Subscribe
Share
Share
Embed
Harald’s Curious Corner is where curiosity meets connection.
Harald chases that question with a guest, gathers perspectives from voices across the industry, and then steps back to reflect on what it all means. The show unfolds like a story arc, part exploration, part roundtable, part reflection, blending imagination with analysis.
The result: trusted insights, meaningful conversations, and forward-looking takeaways that shine a light on where learning is headed next.
[00:00:00] Shellie Grieve: I think the big three frame is that learning won’t be a destination. Nobody will go to training like we know it today. The whole concept of leaving your work context to consume content will feel as dated as walking to the library to look something up. Things are going to look drastically differently. So I think the relationships between doing and learning collapses in one motion. You work and learning happens in the seams. It’s very much in the flow of work and it comes to you. A prompt surfaces at the right moment, a gap gets noticed and before you’ve even realized it.
[00:00:30] Harald Overaa: And that’s what’s on my mind. I’m Harald and this is Harald’s Curious Corner. Hey everyone. Thank you so much for joining this podcast. I’m delighted to have Shellie Grieve on. Shellie is leading learning in the AI era at ServiceNow, a company I really admire and one of the leaders in the SaaS space. She’s been there for 10 years and spent the last five in L&D. ServiceNow has obviously grown tremendously over the last couple of years, so really excited to dive into her thoughts about learning in the AI era, where all this is going, and also how you’re managing such a complex ecosystem. So thanks so much for joining your podcast, Shellie.
[00:01:13] Shellie Grieve: Thank you so much for having me, Harald. This is an absolute delight for me as well.
[00:01:17] Harald Overaa: Perfect. Thank you so much. I’ll get right into it, Shellie. So I looked at your title and it’s pretty interesting to me that it includes both learning innovation as well as emerging technology, which is good mix. How does that change the way you’re looking at the future compared to a traditional L&D leader?
[00:01:35] Shellie Grieve: I think I just sit in a really interesting intersection if you like in what I do. I sit in a very small but mighty team, our learning strategy and innovation team. And one of the unique vantage points that I have is both looking at the innovation that we’re doing in our day-to-day work, but also what’s coming down the pipe, if you like, in terms of technology innovation. And I think a couple of years ago that was very quite slow and we didn’t see a lot of transformation. It was more feature requests and occasionally you would get some sort of buzz, but now it’s just a never-ending, relentless cycle of innovation that’s coming from all fronts. And I think that the part that makes it interesting for me is that I’m looking at both the strategic execution of technology and what’s happening in the future for us, but also simultaneously looking at what’s happening the here and now, how do we keep the lights on and how do we keep doing those things?
[00:02:25] Shellie Grieve: And I think the other part is that our environment at ServiceNow is quite complex. We have an L&D team of upward to 700 people in our org and we service a very complex ecosystem. We have our partner and alliance channels, so all of our implementation partners that are implementing ServiceNow, have a huge ecosystem of customers and end users on our platform every day. And then we have 30,000 employees that we need to service and put all of their development and training needs at the front of our mind as well.
[00:02:55] Harald Overaa: Seems like the definition of juggling several things at the same time there, Shellie- More
[00:03:01] Shellie Grieve: Insanity.
[00:03:02] Harald Overaa: Yeah, absolutely. We were talking a little bit before this and you shared a phrase that really stuck with me, which is what’s called reinvention purgatory. Could you explain what that means to you and why so many organizations are stuck there? Also, things that are coming to mind is things like prompt libraries are usually pretty good, but they could ultimately be a dead end if AI tools are disconnected from enterprise systems and data and things like that.
[00:03:31] Shellie Grieve: Yeah, absolutely. So reinventing … I can’t even speak today. Reinvention purgatory, I think I’ve said that word too many times in the last couple of days, but reinvention purgatory was a phrase that I coined to describe that vantage point I think that I just said that I do in the role that I have. And what it really is, is that we have a lot of people within our organization that we’ve enabled with AI tools and generative AI tools and things like Claude and ChatGPT and Gemini, whatever your tool of choice is in your organization, we’ve given them these tools and they’re able to expedite on particular tasks as part of their workflow really, really fast. And then they will come to me and say, “Hey, I really need this feature on a particular application. Can we get a feature request in with a vendor?” And that innovation is really, really critical for them and their jobs and it’s going to help move the needle for them and make their day much more simpler.
[00:04:21] Shellie Grieve: And so I’m juggling that in one call and then I’ll literally get off and be on the next call and we’re thinking about what is the vision of learning and development for us at ServiceNow in the next couple of years and what does that look like? And we are fully all in that AI native world where we’re fully agentic, everything end-to-end has a lot of bots and agents and things happening. And that is also super important to keep our eyes on the prize and that work needs to start now as well. And so it’s that reinvention purgatory is that balance of needing to innovate in one conversation but completely reinvent in the next. And where do you choose the battles of where that small innovation that makes a difference to somebody here and now versus the long play of where you’re taking what would typically we would say digital transformation is, but in this case in the AI world, we’re really, truly inventing where we are today.
[00:05:10] Harald Overaa: Yeah, that’s resonating a lot. I would imagine there’s a lot of discussions and people having really strong opinions about where you should focus and what’s more important there.
