Latest / Range Podcast / AI Is the New Excel for Compensation Teams
Transcript
- Giac Soliman: Good morning everyone and welcome to a new episode of the RANGE Podcast. Today I'm with Cathy Odds. Cathy is an experienced compensation leader. She's currently vice president of compensation and mobility at Wiley. She brings nearly two decades of experience across Wiley, BNY, American Express, and Mercer with deep expertise in compensation, incentive design, job architecture, and international reward programs. Kathy, good to have you here. Welcome. Kathy Odds: Thank you so much for having me today, Jack. It's great to have an opportunity to chat. Giac Soliman: So you've lived in New York for I think seven years, right? And before that you lived in many different places in the world including London. I'm curious to know in your experience is New York more diverse than London? Kathy Odds: Statistically, I believe it is. somebody living here, I feel they feel very similar from a diversity perspective. you name a country or a culture, you can find it probably in both cities. The way that different nationalities represented is slightly different, inevitably being in the US, New York has more people from Central and South American countries, perhaps in the UK. The UK has a great representation from India, Bangladesh, Pakistan, but what are the things that I love about both cities? And when people p when people ask me to choose between the two, I refuse. I love both in different ways, but I do love the diversity of both cities. Giac Soliman: What's the one thing you miss about the UK after living abroad for so long? Kathy Odds: Pub lunches. Just going to the pub. Giac Soliman: Is there any specific... Kathy Odds: I there's something about sitting in a British pub with a pint of beer and a particularly a good Sunday roast with a fire roaring. Not when we're going through a heat wave, but in the winter with a fire roaring next to you. There's something about that being a very chill vibe that you just don't quite get in New York City. There there's Irish pubs, there's English pubs that try and replicate it, but it's just not. quite the same as going down the pub with your friends, as it is in London. So that's probably the thing that I miss and it's the thing that always springs to mind when people ask me that question. I somehow default to pub lunch. Giac Soliman: Who makes the best burgers, England or the US? Kathy Odds: You keep making me pick sides here. Giac Soliman: yeah, it's part of the game. Kathy Odds: I'm glad you went with burgers and not with pizza. Giac Soliman: That was the reason for that. Kathy Odds: I think that I'm gonna sit on the fence and say the variety is is there in both cities and what I've loved about both is you've obviously got the fast food outlets. I'd say they're the same whether you're here or in the UK. but then what I love is the diversity of other boutique restaurants that have popped up that are specializing in burgers. I will say I do love it's a particular brand. I don't think I want to name a brand in there is there is a chain in London that I used to like a lot for burgers that I don't get here. and I will go back there when I go back to London. but yeah, a good a good burger you can find them in both places. Giac Soliman: I've lived in London for eight years and I never tried five guys but everyone seems to be hyped about them. Are you one of them? Kathy Odds: I've never tried I think they have it in New York. I'm not sure. Giac Soliman: It must be a marketing technique. they fill, if you do takeaways, they fill the bag of chips up until the very top. I think that could be it. they charge you twice of what you would pay for burger, but hey, you get chips up until the edge. So I think that might be it. Yeah. Kathy Odds: I didn't know that. I did not know that though. Giac Soliman: I watched a YouTube video about it some times ago on one of those drunk nights, where you just don't know why you're still awake, you should go to bed, you're still watching videos of burgers and chips. Kathy Odds: I remember when Five Guys opened its first door in Covent Garden in London, however many years ago that was and the lines out the door for that. And I think now Five Guys is probably it's not one place in London, they've probably got multiple outlets. Giac Soliman: Yeah, they're all over the place. Kathy Odds: I will say living in living in New York, one of the great things is sure there are many different big chains that exist, but there's also so many independent small restaurants and that tends to be my direction of choice for food versus the big box chains. Giac Soliman: And no doubt. Now shifting gears a little bit, you're in a, you came, you decided to come to an AI podcast, a podcast where we tell our stories about, AI adoption and how's it all going when it comes to learning and adopting AI. And the key question is why? What is it about AI that interests you, intrigues you so much that you decide to come on the podcast to speak about it? Kathy Odds: I think AI is prevalent every day for everyone. I think try going through a day where you don't hear the words AI at the moment. And if I compare that to just a year ago, where maybe it was it was sitting there bubbling, now it's become very prevalent. And I heard a great statistic yesterday that of everybody using AI, only 10% of people really know how to use it. And so I think the more that conversations happen about it, the more that whether it's informal or formal learning, the more we learn about it, AI is not there to replace us. AI is there to do things differently. And perhaps unlike 10, 20 years ago, earlier in my career, where there was more formal training was the norm, these days I think that learning has become more integrated in everybody's everyday. And so the opportunity to come and chat about it and I think in some ways