welcome to the AIHI project i'm Alex Alonzo Sherm's chief data and analytics officer thanks for joining us today this week we're going behind the scenes of how HR and business tech analytics are formulated today we'll be unpacking the process behind identifying trends and creating impactful predictions before offering up some AI trends and predictions ourselves our guest this week is Zachary Sherto senior research manager of employee experience for global tech market intelligence provider IDC welcome to the AIHI project Zach thanks Alex it's great to be here thank you so much for spending the time with us obviously being the head of research and thought leadership at Cherm you can imagine how excited I was when I saw this on my calendar and thought to myself boy there are 90 different ways I want to take this but before I do that I want to give you a chance to kind of introduce yourself and share a little bit about your professional journey and kind of what inspired you to specialize in HR and employee experience strategies oh it's a good question i often joke with people I fell backwards into the HR arena um my original background actually is in civil engineering out of Miguel University but I graduated into the '08 financial crisis so between project management and transportation for the city of Boston and interning for Kronos those project management skills drove me into work for Kronos for 4 years before I got my first in as an industry analyst working for a smaller boutique firm in Boston that gave me a lot of exposure to the field of vendors and to the transformation that was going on in the HR tech arena to understand how people in technology were coming together i got a lot of exposure to strategy to the digital environment and just to the world of how people were being enabled to work and that led me to ultimately do my MPA and in on the policy side looking at the nexus of labor management and industrial development where I've been ever since it's been a 16-year career as an analyst and I wouldn't trade any of it for a minute that's awesome personally I'm also impressed by McGill i I will share that I have a cousin who's a faculty member there uh so I'm I'm always uh rooting for McGill as one of Canada's finest importers of talent uh so that being said uh let's talk a little bit about some of the things that you're seeing some of the things the trends that are coming up right uh many organizations look to analysts to understand emerging trends can you outline the process that you put forth in identifying and validating trends especially as it relates to HR and AI technologies because that could be up in the air oh yeah i mean it's a combination of both the digital markers on the buyer side and the behavioral markers on the adoption side so at IDC we take a very data oriented approach as our the middle letter in our name will suggest uh to understanding the market that leans into both buyer audiences while drawing on a very closed triangle of outgoing convers and ongoing conversations with vendors providers invest and investors before joining IDC I worked for other analyst firms that built similar databacked approaches from different angles including ROI and survey based evaluation and leading up to IDC I learned and trained on several statistical modeling methodologies that have really benefited the analysis at IDC so the modeling helps us run leaner market surveys during the year that include an annual all HCM survey where we really look at a host of specialized submarkets as well as two additional surveys and employee experience and talent acquisition we also run our SAS path analysis to uncover buyer trends surrounding enterprise applications as well as our future of enterprise resilience surveys that run up to eight times per year to look at the mark at market impact trends for digital and services spend co-odeling between these allows me to relate a lot of the responses and emerging trends across the different survey findings to really home in on what's driving buyer behavior what's unnerving them what's exciting them what's limiting them across different remitts and when it comes to HR the tricky part is is that yeah the CHRO will own a remmit or the different leaders within the subm markets for HR will own their spaces but in the traditional hierarchy of executive leaders the CHRO is often at the bottom of the totem pole so we have to look at the other impact factors like pressures on IT spend what the CFO wants to see what operational resourcing looks like to figure out how employees are being enabled and how that connects into the different trends on the HR side that allows us in partnership with our service services and software tracker teams to look at how market share is changing across the vendor landscape and also to look at our five-year forecast that we publish every year oh it's fascinating because some of what you're describing in my in my world is the large unstructured data problem right we've got a variety of different data that are pointing to a variety of different trends and it's really about making heads or tails and putting that into uh if I could be my inner Taylor Mason for a moment the the algo right how do I put that into my algorithm so what types of data and metrics or si signals uh do you find are kind of the the the worthwhile ones the ones that tell you something's worth pursuing versus something is likely to just fade or is noise in the system so after years of looking at survey data and buyer trends while engaging in all of that active listening that we do during the year um looking at direct customer engagements and all the exposure we get at client events and conferences i'll admit some of the trends just kind of leap out at me out of the data sets when we get them back right away it's kind of that one of those weird savant skills that when you're looking at data for so long eventually the important stuff just kind of leaps off the page i kind of joke that the important data points smell purple but conversations of vendor inquiries drive a lot of the conversations around what the market's seeking to understand but really it's the buyers that need me a lot to understand gaps within and driven by their interests their behaviors and their concerns so there's no better source of guidance than the actual buyer but I also read a lot into broader market and investor analyses while paying very close