How AI is redefining the CSM role with Adam Parsons

Adam Parsons: I'm slightly biased, obviously, but I think that that should be close to 100% of people using AI in some form or other. That isn't me saying it, because I think you have to use AI so that you can say, I use AI in my role, because ultimately that's pointless. You should only use something if it actually has a purpose and delivers some sort of value to you. AI tools deliver value to every CSM in an element of their work.

Grace: Hey, Adam, thank you so much for joining me on the CS Google podcast. How are you doing?

Adam Parsons: Hi, Grace. Yeah, good to be here. Very well, thank you.

Grace: Brilliant. I'm so glad to have you on because we're going to be talking all about AI and customer success. Um, and if anyone's listening, thinking, surely everyone's talked about this, you're wrong. You haven't, you haven't listened to this conversation. Um, and there's a bit of a paradox. Obviously, AI is everywhere. Um, you know, it's reshaping how we think we write, we work. Uh, but obviously in customer success, you know, I think we need to dig into a little bit more about what that actually means. Um, but beyond just the LinkedIn, I don't know, the scare, the scaries that are put out there and, you know, making people think their jobs are going to be taken. So, um, let's just. Yeah, we want to kind of just get into that really. Um, but really, adoption is kind of, I think, perhaps slower than you might expect, which is something we're going to get into in a little bit. But, um, yeah, so kind of, I think this episode would be really interesting to hear what's holding people back and how, actually, how can CSMs properly use it and, you know, the ways they are currently using it and the, you know, the future ways that, you know, it can be harnessed. So, yeah, so I think, to unpack, Unpack that. Let me introduce you first of all. Uh, so today I'm enjoyed by Adam Parsons. He's a client. Client success director at Decreed, and, yeah, is a leader in the CS space and someone really kind of at the forefront of kind of integrating AI into customer success. Um, so, yeah, I'm really excited to have you on, Adam. So would you like to introduce yourself a bit about how you got into customer success and, uh, yeah, what drew you to the AI side of customer success as well?

Adam Parsons: Yeah, no problem at all. Um, I think my, my career history has been partially the constant pursuit of stuff that I find interesting at kind of every stage. So, um, I originally started out My career years ago working in recruitment. So I was actually a recruitment consultant for five years and I, I got to the point where I figured out that I actually hated sales but quite liked dealing with clients. So I started looking for jobs that that was part of. I didn't know that customer success was even a thing at that stage but I ended up working for an E learning vendor, um, as a client success manager. So at that point I was looking after 30 to 40 clients that were um, learning platform implementations, um, corporations and really working with them on how they were using the platform, whether they were getting the most out of it, et cetera. Spent a while in that company, ended up leading the team, um, moved into product development for a while because I became aware that client success often work quite closely with product and so I thought seems interesting. It also seems like even if it's not interesting long term it would be good for me to understand it. So I then spent a few years leading a product team for another um, edtech company. So working with instructional designers, learning technologists, that kind of thing and leading them as we rolled out across UK universities, then moved into degreed as you said earlier. Um, so now in a client success director role, um, looking after a small portfolio of um, sort of very large corporations generally sort of global multi billion dollar organizations, um, supporting the degreed product suite for them.

Grace: Brilliant. It sounds like you've had a hand in it all really. And um, yeah I think was there kind of um, a particular, a uh, particular sort of turning point for you that, where I don't know, was it in the last few years or was it before that where the value of AI and the real, the potential for customer success and, and wider. Obviously we talk about you touching on product development as well, that it kind of really click for you.