normalize that it's okay not to have all the answers. There's a lot of expectations about everybody using AI, whether it be I think in our personal lives it's easier. In work lives it's harder to necessarily navigate exactly what is AI going to do for us and how do we use AI responsibly and what does that look like. And so the more opportunity to have conversations about it, also to normalise, hey, it's okay if you I understand what AI is, but do I understand what AI is? And discover together. Giac Soliman: that rings the bell. We both spent some of our career in consulting, one consulting company specifically. And there's been quite a few instances where I across someone tells me, consulting is changing. Then another day someone tells me, consulting is dying. I wonder where you stand on that, given that knowledge management engine for so long. You think about job evaluation methodologies, they would have a stronghold when it comes to knowledge and essentially they would sell applied knowledge, right? Any project would be, we look at your needs and then we come up with a solution that's custom to your needs, grounding our methodologies, but applied to your context. So have you given this any thought? and say, not because you might want to go back to consulting, but maybe one of those 2 a.m. questions you have instead of watching videos about burgers and chips, you might be thinking, what's happening to consulting? Kathy Odds: Until you ask the question I can't say I th thought deeply about it, as you ask the question it's a really interesting one. I don't think AI is there to replace humans. Ultimately, what I've learned about AI so far is that it still relies on solid data and information. That's what it's drawing from. And if humans are not there to create content and create information that effectively forms the basis of where I where AI is drawing information from. AI isn't that helpful for us. And so I think actually the consultancy role, is it evolving? I'm sure it is. And it's perhaps less around the data side of it and more around the intelligence side of it. How do you draw intelligence from the information that is being created? How do you pull together information that you know the consultancies sit on vast amounts of information across their clients and how do you responsibly use that to support customers. so I think if I were to go back into consulting, it would look quite different from 20 years ago. And if I think about too way back to my consulting days, even then there was already a direction within the consulting world, realizing that there was a consultant in the US probably doing the same type of work as a consultant in the UK and there was already work afoot in that organization to say, well, how do we draw that together to draw better insights for our clients? And I think AI is there to support that. But I think the consulting role perhaps shifts towards more research intelligence from the information. and creating the content but not bread and butter work of comp consultants of job evaluation and job matching and survey production evolves in a way that I don't want to try and predict, but I suspect that it will evolve quickly. Giac Soliman: Consulting is undoubtedly facing some pressures. I agree not there to be replaced by AI, but it's just changing. Equally we're seeing some change in house as well. And as a board leader yourself Where do you see most of this pressure coming from? Is it coming from the top leadership teams telling you go ahead and figure it out? How are going to use it? Is it coming from the bottom team members asking? What's it going to be about my role? How are you going to change your operating model to make sure that my role is not redundant? Where do see the pressure coming from? Kathy Odds: think I'm lucky to work for an organization that is very much leading the way in AI and finding the way to harness its power to advance what we do as an organisation to benefit our customers and so there is very much a strategic lens on AI. the way that I think about it is probably in three layers. There's an efficiency layer Where can AI take over to us that my team can focus on higher value work? So where they're pulling data out of one system into a spreadsheet, number crunching, and then pushing it back somewhere else, is there efficiency that can be found? I think there's a second layer, which is the delivery layer. How do we use AI to give our stakeholders something better, not just something faster, but how do we really change what we deliver? And then there's a layer that I don't even know how to tap into yet, but that's the one that excites me, which is What would be possible if I go from the lens of what solutions would look like for our managers and our colleagues? And what could we build that perhaps looks very different from today that makes the way that reward is delivered feel entirely different from colleagues and managers, where something may now require individuals to look at data, pull data, consolidate, provide an answer? If you layer into that all the legislation on pay transparency, wherever you are in the world, because it's coming in lots of different places, how do we really think about that experience? As it is now and where my team ran out, they're asking me the question. My encouragement is use the tools that we have had available to us from an AI perspective. We have two different platforms that we use. And start to share about how you're using what did you use it for? Importantly, what did you use it for and what didn't work? 