[00:05:19] Shellie Grieve: Yeah, absolutely. And I think I’m seeing that in not just here I am, but in conversation I’m having with colleagues out in the industry as well. So I think it’s fairly common that this is kind of creating friction. There’s a lot of people that are very interested in moving things forward with AI, but there’s a lot of people that have been doing this craft, if you like, in the world of L&D for a very long time and it’s harder for them to see how that transformational reinvention looks like as well. So it really is a tension point.
[00:05:47] Harald Overaa: It’s almost the battle of what you want in AI, the future and kind of those aspects versus what you need from traditional systems. So every organization has got to be different there and the pace in which they’re going to that AI native or tearing down some of the structures will obviously differ as well. So you’ve authored ServiceNow first comprehensive learnings or L&D strategy for an AI native world, which is fascinating to me. And I’m very curious to understand what does AI native mean to you? And I guess what made you realize the old playbook of what has got us so far won’t suffice in the future? What made you realize the old playbook needed to go?
[00:06:29] Shellie Grieve: I think there’s elements of the old playbook that need to stay, but there is far more new elements that need to come into play. So for the last six months, I have been absolutely head down doing a lot of industry research and pulling things together in terms of how all of the different pockets of the teams that support our complex ecosystem, how they’re working, what we could change, what needs to change, where the industry is going. And now I have been authoring this strategy that feels like it’s a never-ending beast, but it is about reinventing workforce readiness. So it’s bigger than just learning and it allows us to focus on one of our North Star goals. So how do we get more learners into our ecosystem and ultimately driving product adoption for ServiceNow. So that’s kind of our North Star goal on that. And really what it is, is about building capability for what our world needs in ServiceNow and what the world is demanding right now, which is fast-paced AI skills.
[00:07:21] Shellie Grieve: So it’s really going from AI literacy to AI fluency across three million learners and how do we build that experience for them? And so what I’ve done is I built a really comprehensive strategy that touches on things across four pillars and those four pillars of our learning strategy is what we build, who we develop, what we offer and how we power it. And this is about building intelligent content, it’s about developing real capabilities, offering governed products that are tightly managed and powering it with intelligence. And that’s the fundamental shift. But outside of those four pillars, we have two fundamental horizontal pillars, if you like, running through this as well. One of those is obviously AI, but it’s AI for how we run our global learning and development team as a business, but also AI, how we teach that to our ecosystem and then how do we govern all of these things?
[00:08:09] Shellie Grieve: And that’s the throughput with our strategy. And probably the most fascinating part of this as we’re starting to do the walking tour and take this out to broader teams on this is realizing that the clock has reset on what we thought was a maturity model that was, you would increase in a point in your maturity model in six to 12 months if you were really focused on it and capping out at a maturity level of say five, that clock is completely reset and we’re starting all over again. So the way that we build our content is back at zero because we now have agents that can do that for us. We can atomize our content and have it dynamically reassembled and we can skills tag that and feed that back to our skills taxonomy and build that profile of our learner. So we also have in all of our strategy a concept called our talent signature, which is really our version of our skills taxonomy and our skills profile of all of our learners and how that is the engine that is really driving everything behind how we actually reinvent our learning experiences and how everything comes together for our end learners as well.
[00:09:13] Harald Overaa: Very interesting to hear. I could probably touch on this for many, many questions, but I’ll try to be succinct. You mentioned there regarding teaching AI skills to customers and partners. I think you’re in a unique position as a company evangelizing AI and also teaching it to yourself at the same time. How are you looking at the drink your own champagne idea and teaching yourself AI can essentially help the world understand AI better as well?
[00:09:45] Shellie Grieve: Yeah, absolutely. So we obviously sell AI products. Part of our vision and our strategy is to be customer zero on a lot of those things. So we have the concept of the AI control tower, which monitors and measures that. So we’re teaching people what that is so that they can purchase that product and use that in their own organizations. But we’re also doing that as well. I had a reminder email 10 minutes ago that I needed to update some statistics in there on some of the AI use cases that we have. So we are already doing that and our future vision of what we’re doing with learning does involve leveraging our platform to the extreme of what it’s capable of doing and being the conduit, if you like, of getting the information around. Because really when you think about what AI native actually means, it’s not AI assistant or AI enhanced.
[00:10:28] Shellie Grieve: It’s really AI being load bearing in what it’s doing, architecture, everything underneath it needs to fundamentally change. So we’re not just shipping courses anymore. We’re looking for signals across a myriad of systems and how do we bring that together? How do you translate that into a learning moment or a training and development need or does somebody even know what that looks like? How do you train those models to do that? How do you build your team of agents behind that to do that? And we’re very fortunate that we have a really powerful platform that can bring that to life for us as well.
[00:10:59] Harald Overaa: Yeah, absolutely. It’s a unique position to be in to use your own product for a lot of these things as well. I think it’s fascinating to understand when you’re writing this really deep strategy report, which is almost like a book in itself, it feels like it’s just almost published first internally and then taken externally. How did you go about doing that research? Was there specific people or publications you looked at? How did you start to try to untangle the current mess or excitement that’s going on?