Adam Parsons: Yeah, it was relatively soon after ChatGPT became kind of mainstream back in 2022. I uh, I started using it just out of interest like uh, like I said, I'm kind of constantly attracted by shiny new things and I thought this is, this is something that's new. This seems like it's going to be pretty big. You know, I'll start figuring out what I can do with it. And then you know, that very quickly became okay, this is, this is a major thing that every person in every modern job is going to be impacted by. And this is before even you know, degreed or other organizations started sort of building out proprietary tools. We're just talking about the, you know, the consumer tools that were out at that time. So looking at like this is going to impact what I'm doing so I better get ahead of it and I like to where I can try and kind of future proof myself. So dug into it a little bit more in the sense that you know, clients success, the client success is the role that I'm doing at the moment. It's going to be changed by AI as is almost every other role. So if I can figure that out now, figure out how I can use it, how, how I can embed it and also not just how it becomes a tool that I use, but how it actually changes the nature of my role, then that will set me in good stead for the future.

Grace: M. So absolutely. So it's kind of a real sort of self preservation thing, you know, Uh, I mean that's how it's kind of how we're looking at it in marketing as well. So much of our roles are going to be automated at a certain point. Kind of getting ahead of that. Is that. Am I, am I right?

Adam Parsons: It's part of it, yeah. I mean the technology interests me just in general. Um, but it's. Yeah, part of it is, is preservation. Um, I think preservation sometimes can sound a bit negative like you're just trying to hold on to something that's dying. But I, I don't think that's in, in client success. I think it's something which is going to just change the nature of the role. So it'll go from being a ah, task based role to a insight based, relationship based, human skills based role. I think that's, that's how it will change.

Grace: Is that the kind of thing that really kind of excites you about it? This or the potential to sort of.

Adam Parsons: Yeah, I think it basically frees up time to do more of the interesting part of the role, doesn't it? Like uh, I don't think anyone gets into client success because they really enjoy doing progress reports or QBR decks. I think they do it because they, they want to actually see a difference in their clients organizations. They want to be able to understand what the tangible impact they're making is on their client's business and kind of being more. Being more consultative, being more strategy based, not just kind of a check in every now and then.

Grace: Yeah, absolutely. No, although I'm sure you will get, we will get the odd person who does really enjoy quite possibly.

Adam Parsons: Yeah, that's a different conversation. But yeah, I think you're right. Um, but yeah, yeah but it's something that I think has. Yeah, I've been thinking about it recently in terms of how CS engaged with AI, it's not just about replacing tasks that we don't want to do. It is about understanding how it actually augments the role in itself. Um, and the requirements of the role become different rather than just replacing parts of the responsibilities.

Grace: Yeah, you sound like you've completely got the right idea about it. You're sort of trying to think a few steps ahead of it rather than just being like, what am I doing with it today?

Adam Parsons: Yeah, exactly.

Grace: How is it making my life easier today rather than thinking actually, how will my role be in the next five years time? Well, it's quite hard to call that

Adam Parsons: one actually because it is a long time in our industry.

Grace: Rapidly. A year then. Even then. Yeah. Um, but yeah, I think that kind of like sort of talking about the individual level, um, because I don't know, a lot of, like I said at the beginning, a lot of there's so much potential. But um, many CSMs I think, well, lots of people in general, but a lot of CSMs aren't quite sure where to start because you know, I've uh, we've got this, we've got our annual state of Customer Success report coming out soon. And I think it was something like, uh, let me try and think about what the actual stat was. I wish I had prepared this, but it was about a third of see IT professionals do use AI. That was in 2013, that was in 2023. Sorry. Then I think in 2024 we saw a bit of an uptick. And then 2025, you expect what we're two, two full years since, I don't know, ChatGPT like we're talking about, you know, we've had a good two full years of uh, generative AI, but I think only it's less than half of CSMs are actually using tools like this. Um, and then I think we, I remember, I remember you know, writing the report, looking at the data, looking at responses. I think these use cases really range wildly. Like you've got, sure your content generation, um, and you've got some predictive analytics, but there seems to be quite a lot holding people back. Um, and I don't know whether that's, yeah, if you're a frontline csm, um, you know, you're not, you know, let's say you're frontline csm, um, you're not a data scientist, you are just trying to do your job better. What, um, do you think CSMs should be reaching for first to sort of better, better use AI today? More efficient to understand what their customers actually want, you know, what we want.

