HR,We have a Problem: From Skills to Outcomes: Where AI in HR Is Actually Going

[00:00:01] [SPEAKER_00] Welcome to the The HR Huddle, presented by Sapient Insights Group, the ultimate resource for all things HR. It's time to get in the huddle.

[00:00:21] [SPEAKER_02] Welcome back to the HR Huddle Podcast for HR We Have a Problem. Another part in this series that we've been talking about, we have a very special guest. All the other guests I told them they're not special. You're the first one. Anna, could you introduce yourself?

[00:00:36] [SPEAKER_01] Yeah, sure. My name is Anna Goldberg. I'm VP of AI Product at Cornerstone, which means I lead all of our AI products and product strategy here at Cornerstone.

[00:00:46] [SPEAKER_02] Amazing. Just that job title is not something that would have existed, what, five years ago?

[00:00:51] [SPEAKER_01] No, actually it just existed for me very recently.

[00:00:54] [SPEAKER_02] That's a good spot to be in.

[00:00:55] [SPEAKER_01] It's a good spot to be in. It's a whole new sort of area and I think it's interesting, like even when we think about the HR space, how we look at what that even means is changing so fast.

[00:01:08] [SPEAKER_02] Yeah, I mean, what are the questions you're getting from customers? Because you're dealing with that a lot, right? Yeah. Because they're trying to make sense of it, I assume.

[00:01:15] [SPEAKER_01] I think what I've seen is that customers were so eager to get AI at first and they were like really chomping at the bit. They were super excited. And so Cornerstone, in addition to all of our friends, introduced a lot of AI features into our products, but then it was like AI feature for features.

[00:01:33] [SPEAKER_02] Yes.

[00:01:34] [SPEAKER_01] So the questions that I get from customers now are like, what can you actually solve or what outcomes can these AI features bring that will add value to us, which is different than like AI for AI sake?

[00:01:46] [SPEAKER_02] Yes. And what are you seeing? Can you think of any examples? Obviously, nothing that's, you know, not allowed by legal. I think we've already talked about that.

[00:01:54] [SPEAKER_01] Yeah, yeah, yeah. Of course. I always get in trouble for that. I see a couple of things. The first is about how AI can help you solve that skills problem, right? Yes. So how can we like initially say like, it's always been super hard to start with skills. It's always been a project. And it's always been the thing that customers are like most scared of. Like if you even look at our own adoption numbers from our customers, skills is one of the lower adopted products because to get started and really do it right.

[00:02:21] [SPEAKER_01] You traditionally have needed tons of buy-in from everyone across your business. You've needed to then get employees to go enter their skills. So it's been super tough. And what we see is AI is an accelerator for that. And customers are really latching on to that idea. And we're super excited about that as well, because that really lays at the data foundation that can enable you to do the next pieces of the conversation, which are outcomes about not necessarily agents for workflows, but agents for outcomes is what we're hearing from our customers.

[00:02:50] [SPEAKER_01] So things like how can you make onboarding go faster? How can you, my favorite one recently was like with some consulting customers of ours, how can you define who's the right consultant to assign to a project or a role and then know their skills gap? Again, that skills is that foundation and upskill them agentically based on that skills gap to get ready for that project, to ensure that project and customer is successful.

[00:03:17] [SPEAKER_02] That's so fascinating. I, before I had a fake job, I had a real one and had to work in HR. And it is interesting because I think about those conversations, right? It was what we were trying to do, you know, like for onboarding, you know, wasn't like, oh, we need to get onboarding done 10% faster. That's just not how you would talk, right? You say, what we need to do is have some ability to figure out what we have and what is lacking and how to fix that and solve that gap.

[00:03:40] [SPEAKER_01] That's part of it. But also then you have retention issues. Like we hear from some of our customers that, especially in healthcare, that retention in healthcare is so tough, right? And so like if your onboarding program takes three months, by that three month mark, like a large percentage of that onboarding class is already left, especially in lower like entry level healthcare jobs. So imagine if we can then tailor and upscale more quickly with onboarding to make it meet that need in the right way and reduce that onboarding time.

[00:04:10] [SPEAKER_01] And therefore meaning you're not having to go through that cycle over and over and over again.

[00:04:15] [SPEAKER_02] Yeah, that's fascinating. I think it was Vincent that drew the parallel, Vincent Bellevaux, that drew the parallel to the sort of original days when you would go to an apprenticeship. And it's the last conversation I had, we kind of found this thing with everything that's old is new again, everything new is old again. But this is what we're really talking about is learning while at work, figuring out exactly what's needed and having this happen using all the skills we've learned in the past,

[00:04:43] [SPEAKER_02] all the knowledge that we've learned in the past, but still bringing in the new tech.