[00:11:32] Shellie Grieve: I listen to a lot of podcasts. I read a lot of research reports. I am connected with a lot of people in the industry that have their finger on the pulse. And so all of those sources are cited in the strategy document itself, but it was looking at a lot of reports like looking at industry leaders like Josh Berson, but also looking at other Gartner reports and things like that and bringing them in together and being able to not take just one source of data and think that that’s how it is to be. It’s to look at all of those things because none of those reports know ServiceNow like we do. So it’s then taking that current state. So I literally scraped thousands and thousands of documents internally to build a very comprehensive picture of what the current state looks like across all of our pillars on all of our teams.
[00:12:17] Shellie Grieve: And then I looked at the research and what that was showing was the pathway and then looking also at where ServiceNow was going with our own product and what we could do on our own platform and sort of finding the middle ground and all of that. And there’s been thousands upon thousands of iterations of this. And thankfully I’ve had assistance with AI to help me build all of this and distill all of this and help do all of this analysis, but it really is looking at what are the things that we can carry forward that are working well for us. Maybe the way that we ship a product today isn’t the way that we ship it in the future, but the product itself does need to remain. And so it’s like looking at all of those things and thinking about the throughput of that and moving the conversation away from that focal point for many people and L&D has always been course completions.
[00:13:03] Shellie Grieve: How many people have finished the course? And it’s about stepping away and saying, okay, well, that’s one signal that tells a story, but here’s all these other signals that tell a story and how are we truly measuring business impact? How are we truly tracking what we’re doing and how does that actually impact the bottom line for ServiceNow as a business with product adoption, customers renewing their contracts with us and things like that. So I’ve put a lot of all of that thinking into the think tank, if you like, to be able to distill out a strategy that I really hope isn’t just shelfware. It’s something that every person that has read it to date has come back with, “This just makes sense. There’s so much in here that makes sense.” And it should because the strategy shouldn’t be something pivotal and new. It should be something that listens to the business at the pulse of where they’re at now and think about where that’s going in the future and then you need to join those execution steps.
[00:13:54] Shellie Grieve: So, okay, we’ve got a fascinating strategy here. How do we actually bring that to life? And so there’s been a lot of work at putting together the right artifacts and things for people to be able to go from, okay, well, here’s the strategy now, what’s next? And so it’s been always about making it something that’s tangible that the average person in our L&D team can wrap their arms around it and go, okay, I know what my role is here and now I know what my job is here and what I can do next.
[00:14:19] Harald Overaa: We’re essentially, as you’re describing as well, our identities becoming architects, architecting experiences, outputs, workflows. I essentially see three separate pillars converging, which is learning knowledge from agents and skills. How are you thinking about those three components and how those things are converging in your world?
[00:14:41] Shellie Grieve: I think that’s really important. I dance around all of those terms all day. It’s like skills and proficiency and capabilities and knowledge and learning and development and enablement. It changes depend for us, particularly depending on the audience that we’re dealing with here. All interchangeable works. But at the core of it, I think of things as what will move the needle in the future is that whether it’s a piece of knowledge or whether it’s a formal learning academy or a program that you are part of, I think we’ll see very much a pivot away to less than 10% will be that formalized learning and most of it will be things that you learn on the job in the flow of work. We’ve built a lot of capability and we’re working very hard on building out ambient assessment so that we can see what somebody is doing in their job and surface that little moment in time nugget if you like, or that little microlearning moment.
[00:15:29] Shellie Grieve: But not microlearning as we know today as a five-minute video, but really, truly, I see you’re stuck there. It’s a litle bit like bringing Clippy back, but in a whole new makeover and it’s a new person that you don’t recognize, but that’s what we’re trying to build. It’s like acknowledging all of these signals of what we can see that you’re doing, what you’re capable of, and then pushing you to your exact edge and just making you that little bit uncomfortable to lean into something that will help you grow your human potential there as well.
[00:15:54] Harald Overaa: From those perspectives, there’s some tensions here between the AI native learning teams and also aspects like governance and data and stuff like that. So if we start on the capabilities or skills, what do you think the capabilities or skills that will define an AI native learning team of the future will be like?
[00:16:18] Shellie Grieve: I think they need to be demonstratable. I think that’s the key thing. And I think that’s quite … I’m in a lot of conversations. I was in several meetings today where how do you measure those skills? How do you change it from being what we’ve historically known as propted exam, multi-choice questions and things like that? How do you pivot that away to that ambient assessment and beyond that to see what somebody is actually capable outside of the four walls of an exam center or, like I said, a quiz at the end of a class or within a SCORM package or something like that. And so we’re doing a lot of work on that, how we can build that out where we can validate a ServiceNow skill within our own platform by seeing the work that you’re doing. We can see that our implementation consultants that are out there building systems for our customer that have done maybe 10 or 20 implementations of a particular product, how do we validate their skills without them ever coming through the front door of the learning platform?
[00:17:11] Shellie Grieve: So it’s the combination of both. It’s the formal learning is there. The knowledge is there, you surface the knowledge to people behind the prompt bar or wherever they are in their flow of work, but then it’s also equally look at what they’re doing and validating. It’s almost like coming in and doing a practical job assessment and then saying, yes, we’ll put the stamp of approval on you. And that has to be an ongoing signal. That’s not a one and done activity. That’s an ongoing observation that we’re making. And so we’re building observer agents for those things and then you need to build the agent to manage what the observer agent is doing to make sure that it’s watching and that it’s continually growing and that what the observer agent is reporting back is delivering on the results. So it’s the team of agents and then it’s the team of agents monitoring the health of the agents to help bring that together and make that happen.