'Cause we learn as much from what we tried and didn't work, if not more than what actually did work, I think my team are all at different stages of of AI adoption from I don't know what I don't even know where to start, Kathy, through to look at these three things I did today, Kathy, and what can we use and what can we not use. and I think it's about continuing to encourage that trial and an error, a trial and experimentation. And I think the piece that I haven't tapped into yet that I'm really excited about as well is how do we work with others? How do you get somebody who's not currently working in the reward team or even within the wider people team to say, Hey, an external lens that gives ideas for what we could build. I think the other big challenge is that I if I go back again twenty years, there was something I worked on at the consultancy, which was creating a database. Now at the time it was in a shared Excel, which today sounds like the norm, but I can promise you twenty years ago a shared Excel was something quite advanced. Updates to that tool maybe came once a year and I had to deal with, okay, well, what are the new features we can now use? Right now, AI and any app on our devices updates daily, weekly, monthly with an that constant learning of what's now available. So just when you think you've understood something and new features launched and now now I've got to relearn what d what does that do? Is it beneficial to me? And I think that, twenty years ago I might have had somebody specifically dedicated to that database and the product manager who can then explain all of the changes and what we could do with it. Now I think there's an expectation we continue to learn. And, as my brain gets older, it definitely It's not it's harder to keep up with every possible feature that comes out. Giac Soliman: On that note, how do you keep up? Kathy Odds: I don't think I do is the answer. look being part of the range community is fab. And the things that I have learnt being part of this community in a short space of time is phenomenal. Small things one of my biggest learnings was ask co pilot to write your clawed prompt. That's really obvious if you think about it. I'd never thought about doing that. And that changed how I operated within AI within twenty four hours. Just one person mentioned that within the community, seeing what others are doing, and I think that's one of the ways I'm keeping up to speed with it. And then it's just where the content looks relevant. So whether it be podcasts, whether it be asking AI itself, I say we have Claude and Copilot. I'll sometimes ask one or the other about the other one. and what are the new features of it and explain them to me. but I'm not sure I do keep fully up to date and fully up to speed. sometimes I discover something new and then that's a small thing, I don't profess to understand all of it. And I think that's one of the important things is not to pretend to be an expert and say actually this is trial and error. And as you discover things, Trying a couple of tools internally and I've sent them to the team and I've said, look at it and provide feedback and be critical about it. What doesn't it do that you thought it could do? Find the errors that it's generating because that's the human in the loop that we still need. Giac Soliman: For the record, I did not bribe Cathy to say those kind words about range, but it gives me immense joy to hear that it's being helpful for you. Speaking about, to your point, yes, you can't really expect to have all the ideas yourself. Sometimes you just need someone to inspire you. Oh, I'll give you another example. Someone in the community told me yesterday, because context management is the new, like it's a new buzzword, right? making sure that memory persists across sessions. Yes, of course, you can start a project in chat GPT or in cloud, but that's gonna, there's limitations how much memory. it's really the synthesis of the synthesis of the synthesis that gets stored in there. And someone yesterday told me, I have just a regular prompt that now, It fires every time I say a word, close this session, and automatically writes it. I simply say, put everything into a handover file, MD format of course, because AI burns less tokens with that. So next time I open a new session, in addition to the project, in addition to the temporary context and all that, it just reads the handover file. And there's one for every project. I'm like. Why didn't I think about that before? We tend to think, oh, AI, rocket science, we have to really squeeze our brain and to get on top of it. It's a simple thing. just someone mimicking what they do before AI, you a notebook, writing down, hand over, this is where I got today. This is I have to do tomorrow. And boom. The outputs I get from AI now are much better. It tends to hallucinate less because it remembers. more but what I Wanted it to remember from yesterday So I guess yes self-paced learning is great. Kathy Odds: I think, you and I both had time in consulting in our early careers and the ability to work with and I I want to say it was thirty odd people that I worked with in consulting, all doing similar things for clients. that learning ability that you had, hey, can you show me what you did for that client? Or hey, you did something really cool in Excel. How do I do that? Can you show me? I don't I'm not saying we don't have that in in-house. I think that it's a different learning experience. And having that consulting grounding and learning from each other was really, really valuable. And I do like I'm a comp professional, I'm gonna talk about Excel for a minute. we're all very comfortable with Excel. We love Excel. we love it's a coding language ultimately is what Excel is. I learnt my Excel while in consulting. I I'll be honest, they asked me, Did I know Excel in the interview? I said, Yep. And at the time I could probably do a sum and an average. And I thought that was knowing Excel. And then day one I'm asked to use some complex formula and I just went, Uh-huh. And then sat there for a bit wondering what now. Googled it, self-taught, had somebody show me, and then over time in consulting, and