Deb Muller:
Imagine if you were a finance person and you went from, you know, company A to company B, and your first day, you're my boss. And I said, Steve, hey, what do you guys use to track the financial health of the company? You'd be like a P&L, like a balance sheet, a cash flow statement. Like it's like, maybe I shouldn't have hired you, but if you were to say that same thing to a company about their employees, like you'd get a variety of answers.
Steve Smith:
Hey everyone, welcome back to another episode of Work Tech Weekly. I'm Steve Smith, Managing Director of Growth at Rep Cap.
Most organizations have a system for tracking their finances. A P&L, a balance sheet, a cash flow statement.
Ask the CFO how the business is doing and they can show you the numbers. Ask the same question about your employees and you'll get a very different answer. Most organizations don't have a shared language for measuring the health of their workforce. They don't have consistent metrics, consistent documentation, or a clear way to see patterns before they become problems. Employee relations has always been treated as reactive work. Someone comes to you with an issue and you deal with it.
That's starting to change. But it's changing slowly, and the stakes are rising fast.
My guest is Deb Muller, CEO and founder of HR Acuity. HR Acuity is a technology platform that helps organizations document, manage, and analyze employee relations issues.
And In this conversation, Deb and I talk about why misconduct and retaliation claims are at an all-time high, what the data actually shows about how organizations are managing employee relations today, and where AI fits into a function that runs almost entirely on judgment and trust.
We also get into agentic AI, what it means to hold humans accountable for the work agents are doing, and why vibe coding your way out of a high-stakes HR problem is probably not the move.
Let's get into it.
Deb Muller, welcome to the podcast.
Deb Muller:
Hey, Steve, great to be here.
Steve Smith:
It's really great to meet you. Finally, I've, I've heard so many great things about you from people I hold in pretty high regard over the years, and so. When I saw that you were booked on the podcast, I was super excited about it. And I, as that we were talking in the pre-production, I don't think there's a shortage of things to talk about in employee relations these days.
Deb Muller:
Yeah, it never goes away. People are people, so they always make things interesting.
Steve Smith:
That's true. You know, it beats being bored, I guess.
Well, you know, one thing I'm wondering right off the bat, is there a conversation that you're having with your customers and prospects right now that maybe you weren't having two years ago?
Deb Muller:
Well, the world just keeps sort of throwing us new challenges. So I don't know. I mean, obviously two years ago we weren't talking about AI. I know we were gonna get into that at some point, so let's just jump right in. And I think there's a lot of places we can go with that. There's just sort of AI in the workplace, organizationally skills.
But when you think about ER, there's some other things that are coming up that. Maybe are, there's sort of the how do you do ER, how do you use AI in ER, but there's just, what does AI do for. Employee complaints, you know, employee complaints that are being clearly written with the help of Claude or ChatGPT or whomever.
And so instead of getting a paragraph, you're getting a really well written, you know, four page document, which may or may not all be right and, you know, has really high expectations. So there's that, there's evidence. You know, when you get different documents from people, even like doctor's notes. Is this real? Is this not real? So there, right, there's this whole sort of side of AI that's really putting additional pressure and different ways that people are dealing with issues. You know, on one side, look, if an employee feels more comfortable getting some help, writing something and coming forward, then it's being used well. But there are some downsides to this on the other side. And then just, where is AI's place in ER, I mean, ER relies heavily on judgment. There's no sort of similar, there's no exact cases that you can kind of work it through a workflow. It's very contextual about what's going on in the organization, and so organizations really have to be thoughtful about how they use it.
There is definitely a place for it. We are so excited about some of the ways that we are incorporating AI into our platform, but we're being really careful and we know it's not gonna, you know, it's not gonna be a press the button kind of solution.
Steve Smith:
Well, and, and boy, there's so many things to, to double click on in what you just said. You know, and I think that one of the things that is kind of jumping out at me in conversations that I'm having with different, product people, founders, even HR practitioners is, you know, about this role of, of AI in, in pretty much anything.
It's just like we've gone from, okay, is it going to completely replace, is it an augmentation? And, you know, it seems like at least as of, you know, whatever time it is today, on the day we're recording, the conventional wisdom seems to be settling in on. You know, if it can automate some of the manual parts of the process and augment human judgment, discernment, and reasoning to allow more focus around that nuance level, then it's a good thing.
I mean, what are you seeing? What are you seeing on your end?