Adam Parsons: Yeah, I mean that, that statistic is, is slightly terrifying if that's accurate. Only 40 people, uh, 40% of people are, uh, using it right now. I Wonder if some CSMs may actually be using AI without realizing it now that it's built into all of our email clients. And a lot of the. Good point. Yeah, A lot of the CRM systems that CSUs have AI built in, I wonder if it's, it's that rather than them thinking I don't sit on chatgpt doing stuff day to day, but I mean, even so, I'm slightly biased obviously, but I think that that should be close to 100% of people using AI in some form or other. Um, and I think it really needs to be. Not, that isn't me saying it because I think you have to use AI so that you can say I use AI in my role because ultimately that's pointless. You should only use something if it actually has a purpose and delivers some sort of value to you. But that said, I think AI, excuse me, I think AI tools deliver value to every CSM in an element of their work. There's even the, I mean the most simple use case you could possibly get in AI is the content generation piece that you mentioned. So almost all email clients now have a, an AI generation element to them. So you can speed up the process of writing emails to clients. That's not to say that you should click generate email and then hit send, because as we all know, you can spot AI generated content a mile off and it completely undermines the human nature of the CSM role that I was talking about just a minute ago. It basically just replaces you as a, as a robot that the client's interacting with, which is not what you want at all.

Grace: Yeah, the copywriter in me as well. Sorry, absolutely hates that.

Adam Parsons: Exactly.

Grace: There's so many formulaic, uh, it becomes just so formulaic. You can spot it a mile off if you're used to it anyway.

Adam Parsons: Yeah, 100%. Um, but I think CSMs should be leaning into, leaning into those where they need them. And there are use cases for those. There are some cases where, you know, if you're writing an email to a client that you know, well, the AI generation is not going to help you because only you know that client, the AI doesn't. It's going to generate an email based on the content of the previous one or something like that. Whereas you're emailing, emailing a person that you know Forget about that and just write it yourself, it'll be quicker. Um, but I think when it comes to getting to know your clients, which is the, the thing you asked a second ago, there's all kinds of tools that you can use to collate information from across the web to really get to understand your client businesses. And I think that's, you know, if you're, if you're a CSM that's dealing with more than say 10 to 15 clients, you can't possibly retain all of the knowledge in your head of what's going on in all those client businesses day to day, especially if they're large clients that have new information happening on a day to day basis, you know, turnover in their teams, change in their strategies, etc. Um, and so tools like ChatGPT, Perplexity, Google's NotebookLM, all of those sorts of tools can be used to collate all of that in a way that means you don't have to sit researching clients for hours on end. You can be going into a call with a new stakeholder, uh, you can run a deep research project on Perplexity to do a landscape, um, review of your client's industry. You can get a report back with a list of. Here are the things going on in your client's industry, here are the things, things that might impact the service or software that you provide to them. Here are the talking points you should think about when you go into that call and you've done it in ten minutes. Um, and it just sets you up for a much more constructive call because otherwise you go into, you know, these kind of new stakeholder intros and it's just a conversation about, oh, who are you, who am I, where have we worked before? What are our priorities? Whereas if you can go in and say, okay, I know that this new regulation is going to be impacting your supply chain at this point, which means that your teams that work in that area need to be upskilled within the next two weeks so that you don't lose money in this area. It becomes value add from day one, which, you know, builds you in a much better position as a, as a CSM for those clients.

Grace: Absolutely. And I think what we're sort of touching on is like a question of perhaps AI literacy. Um, you know, you, what you, from what I thought, correct me if I'm wrong, from what you're just what you just said, then it sounds like research should really be a really White House. CSMs are kind of using AI, so it's sort of or Would you say not so, you know, you don't replace it? Don't replace Google, for example?