[00:17:58] Shellie Grieve: So I think that the skills and proficiency is a really interesting conversation. It’s why our talent signature that I mentioned before is a critical dependency on our forward motion of what we’re doing with our learning platform. Everything writes back to our talent signature, your HR information, your tenure, your learning experiences, the knowledge, your skills proficiency and you can tag all of those other things. So if somebody comes in an academy program, it’s tagged appropriately that writes to your profile, it demonstrates the proficiency level that you’re at, but you can also just do your job and know that you’re learning on the job and learning from your peers and that’s equally important as well.
[00:18:32] Harald Overaa: Feels like we’re almost going back to 70 / 2010 in the AI era where actually 10 is formal, whereas now it’s been way too much. It’s been like the 70% aspect of it. A lot of what we’re talking about here, Shellie, reminds me of a chat I had with Dr. Sandra Loughlin from EPAM where she essentially said skills is a data and an architecture equation. So a lot of those data point, let’s say how you’re measuring your skill is entirely context, team, company dependent. This is where I think organizational alignment and closeness from L&D or enablement to the business is so critical because essentially, otherwise a lot of these measurements are going to be sent from the top and people are like, “Yeah, this isn’t relevant to me at all. Why would I actually put people through this? “ Which then becomes really a mobility problem because if you don’t get that right, then the mobility of moving from one role to another almost becomes invalid because you don’t get the right skills.
[00:19:29] Shellie Grieve: Absolutely. And our talent signature, which we are working on at the moment, we have done that analysis and looked at the weighting of, because we have a lot more signals on an employee versus a customer, signal weighting of what we get from our employees is very different to what we get from our customers and our partners. Our customers and our partners, we are relentlessly responsible for enabling them on our ServiceNow platform, but we’re not directly responsible for everything to do with their career progression and their enablement in their roles. So we can only do so much there. And so those signals weigh differently to the employee where we have a lot more things coming in. And so I look at those as pieces of Lego that you’re building and sometimes you’re building the $5 packet of Lego. Sometimes you’re building the $30 packet of Lego, and then sometimes you’re building the Star Wars ship that’s $500 in the glass case.
[00:20:19] Shellie Grieve: And they’re the different types of journeys that everybody is going on in our ecosystem and we need to be able to cater for all of those. And every conversation that I’m in in that, it’s like opening a can of worms because you start, and it sounds very simple, but the more layers that you look at that. But I’m very fortunate that my team, our learning strategy team partners very closely with our skills team on this and our technology teams in building out talent signature for this. So we’re having a lot of conversations every single day and we’re learning stuff every single day on what that looks like and how to build that out and what that looks like even in six months, 12 months, et cetera. It’s a journey.
[00:20:53] Harald Overaa: Absolutely. Especially if you’re looking at this to the extended enterprise or non-employees. It’s a really fascinating puzzle. How do you measure skills, tasks, competencies for non-employees? So it must be almost a bit of a reality check whether you’re having an impact or actually if people are just doing the same mistakes they previously did.
[00:21:15] Shellie Grieve: Yeah. And we’re looking at things like that too. So how do you validate somebody’s mark somebody’s homework, if you like, of what they’re doing against best practices and how do you … It’s a bit like L&D. Used to be the way that we would build something and ship it as a SCORM package and you were doing the best thing in the industry. And now it’s a lot of different opinions about whether that’s right or wrong, but more leaning to very different in the AI world. And it’s the same when you’re validating these things now because we have people that maybe they went through training 10, 15 years ago and they’re out there doing this, but they’re applying the skills that they learned 15 years ago. They’re not applying current day technology rules and regulations. And we’re seeing a lot of divide in that just even in internal adoption of things like Claude and Claude code and stuff like that where people, you see old school developers that just want a hard program all day long.
[00:22:02] Shellie Grieve: And then you’ve got others that are really leaning into the acceleration and what you can do with something like Claude code. So it’s challenging and it’s why there’s nothing in any of these answers that is black and white. It’s a journey and it is that maturity model that I was talking about. It’s like how do you go from where you are today to the first step in the right direction and then how do you go to the next step and the next step? And each one of those steps seems to unravel something else and peel back the layer of the onions and then you realize that there’s something else there to think about. But there’s no playbook written for this AI native world, so we’re figuring it out as we go.
[00:22:37] Harald Overaa: Yeah, it’s into the unknown, right or unchartered waters, so to speak. I think one of the most interesting aspects of the promise of AI and what we can do is I guess what risk or what willingness does an organization and crucially their security and legal teams have for you to do those things. So one of the real tensions is innovation and then it’s kind of governance around AI and security. So obviously this is something you’re probably dealing with daily. So curious to understand how does that work in your world and how do you balance the speed versus let’s do this right elements of this?