attention to external institutional data around how economic and labor markets are shifting all of that gives me a big broad perspective on what to expect from people movements and from the challenges that businesses will have tapping into talent market intelligence and it's really about leaning into how individual behavior shifts among consumers and employers and employees rather who are really one and the same just starts to help me keep tabs thematically on where I should direct my attention when it comes to the KPIs every survey every conversation I have leans into a set of business KPIs not always the HR ones i lean into how performance has changed in the last assessment cycle how productivity measures if a company measures it has changed what their model is how their revenue has changed customer satisfaction customer retention and we do and I do a lot of modeling to connect some of all of the different um items that we ask about in surveys all the different things we listen for back to those changes and KPIs to understand their impact sometimes they're primary and we get a direct relational factor sometimes they're secondary or tertiary and we have to really dig into how those relationships are oriented um but it really always comes back to getting a handle on the behavior of the market first before you know where you're going to look for the signal and the noise it's fascinating i think about that and I I think to myself so much of the uh the way you describe that is the right way the big data the the data that really stands out those trends they do smell purple or you know they stand out in some way that you you immediately want to do that i'm gonna I'm gonna quote you on that in the future with full attribution of course uh so in thinking about this right we're part of what we're doing on this uh podcast is thinking about AI tools in particular how have AI tools changed the nature of your work and changed the accuracy or the methods by which you you know identify trends predict kind of build algorithms how have they done that i can tell you that AI can be a blessing and a curse and I think that whether you're on the buyer side the investor side the analyst side practitioner wherever you are on it I think that statement is more is more universal than we might think on the one hand when we talk to buyers and vendors and investors it's helped all three of them access information more quickly and improve some of the accuracy of how they engage with us when we ask them for information so there's the intake side on that front while I never really go as far as to demand or openly question you know voluntary participation in our research um it it has made all three of those groups the buyers the vendors and the investors a lot more confident and their ability to answer and engage and that's made it easier to put them at ease when they work with us on the other hand AI models also need to be examined and assessed almost as much as raw data so I still like to see if I can outrun the computer on the analysis side while I will put the the insights and trends into any you know internal gated systems that we might use for analysis and model generation I'm still running those models to the best of my ability both to keep my skills sharp and to to test what I'm seeing against what the computer's putting out because we're still in an era of AI modeling that you need to test it you need to know how the bread is baked um got to be able to do the math manually before you do it in the calculator kind of thing because we also have to make sure that we're looking at not just how on point a trend is but also whether or not that trend is causal or can actually be truly correlative um we have to be careful about that when we're reporting on it so I would say it's definitely improved accuracy and output it's definitely led us to to check on things faster it's definitely enabled me to validate whether or not the signals I'm seeing are the ones that I should be paying attention to but I still do a lot of the math you know either by hand or you know aided by some of the old fashioned tools whether it's Excel or or other modeling languages just to make sure that everything is coming out with the right level assessment so that begs the question though has the AI ever seen something that Zach see um I've experimented with building my own internal um assessment tools off of some of the the public engines that are out there that I can just kind of gate it to my own research um sometimes I prove the computer wrong and mind you every time I do it I go back and and check three four five times to make sure because you know you're only as good as what you know and you don't know everything um and the computer has access to a lot more information and and modeling techniques than you know I do in my brain as an individual um but yeah but I'm going back through it all always to make sure I've caught the computer sometimes sometimes the computer's caught me and sometimes the computer gets it right leads me to the right signal and I take the model the rest of the way very good so what advice would you give a business and think of every business right the small business to the really large businesses the startup all the way to the the big enterprises what what advice do you give them when it comes to kind of leveraging HR and AI analytics for predictions uh and doing it effectively what what are the golden rules that you you kind of laid out you started with you know always questioning and always doing the math yourself but what what else would you throw out there i think the first thing is a golden rule of life is question everything uh the day yeah the day we stop questioning things is the day society grinds to a standstill and we all become the same or start achieving to the same um you know diminishing or top of the diminishing returns curve and it's the day everything just grinds to a halt so question everything is given just because a computer spit it out you have to know the data that went into it you have to know the process and methodology it used and you have to test the outcome because computers are only as good as what we feed them um so no matter how sophisticated the modeling capabilities are you just you have to question everything the second part is go at your own pace for smaller organizations you know good news you're you're in a prime spot to be able to