Adam Parsons: No, I think you can replace Google. Um, I think that's fine. Um, I mean we're starting to see in, you know, even in um, kind of articles and things coming out over the last few weeks, you know, perplexity is increasingly coming up against Google as a better solution because you can do the search and then you can query the content of the search within the same tool. Um, so I, I think research, I think news and I think analysis are three key areas, uh, for csm. So research, as I just mentioned, getting to understand what's going on not just in your client's organization, but in their entire industry. Um, a lot of organizations allocate CSM portfolios on an industry basis as well. So if you're having that type of research project in many cases, it will then impact or be relevant to a lot of your clients if you work across a single vertical or a couple of verticals. Um, I think analysis is quite good for me personally because all of the clients that I deal with generally are kind of publicly listed companies. They produce things like annual reports, earnings reports, ESG statements, all those sorts of things that they have to. What I generally do with those is I just download all of the PDFs, which are usually hundreds of pages long, stick them into NotebookLM, um, and then I can use the voice mode to have a conversation with NotebookLM and ask it for things that would be relevant based on my relationship with that client organization. And you know, how can I take this point and bring it into a conversation about upskilling or, um, workforce transformation that they might be going through? Or what does this thing in the ESG report mean for how they do this particular part of their business in the next couple of years? Um, so I find voice mode particularly useful there because it's a backward and forward conversation rather than just typing queries into a prompt.

Grace: That is genius. I mean, I'm quite late to the game on NotebookLM, but it's such a useful tool to catalog, uh, a lot of different sources and it doesn't tend to, um, hallucinate in the same way I think ChatGPT does.

Adam Parsons: No, no, I mean Notebook LM is, is just processing other information. It's not really as much coming up with its own information. Um, the really handy thing is as well as putting your own sources in. So like I say, with the annual reports etc, you can actually just type in the keywords. Um, so you know, for me, I Might type in the client's name and then it will bring up open web sources based on its own search and you can approve or decline those sources as well. So usually it'll come up with things like the company's website or their, you know, their Forbes or Bloomberg page or whatever might have happened for them. And you can include that um, into the, the sources it pulls from as well. Which is very useful.

Grace: That is useful. Um, I hope anyone listening has like got a pet. Well, I say pen and paper. That sounds so. Okay, but writing that down because that is some seriously hot tips there. Um, but so interesting because I think, I think, yeah, ah, from an. We're just talking about an individual, an individual basis. These uh, tools are really empowering. But when we were chatting about this subject a few weeks ago and about, you know, what angle are we going to take for this episode? Because there's another side of it as well. Like when actual companies build AI into the fabric of their organization, something we're trying to do as well, uh, at the alliance, you know, part of Customer Success Collective. But when actual companies build it in and enable uh, their employees to use it, that kind of takes and takes things into a whole different sort of takes into a whole entire stratosphere. So um, yeah, could you kind of talk to a little bit more about how companies and particularly how csorgs can uh, build uh, this into their processes, you know, with internal tools or.

Adam Parsons: Yeah, yeah, definitely. I mean we're at degreed, we're quite fortunate because we're a learning company. Um, AI fits very nicely with learning, um, especially when it comes down to how you can personalize learning for individuals. So um, probably two examples. One is just um, chatgpt. So we have a team who have built an internal tool on ChatGPT that has access to all of our knowledge center documentation, understands our product, um, our services, et cetera. And we can use that internally to do things like saying, I'm going into a meeting with client X, um, I know that this is their current business objective. Can you give me a feature map of everything within our product that could support this use case for that organization? So then I get a handy table out, ah, the back of that where it'll go through all of the kind of business uses and it will say if you want to achieve this outcome, use this feature within our product suite. And I can use that as a talk track for a meeting with a client to say, have you used this part of the system? Have you considered introducing this or doing this slightly differently? Um, so again it's about doing things that I could go through and prepare manually for, but doing that would probably take a day for one of those meetings. Whereas using that internal GPT takes five minutes, um, to get at least a framework that helps us to provide talking points for a conversation.