[00:23:19] Shellie Grieve: Yeah, great question, Harald. That’s something that comes up far too often in our conversations. We have a very well-defined process on what it looks like to onboard a new vendor and take that through our security process. But with every new thing that comes out with AI and with agents, our security processes get longer and tighter and harder. And so there is the balance of that. So you might meet with a vendor. I had an example of this where we can, “Hey, we can get you onto a pilot in two weeks and we can show you results within the first 14 days.” Well, that’s great, but hold my BOY, just go over here and do the paperwork to bring those people on board. And it is about balancing that because the business need is there. Everybody’s asking us to move faster and do things. And the innovation that’s coming down the line from all of this is truly fascinating.
[00:24:07] Shellie Grieve: The demos that I get to see on a weekly basis, I would love to hand my credit card over for every one of them, but there’s a reality. When you’re in a corporate world, you need to protect your data. You need to make sure that things are being handled in the right way. We’ve seen huge adoption of AI tools here, but there’s some things that have taken a really long time to get in, but when they get in and the data’s in the right place and the security’s in the right way and it’s handled, then you’ve done that due diligence upfront, then you’re able to use it in the right ways. I think I often see posts and tips and tricks about do this and build these agents and do all of those things. That’s probably really great for a small business, but when you have a complexity, when we’re handling customer data, a complex partner ecosystem and our employee data and regulations and things that we need to adhere to, you do actually need to stop and take a deliberate pause because despite how much AI training that you can offer to people and how much coaching, people still don’t understand what’s happening with their data when they put it into all of the different tools that are out there.
[00:25:07] Shellie Grieve: And so we do need to be mindful of that. And that’s part of L&D’s role is even when we have that technology in, whether it’s in L&D or elsewhere in the organization, it’s an ongoing enablement on data privacy and sharing of data and what’s training models and what’s not and what’s right to share. And that’s something that we need to do mindfully as well.
[00:25:28] Harald Overaa: It’s almost like you need to take what I call a global approach, which is, okay, we have different AI tools that we’re rolling out globally, but if you’re doing AI coaching for customers in Europe where you’re going to take their data and make informed decisions about it, you’re going to get into problems with the EU’s AI act almost immediately. So if you’re doing some of those things, it needs to be opt-in or people need to agree to it or some of those aspects. So the idea you almost get stopped in your tracks because of this regulation and you almost just need to keep going into innovation because you can’t not do any of it. So it ends up certain countries or certain areas, you either need to be really careful of what you do or not do anything at all in certain AI aspects, whereas you can roll it out to other countries or regions simultaneously.
[00:26:20] Shellie Grieve: Yeah, I totally agree on that. We have things on our roadmap that the technology is there and we could do it today, but the timeline to implement that has been pushed out knowing exactly that, that people aren’t ready for that change and that it’s more legal conversations that we need to have. We could do some amazing things with the technology that’s here and now, but we have to jump through those hoops. That slows us down. It makes us do more intentional and deliberate work. But as I was saying before, being able to look at that ambient assessment and looking at somebody’s working, some people won’t like that. Some people will think that’s Big Brother, it’s very polarizing. And so with good things comes big responsibilities that we need to sit back and look at all of this as well and make sure that we’re doing all of the right things and that people have the ability to opt out and that we have to consider all of those different regional things that we need to comply with as well.
[00:27:10] Harald Overaa: And this is where I think almost my own public service announment or PSA is when a lot of people are talking about this in public spheres, being context aware that every organization and the speed that they can move is so different is so important because it’s almost like this feeling within L&D that I’m not doing enough or I’m not moving fast enough. And some of it is we could be doing more. But the other thing is you’re not allowed to move at that pace because of where you’re at. A lot of this is going to be held back not by the technology but by the governance and the security and the legal aspects of what we’re dealing with.
[00:27:48] Shellie Grieve: Yeah, absolutely. I spent a large part of my career working here in Australia in our federal government, and so I am acutely mindful of those things. And when we’re in conversations and we’re talking about our customers, they’re the people that come top of mind to me. Who is the most likely to resist this? I’ve been on projects with customers in my early days with ServiceNow where we actually implemented everything they asked for and we literally got down to a week before go live. Everybody was exciting, training was complete, announcements would happen and their security team came in and shut the project down. And that is something that sits with me all of the time when I’m thinking about the necessary evil of going through all of these sort of hoops that you have to jump through and making sure that we do right from the outset that we need to go through this.
[00:28:32] Shellie Grieve: I worked on a couple of pilots last week, which we did paid pilots with some vendors for two weeks and four weeks just to validate whether this was directionally correct for us and they involved five months of paperwork before we could do the two-week trial, but we need to make sure because we’ve got IP in here that we don’t want in the wrong hands, we’ve got customer data, so it’s important. It’s a necessary evil if you like.
[00:28:53] Harald Overaa: Absolutely. And let’s transition a little bit, Shellie, from, we talked about your journey as ServiceNow and how you’re building this L&D strategy for AI native era. We can almost transition a little bit into where’s the industry at and what are some hard truths about what’s happening on that road towards 2030 and reinventing L&D. So the first question I have for you regarding this is a lot of organizations are experimenting with AI tools, but FU seems to be scaling them. Why do you think that is? Is it some of the, I guess, lack of awareness of what’s possible? Is it some of the security and legal aspects or is it something entirely different?