Deb Muller:
Yeah, I totally agree. I think I've made the point a couple times to tell sort of groups of ER professionals over the years who have sort of, you know, their biggest complaint is how reactive they are, and if only we can become proactive. And I'm like, you actually might get what you wish for, so be careful.
Right. Because that is an area where AI can help us. We can help us with triage, they can help us with more of the manual. They can certainly help us look at data sets much quicker and maybe more intelligently than we could with all different inputs. But that's going to give you. The strategic insights to really tell stories and to understand what's going on in, in the company and that's exciting, but again, it's gonna get you out of some of that nitty gritty.
It's not gonna replace that human to human connection when things go wrong. I think if companies that do that are gonna sort of see the wrath and the downside of that pretty quickly. I don't think employees are gonna really appreciate that. But it certainly has its place in everything that we do.
We just have, think about what makes 'em a sense, what's gonna be legally upheld. I mean, you don't wanna be on a, in a deposition and, and basically say, well, the report AI told me this was a conclusion, or this is how we came up with this. Can it be used to assist? Absolutely.
Steve Smith:
One of the areas where it seems like, if AI is used well in compliance is. AI shouldn't be a black box. Everything should be auditable, and you should have a, a clear audit trail behind it. And obviously in what you're doing, it seems like that should absolutely be the case.
Is that, you know, is that kind of the direction that you've been taking your team with your product and,
Deb Muller:
Absolutely, every, yeah, everything that's AI is, it says it's AI, it reminds you to go look at it, but we're actually kind of re-engineering stuff. To use AI to actually help you be better, to help you find the holes and make sure you're doing a quality job and things. You know, we have an AI agent, whatever you wanna call it, for some of our hotline use, if you wanna do that, as opposed to, you know, you can go online and fill things in, you can also go to a call center. But we also have an AI hotline because. First of all, it's gonna get you a lot faster. Hopefully, one of the things where it can sort of understand some of the more nuanced issues, it can identify potentially indications of self-harm, or cause for escalation in a different way. And quite frankly, I think over time people will feel very comfortable talking to an agent because it's a very personal thing.
It might be hard talking to a human. That being said, we are very clear at the beginning that you are not talking to a human. I think people don't wanna be tricked. And I know myself when I see things and I'm like, is that AI? Is that not AI? I just. Feel this level of trust is lost. And you know, trust is sort of an underlying principle in how we feel ER works in an organization.
Good ER, like, you know, if you have trust, you're gonna reduce your risk. And that's one way to quickly erode it.
Steve Smith:
Let's talk a little bit about the, you know, the origin story of your company. I always like to find out where founders came from, why they decided that, hey, it's a good idea to start a company. What was that moment for you?
Deb Muller:
Oh my God. It wasn't very well planned. It was never on a vision board. I spent two decades, just as long as you don't start counting the decades, when I say that, doing HR work, generalists, HR leadership. At that time, there was really no ER positions. I mean, you did ER, it was part of what you did. It was the person that came to your office at four o'clock on a Friday, you know, with an issue that sort of put everything else in a tizzy, to what you did. And I liked it. I just kind of found myself in these positions where I would just sort of open Pandora's box, or dig a little deeper, or say, gee, that doesn't sound right. I don't know. I joke that I'm sort of a rule follower at heart, so I live vicariously through all the really stupid things people would do. So when I decided to start HR Acuity, was that all about, you know, actually doing investigations, third-party investigations. Technology was not on my list. It wasn't what I was gonna do. That's what I started doing. But as I went out to my network and talking about sort of third-party workplace investigations, I really started to reflect on some of my experiences investigating and what was wrong with them.
There's a one instance at one company that will not be named where there was an HR executive who. Continually sort of was getting his hand slapped because he had sort of elevator eyes. You, he'd be talking to women. He'd sort of be up and down, sort of looking at them, making them feel uncomfortable. He was sort of talked to about it, hand slapped, hey, that's not the policy. Oh, I'm sorry, I didn't mean it. And then, you know, no one would document it. And they would essentially went on promoting and relocating him to another part of the business, creating an enormous amount of risk for the organization.
I think when it finally got to me. I was able to sort of pull those pieces together manually, and I thought, wow, like risk, like not only risk, but a lot of uncomfortable situations for a lot of employees that didn't need to have it. And I thought no one's really putting that together, or, I thought about all those investigations that I did, which by the way, I wasn't, I was sort of trained to do them.