Grace: So interesting. And do you think, um, I mean this is not something else we sort of, we learned in our report recently. Um, not trying to, not literally trying to plug. I just found it, I just found it quite interesting. But there's like um, I think like less than half of customer success teams we uh, we surveyed, less than half have a dedicated OPS role, CS OPS role. But do you think, well, what do you make of that in the sense that does. It opens the door to more custom AI tools being m, you know, more widely managed. Um, you know, if we're seeing, you know, year on year we're seeing an increase in ops, the OPS function. Do you think there's a possibility to get that kind of uh, sort of team wide kind of use of uh, AI internally?

Adam Parsons: Yeah. So interestingly, the tool that I was just talking about was actually not um, built by csops person at our company.

Grace: Oh really?

Adam Parsons: Okay. It was actually built by someone in sales who felt like it was a good idea and wanted to see if they could do it. And it turns out they could. Um, so that was then rolled out fully. So I think to your question, yeah, if you've got a csops role in place, great, utilize them, get them to look into it. Csops is going to be a key part of this. But if you haven't, that doesn't mean you can't look at these tools. I think it's about the curiosity, it's about spotting an opportunity to say okay, this could be done in a better way or this could be automated or this could save X number of hours per person per week. Therefore I'm going to find a way of doing it. The tool that I mentioned, as I say it's only, it's built in ChatGPT, so it's not, you know, it doesn't require specific loads of development input or anything like that. But if I were to put an estimate on it, it probably saves me four or five hours a week, um, you know, which is half a day. I'm getting back of things that I would have otherwise been doing. You then scale that up to however many people you've got in your CS team. You can put a, you know, a pound sign or a dollar amount next to that and it's, it's pretty significant. Over a year.

Grace: Massive. Massive. Absolutely. It just, I know probably can't talk too much about it. It's quite, you know, magic's quite early days, but how long did it take to sort of build this kind of this, this your own custom product like that?

Adam Parsons: I mean, they were doing it in their spare time. Um, and so I think it probably took a couple of weeks to kind of get it, you know, get everything together. There's obviously, um, you know, privacy considerations that have to be made if you're giving access to things like internal documentation. So it had to be a localized version to make sure that everything was still as secure, secure as it needed to be. But, you know, the framework is there. OpenAI make it quite easy to build on their, on their framework. So you can, you can start, you know, with a few YouTube videos and tutorials, building your own stuff out like that beyond just, you know, the, the UI custom GPTs that you can build.

Grace: That's amazing. I mean, it's just there's so, there's so much possible. Um, um, yeah, absolutely, yeah. I mean, kind of, um, looking forward at like, I don't know, the future of, you know, so at the moment I think, yeah, research we've done and research I've seen, you know, on other sources, it's like very few teams are automating. Um, you expect the number to be higher. I think, you know, we recently found out something like less than 20, but you know, let's hoping, you know, say that doubles. How do you think the CSM role evolves? Um, because it's a, it's a very unique role. It is. Um, but how does it kind of remain that sort of special kind of human. I'm trying not, I sound like I've got rose tinted glasses on the special human role. But, you know, it is very, Is it a very, it's very unique role? What, how does that, how does that role change?