go from some manual processes is you're in a position at least on the HR side where potentially your HR teams can know and interact and see most of the workforce on a daily basis even if it's over a remote environment so your HR teams probably know you know at least half of your workforce by name that means that some of those human insights you can rely on them for a for a certain time and you can dip your toe in AI data enablement performance enablement you know or structure modeling whatever tool it is you want to use you can really do it at an experimental level and not worry about going all in all at once um human human nimleness is still an advantage that small organizations have for mediumsiz for for SMBs you got to think about which category you're in when it comes to diving into AI transformation if you are a organization that embarked on digital transformation as needed over the last 10 12 years you're going to have some of your data environment that you got to clean up first before you can go full-fledged into any kind of AI enablement or system consolidation if you were latent in digital transformation you could actually consider bypassing the whole DX phase alto together and go straight into AI centralized tool enablement because the vendor is going to be all too happy to take your manual data sets scrub them put them into the system and consolidate you on day one so within the SMB market we're actually seeing folks that were latent with digital transformation leapfrog ahead into AI transformation it's kind of a very funky phenomenon when you get into the large organization it gets a whole lot messier you get a bunch a couple of companies that are that have very very siloed or disperate systems either within just within the company you know across national operations or even globally that are you know that are looking at deploying AI capabilities provided by their vendors for the part of the organization that they're using that vendor my recommendation really is take a comprehensive look and chart a strategy out either with a consultative or SI partner to look at your entire global organizational data model see where consolidation can really benefit you and then get into the vendor conversation around AI insights because with the data that the enterprise organizations sit on is one of the ones is one of the data sets that I'm really excited about um a lot of the large vendors serving enterprise organizations are finally at a point where we can start to look at performance as an outcome which means you have to relate a lot of subjective and qualitative data across a lot of different systems so take a look at that consolidation build a strategy out first and then dip your toe into some of the more sophisticated um AI models and and AI insights enablement because if you get that data that data set construction right on day one it's just going to be so much more powerful on the back end you know it's fascinating because you you started off with I think probably the best golden rule which is to question everything and at the same time you also kind of caveed that by saying go at your pace um not to not to ignore the other things but it takes me back to grad school and one of my adviserss who always said "Think about your analytic plan and always remember that your eyes should never be bigger than your stomach right your appetite should for for a really interesting finding should never be bigger than what the data the data you're ingesting is old kind of analogy u I think about that especially as it relates to the work that you do or the work that I do and it does happen very often it's frightening how much people really do want to go too fast or think broadly or overemphasize a hypothesis in their brain right and and and so uh I I always think back to those kind of opportunities and then I I like how you rounded that out with the notion of understand the amount of data that you're generating and what it is that you have and what you sit on because oftentimes you forget about how all those things can relate even if they're not technically related in some way or statistically related well and that's the big thing for for AI you know we at IEC we taxonomize AI into you know corei modeling that's at the center of the ecosystem and the whole unified data play we also then get into the operational AI the functional side when you get into assistants advisors and agents and with all three of those yeah you've got to be careful and stay on top of the data that's going into them you've got to be mind you know mindful about the purpose that they're supposed to serve what is it they're supposed to put out you know assistants are about insights advisors are about you know human-led action and agents are about autonomous action um and in all cases they are designed to pull together both the related and the unrelated data which means you got to be really on top of your governance to be able to get to a sophisticated level of using them cleanliness compliance and governance are so paramount for for buyers and for acceptable and equitable use cases before you can get anywhere with any any you know any AI enable yeah and it's that kind of conditional relationship that we don't always think about we're so focused on that correlational or that causal relationship but sometimes it's that layer that is the conditional components i you know you raised an excellent point there u so it's fascinating we've been talking about a lot of data and especially the analytics that you do is as it relates to HR techch and more importantly what you've seen over the last 16 years as you described uh selfishly I sit on the Sherm Labs in strategic investment committee and uh I was the initial executive sponsor for Sherm Labs i selfishly I care about what's going on in the world of HR tech one of the things that strikes me is I I I'd love to get your take on what it is that you've seen over the past 5 to 10 years uh in the world of HR tech and and what do you think has been it the the top two or three kind of milestones that that relate to that i think overall the thing that I'm most excited about certainly that got catalyzed by co I mean don't get me wrong I'm I'm not about to celebrate something like co but this was a good silver lining to come out of it is that the digital environment is finally pivoting to actually listen to employees in the line of business um and