[00:29:35] Shellie Grieve: I think some of those things are very true that the data security thing is one thing, but I think from my observations both where I’m working and the conversations I’m having in the industry, I keep seeing a theme emerge in all of this and it’s really that we are trying to move forward in a world that’s completely different to what it ever was before the architecture has changed. And so when I think about what I gave the Example earlier that I might have an instructional designer that comes up with an amazing prompt in Cord, for example, and that’s going to accelerate their journey. That’s wonderful, but really we need to step back and look at end-to-end what is the role of an instructional designer when you have technology that can do those things? How do we pivot and change those people’s skills? This is as much a human transformation as it is a technology transformation or reinvention because we are having to pivot people’s skills.
[00:30:26] Shellie Grieve: We have to pivot the infrastructure and the technology that we’ve always had. But when I look back to your question, when I look at what’s going on in the industry, there is a lot of things going on where I see tried and true vendors that have been around forever, that they are held back by legacy architecture that doesn’t allow them to move and change things because they’ve got to keep the lights on with existing customers and capabilities and things that are … It’s a mess to untangle and redo that. Then you’ve also got this emergence of startup companies that are able to move so fast because they haven’t got the handcuffs and the shackles on that are stopping them. And it’s like we’re able to work with those startups, but they don’t have the background and the security of what they’ve done before. They also don’t have the runs on the board with organization as complex as the one that I’m in.
[00:31:11] Shellie Grieve: So there’s a lot of different tangents going on, but I think that part of it is that we can’t apply what we did yesterday to what we’re doing tomorrow. And that’s really at the core of everything. Like I said, our humans need to adapt and change into this new world, but so does the technology, so does our business processes. You can’t know that if we continue to chip the course as a product, what it was before is changing now. The way that we package it together and ship it has fundamentally changed. The way that we can measure it, we’ve got better access to data, we can do different things, we can provide alternative experiences to people like it used to just be a one size fits all. And so all of those things combined have fundamentally changed everything about what we do. And that’s really why we have had to reset the maturity model back to zero again because you have a choice here, you can throw the baby out with the bathwater or you cannot.
[00:32:02] Shellie Grieve: And just to add one more point there, it’s also about thinking differently and being willing to go in that. But a lot of what we’re seeing also is bolt-on capabilities where a generative AI feature can get you a step in the right direction, but an agentic AI experience can get you the whole end-to-end workflow and just not those little band-aid fixes along the way. And so all of those come together for me and that’s what I’m seeing in the industry. That’s what I’m seeing where we are as well. We’ve got pockets that are moving very fast and we’ve got pockets that are really change resistant and moving very slow and tied up in legacy systems and things like that.
[00:32:36] Harald Overaa: When we’re thinking of changing roles and identity a little bit of learning and development or enablement or education teams, one of the things I’d really advocate for is let’s say you’re an instructional designer. As L&D people, we are trying to tell the business we need to upskill ourselves and change roles. I think a lot of that we need to do within our own teams as well. When you’re mentioning about learning systems and where things are going, I think fundamentally how our consuming information is changing. All of this is context dependent and user dependent, but there’s still a role for a really robust backend and structure of users, courses, information. How are those things from how do we upskill our own team to the fundamental changes in learning systems and skills architecture? Are you seeing similar things on your side?
[00:33:29] Shellie Grieve: Yeah, absolutely. We’re doing a lot of intentional things even from within my own team on how do we support those people that are in those roles. We’ve stood up just some things in recent weeks to look at. We’ve given our introductional designers a toolbox of AI tools. So what? That’s great. We’ve got some people leaning in, we’ve got some people that aren’t, but now we need to be more intentional about the transformation that they need to go under and agree that that’s more of an architectural role. We’ve got a very unique team of people that work in our environment too, that are highly technical curriculum developers. And so they’re building labs for our highest level certification courses. They’re building guided things that are different to somebody shipping a compliance course. So we’ve got a mixture of all of those and those people are prime candidates for that architectural shift that could be training agents to do more intentional learning work.
[00:34:20] Shellie Grieve: So there is that acknowledgement. We are doing some work around that, but it’s again, when you’ve got to keep the lights on and keep the content flowing when we’re shipping product at a rapid rate as well, how do you give meaningful time to those people to actually make that true transformation? So I think that’s important.
[00:34:37] Harald Overaa: Yeah, especially as this is all happening in hyper growth mode and you’re definitely not standing still. Thinking about that AI maturity model that you mentioned there and almost resetting that, what are the different steps you’re seeing on that AI maturity because you’re saying we’re resetting to zero. Let’s say we’re going from one to three or CO to three or zero to four. What are the steps along that journey and where do you think L&D teams get stuck when they’re trying to do those aspects?
[00:35:08] Shellie Grieve: I think part of the sticking is that not giving the time to do the transformation and being able to be meaningful on that. As I mentioned with our strategy, we’ve been very intentional about working out what the stepping stones are between those pathways and we’ve done a lot of detail on the next six to 12 months. So we’re already planning releases of what we’re doing for our systems for 12 months out now. We’re neck deep in that planning. And I think from that maturity model, some of the steps that we are doing is you need to know, like I mentioned before that I’ve done a lot of work to understand the current state. I’ve been afforded the luxury really needing to step back and say, here’s where we are and here’s where we want to get to. We know the destination, but we need to understand where we are on the map to know how to chart out a pathway forward.