Adam Parsons: Yeah, I think, I guess it's two parts. It goes back to what I was saying about in the short term, AI will just replace tasks that people are doing with a quicker way of doing those tasks or mean that they won't have to do it. So, um, that just means that CSMs can get more stuff done. Um, but I don't think that is the end state of AI within cs. I think that's just based on what you can do right now using consumer tools or if you've got, you know, proprietary tools built into a CRM or things like that. It's just making the access to and analysis of data easier. Um, so that for things like QBR decks, stuff like that. Ah, we will get to a point where you can create a QBR deck by clicking on generate deck and it will pull in all the data from your CRM, from your analytics software, whatever it is that you use internally. So that will go away, you won't be spending hours doing that anymore. Um, lifecycle emails like you're coming up for renewal. All those kind of standard milestone things can be fully automated now. Um, so there's no reason someone needs to be doing any of that manually. Um, the thing that I think will change as a result of that is that the CS role will become more consultative and more strategic. So it will end up focusing more on, um, demonstrating to a client how they can meet their business objectives as a result of using whatever that solution is. So whether that's an isolated solution because it's a niche product, or whether it's a wholesale organizational transformation, it doesn't really matter. But it will become more about as a csm, how do I consult with the client to say, you know, basically why did you buy from us? What are you trying to do? And how in a year's time are we going to demonstrate to you that it, that you are closer to that goal than you are now? Um, because that's ultimately how you get renewals, how you reduce churn. Um, you know, it's about keeping clients and expanding clients. Right. So if you can get to a position of fully understanding the link between what they have now and where they're trying to get, you can look at what the gap is. You can also look at what the data is that will illustrate whether that gap is closing or not over time. And it might be, if I think about some of my clients, for example, they're running through broad scale learning programs and when you first talk to them, they might say, okay, we want to make sure that people are upskilled in this area for the future. And you say, okay, well why do you want to do that? Well, because the role is going to be changing. We think the nature of the industry is going to be changing. Okay, so how are you going to know if you've done that? Um, well, we're going to have more people that are upskilled and we, we're going to have better succession planning, etc. Like, okay, so what does that mean for the business? Like, well, we need to reduce hires, we're going to reduce external hires, we're going to increase career mobility. Okay, what does that mean for the business? Right. It means it costs less to hire people because we're spending less on external costs. And then you start to get to, okay, right, so how do you put a dollar amount to that? So if you can reduce your external hiring by 10% as a result of using this, that's X million euros per, uh, year that you're saving. So now you can go back to your shareholders, stakeholders, whoever, and say, by implementing this solution, we're saving X million euros per year as long as you can make that link. So that's where I think the CS role really starts to deliver value of helping clients see that journey, um, and helping them look at, you know, the bigger picture of the overall impact and the overall business change that they're bringing about.

Grace: Oh, fantastic. I mean, that sounds like something. It sounds like the ideal of CS at the moment. But from what. Obviously you are the expert here. That's not being properly realized at the moment because people, I guess, being stuck in the weeds a bit too much with, you know, things that perhaps could be automated.

Adam Parsons: Yeah, I think people are. People are used to the way that they're working and so they carry on doing it. I think also, to be honest, AI overwhelm, you know, that is a real thing. Um, I subscribe to a couple of AI newsletters that I get daily. And every single day there are seven or eight major news stories about some big AI revelation. Like, that's a lot to keep track of. So I think if you're doing a role where you're not, you know, you don't have any kind of responsibility over AI solutions or things like that. It's very easy just to feel like you're behind even after a week, you know, um, and so it then puts you off thinking, you know, I can't start anything because I don't know enough. Everything's moving too quickly. I'm going to get left behind and you sort of get stuck in, you know, in your own head of just, I'll just carry on doing things the way I've always done them.

Grace: No, completely. And on the overwhelm, we've got another assisted community called AI Accelerator Institute. AI AI. And my God, the. Just the constant developments that are. Uh. My colleague. My colleague, Marissa. Yes. She's the content lead for that. And she's like, I just can't.

Adam Parsons: Just can't.

Grace: I cannot keep up. I'm not, I'm not. It's so. I completely get that. Um,

Adam Parsons: you don't need to keep up all the time either. Like as ah, someone working in cs. You don't need to know about everything that's happening in AI, you just need to know about the stuff that matters for your role. And I think to make it a little bit meta, you can actually use AI to help you with that. You can set a task in ChatGPT or a task in Perplexity or something like that and you can say look, here's my role. You can even go as far as copy and paste. Here's my job description. Give me a daily digest of everything in AI that matters based on what I'm doing. And ah, you can personalize it to yourself. You don't need to know about new developer tools that Google have released. If you're not a developer, it's never going to be helpful. It's just going to be noise that distracts from the genuine value that's seriously smart.