that's something that we talked about employee engagement we talked about measuring it before co but it was a bit tokenized it was a bit driven forward by the line of business not necessarily formalized across the organization by HR and you know before before I got into the ex side before AI was a thing I started my career when we were talking about the onset of all-in-one turnkey HR suites that were really designed around automating core HR processes employee experience wasn't a thing yet um or if it was it was being very very human- driven there weren't purpose-built tools for it what voice of the employee did was transform the organizational landscape so that employee sentiment employee feedback all of the insights could come back to tell HR where employees were being resourced properly were being enabled where the learning was working or not so that performance drops weren't totally their fault or failure to achieve OKRs was not their fault we haven't had that kind of connection between the line of business and you know executive OKRs since the end of the 1970s when we started decimating middle management um in favor of automation now we're really challenging the top down management models and it's for for a good outcome you know it's fascinating you raised that because I think to myself there there are several different uh kind of watershed moments that we think back upon and I always think about the great reset that took place in 2020 and how it made us think about not just the tech but also the things that we we do from a management uh perspective and and how we analyze things even differently so uh certainly that's that's a a powerful moment is uh as I think about it one of the things that I think about is I think back to 2023 and obviously or 2022 November of 2022 and I think about the the big announcement and and how 5 days led us to a million users of ChatGpt thanks to OpenAI and their and their social media game uh I I I wonder specifically when you look at HR and AI where is it that you see the biggest impact so far where is it that you see that great rate of adoption and what what parts are really resistant learning experience management continues to be the top area where AI has come into play it's the earliest place where AI personalization has taken off through you know behavioral science modeling engagement data all the rest of it um and it's been a big catalyst for learning and skilling while also a big proof point for operational streamlining so that the learning team produces maintains and supports only what employees are engaging in the way they're engaging it um that's had extended you know value cases out to operational resourcing it's given HR value to to tell finance and ops teams where the organization is and is not empowering employees correctly um and it's led into other avenues like personalized data delivery in pay compensation acrals performance etc um and you know after the LXM environment talent acquisition has been a big place where um where particular AI assistance has come in for candidate communications automation nudges and talent CRM to personalized workflows within the talent acquisition process flow and it's it's been a big big catalyst and an easy place for adoption ai ultimately faces opposition not in any one area of the HR remit but in places where it gets either rolled out too noticeably or too fast so there are a handful of vendors that are that are either acquiring advisors or you human advisors or working with their SI partners to build uh change management or behavior change strategies to phase AI enablement in through insights enablement first to make employees more aware of the outcomes enablement of AI rather than you know saying we're going to take this process away from you don't have to do this anymore so it's more about getting the behavioral shift into using these these enabled tools rather than necessarily any one tool saying you know resulting in an employee or an HR leader saying I don't want that you know it's fascinating uh I I think about this and I know I say that a lot i I fascinating maybe a crutch word for me uh one of the things that strikes me is I I've fielded a transition myself from HR leaders uh in terms of what they're doing around LXM right to your point around how learning experience is changing how they're engaging people and the creating the experience and I had somebody who uh a year and a half ago would have told me there's no way in hell we're ever going to do anything with deep fakes i don't want anything to do with deep fakes and they came back to me a year later and said "What if we were to consider the positive side of deep fakes as an example and say we created a personalized learning experience for onboarding?" And it was personalized with a a communication from your CHRO and a communication from your CEO and other executives that made it possible for you to really leverage and feel personalization in your own u in your own workforce now promise me you've never seen anybody actually apply uh deep fakes in that regard unless it was really in a in a warranted way but what do you see in that well you know I mean I think it comes down to governance and consent ultimately i mean I I don't think you're going to see a CHRO or learning leader or a talent leader you know have a problem with the with their voice being used as long as they can be shown where the the gates and the guard rails are around what that you know Genai voice can say so as long as they're they're they consent to the language that's being used to come out of their their persona simulation I don't necessarily see anything terribly out out of sorts with being able to use it that way but like I said that that whole you know human digital twinning concept is is a little interesting to me i think the better direction you know for employees is to proverbally create a data shell around them that moves with them even when they're new hires and attaches the right resources to that shell based on how the behavior is moving so personalizing the experience I'd rather have rooted in behavioral science for resourcing first before you get to those those augmented automated communications it may send a nudge to that CHRO that you know employee A needs this standard communication blasted to them um and when they see it coming in from the CHRO usually that's good enough it's not always about what is said but