[00:35:53] Shellie Grieve: And that has been hundreds of hours of work of me interviewing people and looking at processes and understanding where things are. And I think that’s really important work because it’s all very well and good to come in and say, we are going to be AI native and we’re going to reinvent everything about L&D, but you need to be mindful of the lights that you can turn off. You need to make deliberate choices of what you can stop doing and you need to have that sponsored from the executive leaders all the way down. And we can have our instructional designers tell us that they’re spending 10 hours a week on administrative things. Okay, we hear you. Let’s turn some of those things up, but we need to make those intentional choices. And that is all part of that maturity model. You’ve got to stop doing things the old way and you’ve got to simultaneously start setting those things up.
[00:36:36] Shellie Grieve: And back on my reinvention purgatory thing, we’ve got some things on our pathway as well where we are doing that iterative innovation very intentionally and keeping the flow going of what we’ve always done, but just slightly different and maybe a little bit faster and maybe a little bit better. But then separately to that, we’re also got the eyes on the prize of how do we do it completely differently and how do you run those two things in parallel?
[00:36:59] Harald Overaa: So let’s say for a second here, Shellie, that you wouldn’t have all the resources that you have within SN or I guess a luxury having a great product there. If you were to design something for Scratch and L&D team ready for the AI native era, or maybe you’re entering a team with fewer people, what are some of the things you’d be focusing on and what would you be doing differently than in the position where you’re currently at?
[00:37:26] Shellie Grieve: I think my answer to that would be dependent upon where I would find myself. But I think if you were starting from scratch, you’re actually in a much better position because you don’t have the legacy that you need to reframe and go forward with. And so I have jokingly said many times in meetings over the last little while, imagine if we could just build this from scratch where we didn’t have to think about the migration plan, the re-skilling of people, the data transformation, the thousands upon thousands of courses we have sitting in Scorm packages that we would love to train an AI model on. So there’s things like that where there’s a benefit to that. If you’re in a small team, you could be more mighty because you could build the agents to do things exactly as you want without the shackles of data security being as stringent as they are here.
[00:38:09] Shellie Grieve: But we are doing both of those things. It’s just that we have to move a little slower and more intentionally in those. So my advice to anybody in those small places or doing the startup, there is so much at your feet, do the research, understand what you need to get, but always have the goal. I have very central on my strategy by design at the top, what’s the north star that we’re always anchoring on and how are we going to get there is the steps below that. But we need to always have your eyes on the prize and think about that. The technology is there and it will come and you’ll find the right vendor or you will build the right thing or they will customize it for you, but you really need to be intentional about what you’re trying to achieve, the transformation or as I like to say, it’s not just transformation now, it’s reinvention.
[00:38:52] Shellie Grieve: You really need to be intentional about that and you really need to constantly come back to that helicopter view and say, “Here I am at the edge of the cliff. I’m ready to dive in and take what comes with this. “ And so with Great Power becomes great responsibility with that. But I think that the more that you can say yes to things and say no to things and being very intentional about that will serve the purpose there a lot as well.
[00:39:14] Harald Overaa: I often think about if you’re coming as a new organization or new leader and you’re looking at technology and the landscape, how would you start those, I guess, steps or research before you’re getting to vendors and demos and a lot of those things? I think the fundamentals for me is you need to really understand your organization and what are critical requirements we’re not diverging from versus these are some nice things. If you’re going to teams, it’s almost like they’re going to ask you to get everything because they want their own lives to be easier. And this is my approach. I probably say no to more projects than I say yes to because I don’t want to go on a journey where actually I don’t think I can help you. I’d rather send you where I think you can get good advice. And that kind of radical candor I think is definitely needed.
[00:40:05] Harald Overaa: But let’s say you don’t have those constraints. Maybe going for legacy systems if you don’t need that is also not a good fit. So I think you really need to have that homework done and really scrutinize that and being close to the business is going to be my recommendation there.
[00:40:21] Shellie Grieve: I was just going to say, it’s the shiny object thing. It would be no fewer than 10 times a week would somebody say to me, “Hey, have you seen this vendor?” Or, “Hey, this is really cool.” And I always stop and say, “Well, what’s the business requirement here? What’s the goal that you’re trying to do? “ And it’s the same way when you develop learning, you should really start with what is the needle that you’re shifting here? What’s the business impact of starting that up rather than, “Hey, we need a course.” Well, how do you know you need a course? What is the data that’s saying how are you going to know that this has made an impact? It’s really about having those intentional conversations upfront and some people don’t like that. They just want the course or they want the shiny object or whatever.
[00:40:59] Shellie Grieve: And I am very forthright in saying that, tell me the problem that you’re trying to solve rather than giving me the solution. And that’s part of our innovation workflow. We have an ideas portal where people can submit ideas and it’s the very first thing that we ask is like, what is the problem that you’re trying to solve or what is the opportunity that you see here for innovation to be able to solve rather than tell me the vendor that you went and saw and how much it is and we’ll just go and get it for you. So they’re very different conversations and when you have those conversations, it yields far better results.