Grace: I think that could that applicable for any role though. I think that's fantastic. Yeah, um, I'm definitely going to have a look at that as well. Um, but I think also just uh, to draw back to what we're talking about, the sort of literacy and the sort of AI kind of overwhelm, um, do you reckon there's a little as a small part of people that kind of have um, concerns over, I don't know, ethical concerns maybe? Um, do you think that you think a little bit of that turns into it or do you think is that not something you've really experienced that much? Because obviously we talk particularly for talking about, you know, internal built tools and stuff or, or just the general, I don't know, roles. Yeah, roles may be being reduced. Do you think there's a, there's a little bit of that or is that just a bit scare mongering?

Adam Parsons: No, I think it's a genuine concern for people. I think um, you know, people don't want to think that they do a job that can be replaced and I think that is a perfectly valid concern. And I think people can then say, you know, ethically, should organizations be pursuing replacement of Starfire AI? That's not an answer I can give one way or other. I think everyone will have a different opinion. But I think for people in client success it's about future proofing yourself to say not I think my job's going to be replaced, but actually how can I continue to add value? And also what does value mean in the future where uh, AI is everywhere compared to what it means now? Because now delivering an effective QBR meeting might be valuable for some of my clients in five years time, I don't think that's going to be a thing. I think it will be completely different. And so the job for people in CS is to, is to predict the future of what is going to be valuable to businesses based on how the nature of business is going to change over the next five years and how do we make sure that we are able to continue delivering that value.

Grace: I completely agree, I completely agree. And um, yeah, touched on some really, yeah, really significant points there.

Adam Parsons: But I think on the, on the overwhelm point, you know, I guess to sort of think about how do you, how do you get out of that? Because you can get very stuck in that. I think just pick a problem, pick one problem to start with. Of, you know, what do I find, um, what do I find annoying in my role? What do I think sounds like it could be automated. You know, what's, what's taking me longer than I think it should. Um, and start really at just that hyperlocal level. You don't need to think about how do I impact all of my clients using revolutionary AI. It's just I find writing this weekly progress report annoying. Can I automate it using AI somewhere? Probably yes. It's about removing friction from the stuff that you're doing. Um, and so I think at the most basic level, just think through where do you allocate and where do you spend most of your time? Is it on things that are genuinely adding value to clients or is it on process based tasks? M. If you spend a lot of time on process based tasks, those are probably the easiest to look at. Automating to start with. And you don't need to get hung up on looking for a particular tool up front or thinking I ought to be using this solution. It's just about what problem have I got to solve? And then go to what are the options of how I can solve it?

Grace: I mean, Adam, you seem seriously switched on about this, um, about integrating AI more. But do you think, and obviously it's everyone's response for themselves. But do you think there's, do you, in your experience on conversations with peers, do you think enough people, managers and team leads and leaders, CS leaders are kind of enabling their sort of their team to, to do that. Like, you know, maybe some people, uh, if people listen to this podcast, they'll, they'll know. But do you think sort of people, managers and sort of leaders are kind of setting that standard at the moment, or do you think there's still quite a lot of internal training to do. And managers might not know that they should be suggesting this to their staff.

Adam Parsons: Ah, I think where we are at the moment in the sort of AI revolution, AI takeover, whatever you want to call it, you know, the stage we're at at the moment, I think everyone needs constant upskilling because like I said, it's changing so fast. If you have some internal training next month, that's going to be completely different. Um, you know, there'll be new things that you can start doing. I mean, two months ago, the example that I gave with NotebookLM wasn't really anywhere near as good, and so I didn't do it as much. Now it is because it's changed. So I think managers need to be thinking, firstly, from the perspective of how do I develop my team to ensure that as their line manager, I have some level of responsibility for their career progression and their development. So I need to make sure that I'm giving them the tools that they need to flourish in their role and into the future. But secondly, just as a, as a team lead, you have a responsibility to make sure that the clients in your overall portfolio are receiving the best level of service they can. And ultimately that can be improved by using AI tools. So I think you have a responsibility to at least do the. At least do the work up front to say, are there areas that can be improved? There might be. I've yet to find one, to be honest, but I'm happy to be proven wrong in it. But there might be some industries where CS teams will look at it and go, okay, there is genuinely no area of our role where AI can contribute or improve the service. And if that's, if that is honestly the case, then fair enough, don't bother. Um, but I don't think that is the case. I think there are always areas at the moment where for CS professionals, there are things that can reduce the time you spend on stuff. There are things that can make understanding your role, understanding the data you're seeing, understanding the insights you're giving your clients, easier and better for you to present to them.