the fact that it is said so that's fascinating i I I really appreciate that uh all right let's talk about the future because I'm I'm really in in interested in what you have to say about the future and really as you think about every bit of deep expertise and analysis that you've put forth what what is exciting you the most about the future of HR tech what is it that that lives out there that's Zach is pumped about i think I alluded to it earlier but the fact that we're finally moving in a direction where performance can really be an outcome versus a process is yeah I'm I'm totally nerding out over it we're moving into an era where AI can relate both directly connected and and disconnected or or seemingly unrelated data sets across multiple remits so the concept that we could have a data asset manager you know like a that builds a common data platform between the HR systems the op systems the finance systems everybody working from the same consolidated business performance model that ties in all of the workforce engagement and performance insights is a huge deal and you've got a lot of vendors particularly the big vendors that are driving towards that um and it's what's exciting about it is nobody likes a subjective performance review we we've stopped at the halfway mark kind of like getting to nuclear fusion but not quite at fusion where you know employees and managers are enabled with the same data insights about an individual about their individual performance so that they're on the same side of the table when they're charting the next steps and goals but we're really poised to go so much further than that this these systems will nudge for when an employee needs to be trained or retrained on something because their skill and performance is dropping a little bit maybe they need to be working with a better optimized team relative to their historic performance so they're going to get slightly moved internally and all of this is feeding into continuous workforce planning that's agentically driven um so that organizations can better adapt to changes in their market without needing to always stay on top of you know I'm going to try to control my market segment and push risk down to the front lines of business because I just need to control everything I can and keep the company on track no we're going into more fluid business planning cycles and for HR that's increasing their relevance at the table with finance and operations because they're the third leg of the stool if you don't have workforce performance embedded into those models you just can't get a handle on the entire frame for business performance so HR may finally get the seat at the table i always joke they were HR when they realized that they're sitting on a lot of operational insights from voice of the employee you about resourcing not working or working they bring it to the table first only to find out that they didn't have a seat at the table finance and ops move their seats aside so HR could bring a folding chair as long as they towed the line and we're starting to we're at the very precipice of seeing that start to change and all of that hinges on AI enabled performance as an outcome and I'm very excited excited about it i really appreciate that perspective because I've been on the road talking specifically about how AI agentry and human agency are the key levers to productivity right and if you're not talking about productivity you're missing half of the equation and it has to be performative you know that we can no longer measure productivity on just what the standard operational output is it has to be performative and it has to be behavioral agreed couldn't agree more so I'm going to put on my CHRO hat and I'm going to ask you one final question here what's that one wisdom that one takeaway that you want to every CHRO every every seuite leader to walk away with and think about and say if they're talking about HR and AI analytics what would it be what's that one guidance or wisdom well the first one is I don't think that this is anything new but HR leaders fight for your seat at the table you're smarter than the other stakeholders often leads you to believe um and you deserve your seat at the table the second one is it is going to be your friend as AI comes into play HR is one of the first places that vendors are experimenting with AI capabilities why because HR leaders and stakeholders often need their solutions to be more endto-end complete and out of the box ready than their counterpart stakeholders yep your technical literacy is not often as strong as your counterparts but that works to your benefit to be a first experimental house for AI enablement that means that in partnering with it learn how to speak each other's language because when you've got to carry those behavioral insights into the other remats it is going to be the partner that's going to help you get there that HRIT partnership is fundamental to success in the future of a digitally enabled work environment and then the last piece is carefully monitor and and yeah take the gold that's given to you when it's time to escalate your rise in the hierarchy of stakeholders finance is going to resist giving up their seem that the control they feel they have over the organization operations is going to resist even AI to a certain extent because they don't want to give up the blank check that is procurement and right now neither one of them know which end is up because of global compliance disruptions you've got a chance right now to really take a lead on organizational transformation step into that role embrace it i really appreciate that and more than that I think specifically about the power that is that reset continuing that we talked about 2020 being a great reset and this continues to be the next layer of our reset for this entire profession that's going to do it for this week's episode a big thank you to Zachary for sharing his expertise and deep insights with us zach I can't thank you enough before we say goodbye I encourage you to follow the AIHI project wherever you enjoy your podcasts if you enjoyed today's episode please take a moment to comment leave a review etc anything that you want to do to let us know how you feel about today's episode finally you can find all our episodes on our website at sherm.org/ihi thanks for joining the conversation and we'll catch you next [Music] time sherm
2025-04-23 02:24