[00:04:47] [SPEAKER_01] Yeah. I mean, you know me, I've been in this industry for like probably 15 years now. Since you were 10. Since I was 10. Right. Yeah, exactly. And watching us go through this transformation and like thinking about what was possible before and then really taking it to what's possible now because of AI has been the most exciting part for me. Yeah. So when I, for those of you,

[00:05:11] [SPEAKER_01] for people who don't know, which may be on the podcast, many don't. But you know, before I was in product, I was a customer success manager. And before that, I was a customer myself. And the hardest part, and I know your guys' data says this as well, was getting HR a seat at the table.

[00:05:28] [SPEAKER_02] Oh yeah.

[00:05:28] [SPEAKER_01] And like proving that your data was relevant and important and that what HR was doing was so important to the business. And one of the most amazing things now is now we can actually, because of the agents and the skills based underneath, we can actually tie what HR is doing to real business outcomes. Love that. Cost savings. And to me, that's like the holy grail of like how we can tie the amazing work we do to develop people, bring them up to their full potential

[00:05:56] [SPEAKER_01] in HR and tie it to business outcomes that drive at the C-suite level is what I'm super interested in. And I think what AI can really unlock.

[00:06:06] [SPEAKER_02] Yeah. And you know, so it seems like this sort of complex issue and it is, but you guys have come up with a wonderful framework, the dancing with the stars.

[00:06:15] [SPEAKER_01] Dancing with DWTS. Yeah. I shouldn't have said that.

[00:06:17] [SPEAKER_02] It sounds like it's going to be something that you shouldn't look up on Urban Dictionary. I was like, well, what's DWTS? What's this something?

[00:06:23] [SPEAKER_01] So DWTS is not dancing with the stars, but I wish it was sometimes. It is the domain work task skills framework. And what it does is it allows us to contextualize who you are as an organization within the larger labor market, and then use that to build the data foundation to power the rest of your learning and talent experiences towards those outcomes. So domain, literally, I was just talking to a higher ed customer upstairs, like domain literally means the domain. Where do you work?

[00:06:52] [SPEAKER_01] Get that initial context because a job title in higher ed is going to be different than a job title in oil and gas or somewhere else. Right. So that initial piece, then the work, the person actually does from their job title and description. Then from there, we can infer tasks and skills based on labor market data. Because the fact is, is that my job title, whether you add AI or not, has pretty much been VP of product for a little while now. Right. And, but the things I do

[00:07:20] [SPEAKER_01] underneath the covers have changed. So my day to day is different. The tasks I do are different, which means that I need to develop tasks and skills for other jobs. And just by looking at the org chart, you wouldn't see that. Right. But by getting to that granular level of tasks and skills,

[00:07:36] [SPEAKER_02] you can really see that change happen. Yeah. Again, it's one of those things that has been talked about forever. Forever. And it's wild to see it actually being put into place. Yeah. What are you seeing? I mean, you're, you're, you're getting the behind the scenes, obviously, working here at Cornerstone, but what are you seeing as sort of the most exciting thing coming out right now? Obviously what you've got going on, but like, is there something you've noticed that customers really latching onto or something you feel like they're going to take away and start implementing right away? There's so many things.

[00:08:07] [SPEAKER_01] Um, I think the ability to, there's a couple first is like the ability to connect agent to agent. Oh yeah. So every customer I talk to, they are like, we love what you're doing. Can you put it in our co-pilot? Can you put it in our Gemini? And the idea that you would have to go through an actual training before, but can now actually like surface what you need and co-pilot don't have to go to the learning management system or don't have to go to learning. That's like the first thing they're

[00:08:36] [SPEAKER_01] really excited about. Then what I'm seeing, and I love to see it is customers are really getting excited about becoming a skills-based organization because they're seeing the easy switch. Yeah. Right. So when you say like for the agents to be better, for those recommendations to be better, you have to start with a found a strong data foundation. They're like, Oh, we've not been able to do this before. And now we can do this a lot easier. And they, like, it's really cool to see that happen. I think in a lot of ways that's already happened in Europe and we're catching up

[00:09:06] [SPEAKER_01] here. And so it's really cool to see us catch up in that way. Yeah. I really love that. I think,

[00:09:12] [SPEAKER_02] and not just because it's a reflective of our data, but we know we have shown that, that AI already existed in organizations, right? They were already using these tools. And it seemed like a lot of organizations said, well, how do we stop them from doing that and use our tools? And instead he said, no, we'll just connect to it. We get it. You're using those tools. We have our systems and those can all work together to be transformative and take the fear out

[00:09:33] [SPEAKER_01] of it. And your data also shows, right? That if it's just like you're trying to implement AI without that strong data foundation, it falls down, right? And is less effective. And I see, I think our customers are, once they hear the story, they're seeing that too. And they see that

[00:09:47] [SPEAKER_02] they need that foundation to start the journey. There was an offhand comment yesterday when Stacey was speaking at the customer advisory board where someone said, yeah, it's garbage in, garbage out. It's been the problem forever. Forever. And now it's that people even in HR need to understand that.