and I I know we worked in the same team, just at different points in time, there was a bit of a community around learning Excel from each other. The pivot point here is AI is no different to Excel. You just have to learn a different language to get it to do what you want it to do. So if we're talking about comp analytics, which AI can do assuming everything is good from a data perspective and sensitivity. It's a shift in mindset. So instead of being able to go in and tinker with your V lookups and your if formulas, you now have to retrain your brain to learn how to prompt your AI tool to generate the same outcome as what you can do in Excel. And I'm sure we can talk about data sensitivity, but assuming it's data that you're happy and allowed from a company perspective to put into to Claude or Copilot or whate whatever it might be, it has that capability to do analytics as quickly as Excel. But what I have to get my team and I and peers is we all need to be on that journey that we need to be comfortable not doing things in Excel anymore. We need to shift that. And I feel this is probably what people went through in the eighties when they were probably working with calculators and Excel came in as a tool and everybody's like, Excel's never gonna work for us. It's too much to learn And then, twenty, thirty years, we're all super comfortable in Excel. So I think we'll get there with AI as well. But that learning from others is critical to it. Giac Soliman: think people that used to do calculations manually in pieces of paper, once they've seen auto sum in Excel, it must have felt the same way when I asked Claude to draft a tensed light deck in this format with this brand palette and my god. I remember the first time I saw that, it was mind blowing. It does require a shift in mindset. and, that initial amazed feeling is just the beginning, but then it looks pretty much like this. goes from amazed and fear of missing out or getting stuck and then actually stuck and then back to recovery. And then, it's like a circle of emotions. And what makes it worse or better, depending on your point of view is that you mentioned being able to prompt the way I taught myself how to prompt the year ago is different to the way I'm telling myself. Is it exactly the opposite? A year ago, I've been told you have to be specific, be specific. And now people are telling me you have to be specific, but you have to make sure that your prompt of very succinct. Because if you are too specific, you're going to hinder the imagination ability. You're going to confine the AI into one guardrail, one lane. It's going to fail at what you want to do. you want it to be more of a sparing AI, I entered a sparing partner. So it's an ever evolving skill, which makes it challenging for someone that is more of a mindset. Oh, I want it to be done now. I want to be able to say I've earned it now. So it's a constant change and starts with a change in your mindset. Kathy Odds: Yes. And I think where people are comfortable, Giac Soliman: So. Kathy Odds: We do what we do in a certain way 'cause that's our level of comfort, that habit. And it might take ten steps to make a cup of coffee, but we're comfortable 'cause that's our habit and that's how we do it. And when somebody points out but if you put the coffee cups next to the coffee maker, you don't have to walk across the room to pick up the coffee cup and wouldn't that be quicker? I give that example, that was my parents. Giac Soliman: you Kathy Odds: If you had the coffee maker on one side of the kitchen, mugs on the other, and I was like, why don't you put it and my dad struggled to adjust when I moved the coffee cups because he was used to that habit and I think that's the same for us and when in our work lives, when we are busy and we've got a lot to deliver and there's lots of questions coming in and we're going through salary reviews or salary pay range refreshes or job architecture redesign. It's really hard to then at the same time also be in a period of change and retraining our brains to operate differently because it's like, you know what, I know how to do this with Excel. I'm gonna keep doing this with Excel. I don't I don't have the time and that that's change management. And I think everybody feels that way. And it is I think that's one thing that I struggle with and I try and carve out somewhere in my day. the administrative pieces of efficiency I use AI for every day, but I do try and carve out as often as I can 15 minutes, just try something. throw it in there if I don't like it doesn't matter, but I've at least thrown something in there and try to find that time to do different things so that when I do end up in a super busy season, that there's things that I've learned along the way. And rather than trying to tackle it every whole day assigned to trying to learn this, doing it in bite sized chunks is what works for me. Giac Soliman: I think... more and more people will be interested in neurology. the study of the brain, cause what I've noticed is that we know we can get stuff done much more than what we could in the past. you still have to, go to it today, but with different questions, we would need to go to a software engineer. We need to go to my IT department to get some stuff done and we'd need to wait. And then we would get pissed because, I need to wait two weeks. I'm just a drop in the ocean. Now there's so much that you can do with AI without actually having to rely or depend on other teams. But just because you can doesn't mean you should do everything at once. we're obsessed with productivity, AI will make us more productive. If you use AI the wrong way, it's actually gonna make you less productive. So back a few weeks ago, I would look at every day of the week the same. I'm going to do, I have to do 10 things. I'm to do a bit of every, a bit of all 10 every day. And so I had