Grace: Definitely. No, I completely agree. Um. Oh, such an interesting conversation. Um, I think sort of drawing to close though, and sort of summarizing everything and synthesizing what we've spoken about for somebody. Um, I don't know whether they're an entry level CSM or perhaps they're at the sort of start of their as. As, sorry, AI journey. Um, do you think. How do you think they should approach that? You know, I think what we've. I think what's this, the thread throughout this conversation? It's. I think it seems to be going through, going into it with purpose as well as curiosity. I mean, there's a good to be. It's good to be curious about these things, but I think, you know, it's getting to that point now where it does matter. Yeah, it's not going anywhere. It does matter how you integrate this and when and how. So. Yeah. What's the kind of. Yeah. What's that kind of mind. Mindset. Mindset shift people really kind of need to make and. Yeah. What, what would your advice be for somebody?

Adam Parsons: I, I think to. To get started? Um, yes. Step one is pick a problem, um, that, that you want to solve. Don't try and just overhaul everything, or don't try and just bring in AI so that you can sound like you're on the cutting edge of something, when actually you might end up solving nothing and introducing more complexity for no reason. You pick something that's actually going to be fixable. Um, I think for individuals that are using consumer AI tools like Gemini, Copilot, ChatGPT, whatever they are, the mindset shift is to not treat it as just a really good search engine. Treat it as a colleague. I, um, can't remember the exact number, but it's something like AI has an IQ of around 150 at the moment, which is the equivalent of kind of a very intelligent person, but that's an IQ of 150 in every single topic known to mankind. Um, so it's an expert level in everything. So you can be having a conversation with it like you would a colleague. And if it says something you don't understand, or if you want to probe or if you want to say, okay, but what does that mean for my role? You can say that you're not just. It's not just a data exchange, it is a conversation that you can have. Um, and then to your point as well, I think being intentional about it is important. You know, run those small experiments. If they don't improve what you were hoping they would improve, stop doing them. You know, kind of learn. Learn to fail quickly and replace with something else. If the initial problem that you were trying to solve doesn't get solved because the solution's not right, stop using it. Pick a different problem. Or if the problem is the tool, you know, you can go back to researching that. But I think curiosity and intentionality have to go together there.

Grace: Yeah. Oh, completely agree. And, um, yeah, very, very eloquently put. Um, yeah. So I think there's so much to unpack here and I think there's so much that I'm actually also going to take away. Think about how I'm sort of, yeah, applying to my role as well, which is not CS for anyone who's listening, it's marketing. But yeah, really, really interesting conversation here, Adam, and thank you so much for coming onto the podcast to talk about this. Um, I, I anticipate. Yeah, we, I mean, we've probably barely scratched the surface here, but I anticipate listeners probably will have questions and perhaps will, you know, almost certainly want to follow up with you. Um, where is the best place for, for anybody to continue the conversation with you and, you know, learn a bit more what you're doing, uh, particularly within your teams as well and how. Yeah. What they can learn from you.

Adam Parsons: Yeah, um, you can find me on LinkedIn, so LinkedIn. Adamlparsons. Um, I also publish a newsletter there called Artificially Intelligent, which is about how CS teams can implement, um, AI tools and how it impacts what we're doing day to day.

Grace: Brilliant. I'll link those in the show notes and people can access it when, uh, the episode goes live. But, yeah, brilliant. Thank you so much. And, um, yeah, thank you so much for being fantastic guest and a seriously interesting conversation. Something that's so important to today.

Adam Parsons: No problem. Thank you.

Grace: Thanks for tuning in. Be sure to check out our other episodes and go to customersuccesscollective.com for even more customer success related content. You can also join our global community on Slack and can find the link to that on the CSC website. But until then, see you next time.