[00:10:03] [SPEAKER_01] And they are. They are. It's actually like the first question we get with our customers when we're talking about the agents is they'll be like, well, what if our data's bad? Right. That's the first question. And then it's like, okay, we're here to help you make that better. Here's what we can do

[00:10:17] [SPEAKER_02] collectively to take you there. I love it. So that's what's going on right now. Any predictions? I mean, you've got the AI in your title now, Siri. You've got to be a prediction business. For a few years from now, what do you think the conversation will be like as we're sitting here?

[00:10:35] [SPEAKER_01] One of the things that getting that strong data foundation allows you to do is to think about what replaces, what AI replaces and what you maintain even with the change. And so I think there's some interesting things in the future about making sure your agents are compliant, for example, or how we workforce plan differently. Like when we think about build by borrow bot for strategic workforce planning are some interesting things in the future that having this strong

[00:11:03] [SPEAKER_01] data foundation that the AI can help us leapfrog with will unlock in a new way that will allow organizations to plan and think better. And then also think about where their people go and where they can be most effective. So I am a, you know, this, I'm like a strong proponent of human in the loop. And then I think there's a lot of things in HR that won't be replaced with AI, but I look forward to the days when a lot of our HR folks can be freed up from the day-to-day tasks because their jobs are

[00:11:30] [SPEAKER_01] augmented by AI to really work on the more strategic pieces that matter to the business.

[00:11:36] [SPEAKER_02] And speaking of that augmentation, you know, we're in the, you can't see it at home, but we are actually in the middle of the floor. Maybe you can hear it of all of the different partners that are here as well. There's some augmentation you guys have with your partners as well, where you can find those areas where maybe Cornerstone can help hand it off. Are there any specific examples you can give or some areas where you think that those partnerships will become critical?

[00:12:03] [SPEAKER_01] I think one place, and I, you know, I don't think they're here, so, but that's okay. I'm really excited about our partnership with AgentForce because it will put the Cornerstone capabilities directly into Slack where people work. And I just think that that ability of using the Slack MCPs to drive experiences across will really change how we do business for not only the folks in HR, but the folks who we serve

[00:12:29] [SPEAKER_02] every day. Yeah. Yeah. I love this. And again, I'm saying this at a place of love because I'm an ex-HR person, or I guess still technically, but it is still fascinating to me when I hear HR people talking about data governance and MCPs. I was talking to a few people at the dinner event last night. Yeah. We're customers. And hearing them talk about headless architecture and MCPs, I was like, you're in HR? You understand these words?

[00:12:52] [SPEAKER_01] It comes up now in like every customer conversation. I would say like two months ago, no one who we talked to knew about three months maybe knew what MCP was. And now it's like, oh, well, can you embed this into my co-pilot? Can you embed this into Slack? And so it's now becoming more and more of a prerequisite than anything else, which is one of the things that with Cornerstone's non-jealous, like open architecture makes it so different because we're willing to sort of go make those connections with whoever wants to play in the sandbox with us.

[00:13:21] [SPEAKER_02] Yeah, exactly. It's still that, you know, for so much of this technology discussion, and this is sort of a recurring theme, it still keeps coming back to the humans that are there and not forced. It's not like, oh, we should get a human in the loop, but it's that, no, that was never not going to happen. You know, there was some nervousness at first. And I think people have now gone, okay, that we're, we know we need people for so many of these things. And it's more a matter of helping them understand and trust and maybe take a little of the fear out of what's a fairly exciting time.

[00:13:51] [SPEAKER_01] Yeah. And I think that's why, I know now I'm like a broken record, but the data is so important, right? Because then it gives you that audit trail to take out the fear also.

[00:14:00] [SPEAKER_02] Yeah. This whole podcast sponsored by data. Thank you to our sponsors. Yeah. Thank you, data. The core concept of data has been a great partner with us. Anything that you personally are hyped for? I know that you've got plenty of meetings to go to, but I would love to hear what, what gets someone that's so invested in the AI world hyped for the coming months?

[00:14:25] [SPEAKER_01] I am just super excited about the potential it unlocks in our people. I mean, I have, I'm, maybe I'm unique in that I'm not in the tech business for tech. I'm in it because I'm, especially I work in HR tech because I think it is the great unlock in how we create a more equal workforce. It's where people can develop and grow their skills. And it's how people improve their

[00:14:53] [SPEAKER_01] economic livelihoods, right? By growing and learning at work. And that's why I do what I do. And so when I think about what AI can unlock, I think about how we can better target training and upselling for those folks to give them more potential in their next step. And I'm super excited about that. So I just think about like what we always are like, Oh, the robots are going to eat our face. Right. But like what I'm excited about is the human side of it and how it really unlocks

[00:15:18] [SPEAKER_01] our ability to open up human potential to more folks and let them grow in different ways.

[00:15:23] [SPEAKER_02] Yeah, that is wonderful. Well, you have helped us all grow in life potential just by sharing your knowledge with us. Thank you for taking the time to speak with us today.

[00:15:32] [SPEAKER_01] Yeah. Thanks for having me.

[00:15:34] Thank you.