this constant switching going on that was, hindering my productivity badly. Now instead I keep one or two days, Monday and Tuesday for two specific tasks, Wednesday and Thursday for this. Friday is meeting free and it's for that. And I've noticed that how much I'm able to get done has increased dramatically, but I haven't changed the tools. I'm always using Claude and yes, in the past two weeks, there's been updates and new models, but Claude didn't change that drastically. What changed drastically is the way I use the AI tools and I decide, you how I decide to use my brain. And I was wondering if you... If there's something you noticed about your own behaviors, the way you behave with the eye, that it changed over past six months, not just the way you prompt, but also just the way you interact, or these rules that you put in place for yourself to be a better version of yourself. Kathy Odds: Wow, that's quite a question. I'm gonna go a bit beyond that to say that what I would observe in the past perhaps I'm not I'm gonna say seven years, which happens to coincide with my move to New York, but that's not the coincidence of it. I think in the past somewhere between the past five to eight years, the way that technology more generally has evolved, putting AI aside for a second, the expectations of multitasking and there's five apps sending me messages in different places at different times to keep up with all of that has evolved enormously. Again, I'm gonna go back to my very first, twenty years back when I started working. We had email, we had the internet. At some point I think I had a BlackBerry, but we didn't have smartphones. Well, I guess BlackBerry is a smartphone. And that felt like a lot at the time. If I think now it's everything that comes through my phone, everything that comes through my computer, there is a lot of expectation of multitasking. And then we introduced AI. And I think that's another layer that can help with that and there will be efficiency. But I think that in terms of how I operate with AI, is it making me more productive, less productive? I don't know yet. I think in some ways it's making me more productive because it has sped up certain things In particular if I'm thinking about anything that's communications, whether it be communicating with my team, communicating with business partners, communicating with the organization, I am more productive because I now spend the time reading and editing versus the time that I would take in thinking. So I think there's a productivity gain there, because I don't want to fall in the trap of just, let AI generate it and say that that's what it is. Right, I'm the human in the loop that's re reading through that. But I think in order to be productive with AI is that learning curve, at the beginning of a very steep learning curve of what it can do. And I think that's the The Jahari's window of the unknown. The stuff that I don't know, that I don't think anybody else yet knows. And until we've unlocked those items, where it the productivity gains are there in certain areas. I think in other places it slows me down because I think I want to get through this learning curve to make AI help me with this, but that's going to take time to do it. And so with my team, that's one of the things that I encourage is you have to try things. We're gonna, we share what people have tried, what's worked, what hasn't worked. Because from that we then say, okay, well this is the one we'd like this to be done by AI, to get there as a journey. Let's not try and replace the whole thing. Well some quick wins. But yeah, it's a good question on productivity and the overall how I interact with it. I think it evolves. I started with Copilot a couple months into this role, I then also got Claude and I don't I use both for different things and I think maybe that's the biggest shift is learning which one is gonna do what I want the most efficiently. Giac Soliman: it's Kathy Odds: And a good one. Giac Soliman: You don't always need to have the answer ready, right? that's the whole point to be able to admit that we're gonna take a stab at this a crack at this We don't have all the answers. But what's important, it's keeping that mind. It's focused on evolution and continuous learning We're getting to the end of the podcast so one on that note, I want to look at the future and I want you to think about one of the biggest headaches you had in your career in compensation. If you had the magic wand, what would you automatically fix with that wand? What's one thing that you said, I don't have to worry about that anymore because the way I design the eye is going to take care of that for me with minimal supervision. Kathy Odds: I think any of the more administrative tasks as it relates to compensation and career frameworks, job architecture, benchmarking, matching, job evaluation, I think that AI is going to make that very efficient so that the time can be spent on interpreting the insights from that information. What excites me even more is I think that I'd like to see it change what we do. the way that we've been benchmarking designing pay ranges, Those practices have been around for a long time and is that where the future is? I don't know, but I'm excited that actually AI might offer opportunities to really rethink compensation structures in a way that we can't fathom today, that starts to address some of the systemic issues that are being addressed by legislation in certain countries, pay equity, pay transparency legislation, because when legislation comes in, it's like somewhere something hasn't been done the right way and how could it be designed differently? and I'd love to have the answers as to what the future of, comp structures is. I don't, but I do think that AI is going to offer opportunities for us to think in ways that we don't yet think on compensation structures.