase and whether or not they call? 

Adrian (00:38:41) - So is it? 

Simon (00:38:41) - The people aren't quite understanding what they got. Is there a direct correlation between how long it takes to get delivered and propensity to call? Is there a correlation? You know you're looking for all of these correlations that can then lead to causation potentially, but you know- and going back to your other point about some of the crazy stuff, I see government departments, like the tax department, asking for an NPS score. Yeah, are you likely to recommend us to someone who's gonna recommend a tax department? And if they are, what? 

Simon (00:39:14) - How many competitors are there for the national tax department? So, yeah, I just see some of these things that you just look at and they're absolutely stupid. You know a police service. Did you enjoy your call today? Well, nobody calls the police when they're having a good day. It's like a windscreen company, right, you don't call them. Hey, I'm having a great day, I might call up the windscreen company. No, you call them because some idiot threw a rock at your car and broke the windscreen, right, yeah, yeah. 

Simon (00:39:46) - So you know, I think we've got to take all of this stuff into into consideration. I really do. 

Adrian (00:39:52) - I love it, simon. Look number one. This has been absolutely fantastic, for I let you go, my friend. Where can people find out more about some of the incredible work you're doing. Where can people find your book? Where can people get in touch with Simon Chris to learn more about some of this awesome stuff that you're doing and some of this, some of this stuff that we're gonna all have to understand how to leverage and how to think about how to use in the future? 

Simon (00:40:12) - Yeah, look, the the book is available on Amazon. It's called the AI empowered customer experience. For those that are watching, that's what the cover looks like. It's an AI picture on the cover, of course, and I'm working on version 2 right now, because I wrote that book six months ago- seven months ago- and seven months is a very long time in AI. So I need to, I need to update the book, but the best way for people to get hold of me is via LinkedIn. I live on LinkedIn, I post a lot of stuff there and I'm always happy to entertain questions. 

Simon (00:40:50) - You know people can just DM me with a question, mostly because nobody kind of has a degree from a university and in CX and call center management, or very few, most of us have learned to buy a little bit like snow skiing. You know, you learn the most when you fall over. Yes, it's the same type of thing. So I'm a big advocate of sharing and, you know, helping each other along and answering questions, so if I can do that for people, I'm more than happy to. 

Adrian (00:41:21) - I love it. Well, Simon, it's been our absolute pleasure having you in the CH chronicles podcast. Best of luck to you in the future. My friend and I'll look forward to continue our conversations in the future. 

Simon (00:41:29) - Thank you, sir. I'd love to keep chatting.