[00:41:27] Harald Overaa: I think we almost need to scrutinize. What is the role of formal learning when AI can provide those immediate answers? And curious to get your take on that, Shellie, my own is our job is to fill it with the context-dependent knowledge sources and just fuel that engine rather than try to replace it and shoehorn everyone in to do a X amount of minutes course when that’s not going to be how we conduct learning in the future. So how do you see those things like what do we throw into fire versus what do we bring in?
[00:41:58] Shellie Grieve: I think that’s where skills validation and checking somebody’s proficiency is really important because I could have a hundred conversations with Gemini or Claude or ChatGPT or whatever. I’m learning so much because it’s giving me step-by-step and I’m seeing and I’m doing and I’m sharing screenshots and I’m doing those things, but I might not have ever taken a formal learning course. And I think that’s really important to know that the front door of the learning platform isn’t where everybody’s learning. We’ve always done on- the-job learning. We just now have that AI assistant or coach or tutor sitting at the door. I’ve been saying for a very long time, I remember when ChatGPT had the surge that it did when it hit the market and everybody was jumping over there, I said, “That’s our competitor there. The competitor isn’t the historical competitor that sells a similar product to us.
[00:42:44] Shellie Grieve: We’re now competing with that. “ And I did a bake off probably two years ago of doing a search for a simple thing. Like I said, how do I run this report? And our learning system couldn’t give us that and Google gave me a half answer on it. It’s much better now than what it was when I did that, but it directed me to a video on YouTube and eventually I found the answer deep within the video, but then I went to ChatGPT and asked the exact same question and gave me step-by-step instructions on how to do it and suggestions on how I could learn more with it. And so that’s what we need to know. People are so time poor that the prompt bar is where they’re asking and getting those answers and being able to learn and do that. And so we need to make sure that the knowledge that we have of our products, our company, everything that we’re doing, we need to make sure that that knowledge service is in the right moment and then be more intentional about when we do want people to do dedicated learning, the 10% or whatever number it is at this point in time that we build the human connections in those programs and we build the right meetings and we allow for situations where there’s peer-to-peer learning and peer feedback, that’s just as important as a formalized training course or a compliance course or that prompt bar that’s giving you that learning in the moment as well.
[00:43:55] Shellie Grieve: So I think it’s balance, it’s always balance and we need to put the right weighting on that and it comes back to that skill. How do we validate that somebody has increased that skill proficiency, their capabilities, they’re performing better, find those business metrics that support those arguments. And it’s then also being able to surface that knowledge, but then retain the history and know that you’ve already surfaced that knowledge so that now you know that the proficiency level or the expectation of where you can challenge that person is here or maybe they can only understand it in layman’s terms versus highly technical terms or they want to cut to the chase and get to the code or whatever that looks like, that needs to be part of the design of what you’re moving forward as well.
[00:44:33] Harald Overaa: Absolutely. As a final question here, Shellie, you talked about your AI strategy or the L&D strategy for the AI native era. If we’re having this conversation in 2030, what do you think would’ve changed most dramatically about how people learn at work?
[00:44:49] Shellie Grieve: I think first and foremost, I think we don’t know what 2030 might look like in terms of the technology is evolving so fast, but based on what we know and the trajectory that this is all going on, I think the big free frame is that learning won’t be a destination. Nobody will go to training like we know it today. The whole concept of leaving your work context to consume content will feel as dated as walking to the library to look something up. Things are going to look drastically differently. So I think the relationships between doing and learning collapses in one motion. Your work and learning happens in the seams. It’s very much in the flow of work and it comes to you, a prompt surfaces at the right moment, a gap gets noticed. And before you’ve even realized it, that’s actually been resolved for you or it’s been surfaced there where you can do that and it’s very personalized to you.
[00:45:32] Shellie Grieve: And I think that that’s what learning, enablement, development really is going to look like. And I think by 2030, organizations will have moved from measuring what people completed to actually understand what people can do. And so that’s what we’re doing with our talent signature and our intelligence layers that we’re building in our team of agents. It’s about evidence gathering, it’s about signals, it’s about taking that information and doing something intelligent with that rather than collecting completions or certificates.
[00:45:59] Harald Overaa: I think the honest answer on that, Shellie, is there’s things you’re going to have to do. Let’s say it’s compliance or taking training to ensure that you do that in an internal role. But if you don’t do this to keep up with people’s consumer habits, if it’s not as interesting to learn something as it is to go to YouTube or ask your LLM of choice, people are simply not going to do it. So the bar is just getting higher and higher so is the owners and L&D and enablement teams to keep up with that and ensure that we’re building products that people actually want to use. So the bar is higher, but sowers are tools that we could actually use. So it’ll be an exciting thing to follow here as we go along. Really enjoyed the episode, Shellie. Really fascinating to hear your thoughts about building an L&D team for the AI native era, some of the things you’re struggling with, the reinvention purgatory.
[00:46:51] Harald Overaa: I think a lot of people are going to find value from your perspective. So thank you so much for joining the podcast today.
[00:46:57] Shellie Grieve: Thank you very much for having me, Harald. This has been great fun.
[00:47:00] Harald Overaa: Likewise. Thank you so much. Cheers. So that’s it for this conversation. I’m going to keep going down rabbit holes as we explore the weird and wonderful world of learning and development. If you want my full wrap up and takeaways, subscribe to my newsletter where I’ll connect all the dots.