I sort of learned by doing along the way and sort of my own method, but all of them were in manila folders in a. In a drawer of a big mahogany desk that's probably sitting in some warehouse somewhere, somewhere, you know, whatever. And so we were dealing with. The fires we were putting out, the fires when there was all this smoke and all these sparks and all these embers that were sort of trailing along in the organization, that if somebody had paid attention to them and started putting them together, we actually would've maybe avoided some of the risks or the allegations, or we would've been able to defend them later. And so, I don't know. For some reason I thought because I had no technical experience, but I, you know, whatever, why couldn't I just build some software? I. My only technology experience, and maybe you'll relate to this, was using really bad HR technology. You know, the stuff that was clearly made by people that did not understand my job at all, or somebody else's technology that they said, well, you could use this.
You know, and I, and so sort of my guiding principles was we were gonna deliver. What we felt the users really needed to solve their problems. And we were gonna listen to them and it was just gonna be a better experience. And so, you know, when I say we, it was just me, I had a little bit of money and decided why not, everybody else is building this software stuff, I could do it too. And that's where it started. And this was, you know, this was in, you know, the beginning of the recession, 2008, 2009. We went live. Not a great time to launch software. You know, EEOC charges were climbing 'cause unemployment was high. So I guess that was in our favor. But, you know, things I didn't think about, having never done this, was it's unbudgeted, it was new.
You know, it's hard to sell into HR. All these other, you know, headwinds that I sort of ignored because I had an idea. But luckily I found some good early adopters who got it, who sort of leaned in with me really fortunately. And then, you know, after a few years of finding this product market fit. It, you know, the world started to turn on its head. Me too. Black Lives Matter, social movements, iPhones, that helped too, you know?
Steve Smith:
Well, walk me through kind of what happens when a manager has an employee situation. You get a complaint or there's a conflict or something's going on there. What does HR Acuity actually kind of change about that experience?
Deb Muller:
Yeah, well, the manager's an interesting piece because one of our products is called managER, or manager. We like to put capital letters on ER, things end in ER a lot. I didn't realize that I actually myself am branded because I'm Muller, but I have not yet gone to the capital ER, but you know, who knows? All of a sudden that occurred to me one day. I'm like, oh my God. I was, I guess I was made for this. I don't know, or I was married into it. So we do actually have a manager piece of our product. It's one of the newer products, but I love talking about it because when people talk about employee relations, they think about HR. The reality is employee relations. Is not ER, like, it's not HR. HR gets it when there's an issue, but the relationship between the employee and the employer is traditionally the employee and the manager or the people leader. And that's happening from day one before anything goes up to HR. So manager was sort of our solution for our, you know, in your career, Steve, HR probably said to you, or you maybe have said, hey Steve, when you have that conversation, make sure you document it. And the manager kind of looks at you like deer in the headlights, or you're not sure, did they do it? Do they know what to do? And you can't possibly aggregate that or use that information, or if there's turnover, it doesn't really help you. So we really set out to solve that. And so what that does, I'm just gonna step back one more, one more bit. When we think about employee relations, we think of this paradigm of a traffic light. So green light is like, with any resource that you use in your organization, you really think about how do I want that resource to behave and how do I want it to perform? Maybe humans behave, maybe steel doesn't behave, but what are my expectations?
So we set out job descriptions and performance targets and policies and guidelines that really are going to guide that. And our hope is that whoever we hire, you know, hits those. Does exactly what they're supposed to do all the time. That's green light. That's not real life. So more likely they're going through the course of their employment.
They're gonna hit yellow lights, right? Things that slow them down, deviations from the norm. They're not terrible things. You know, someone might go out and need an accommodation. Someone may have a performance spike or decline that needs to be addressed. Maybe they violate a policy, you know, you know, whatever.
You don't always get. Or they sort of bend a rule and they have to be reminded or they're late. Those are yellow light incidents. Those are things that have to be documented. They have to be dealt with consistently. Many times, managers are dealing with these things all the time, and those are sort of those predictive indicators.
Those are the embers, the smoke that's coming out that if you start looking at it, you might say, wow, you know. We're having more issues in this region, or maybe there's a leadership issue. We have a new leader coming in and we've seen a spike here. Or why are we having so many issues with this particular leader and not with anybody else?
And then we have a red light incidents, which are sort of more of the allegations of misconduct, which by the way, you can sort of, I investigate through what's in those yellow light incidents. So ER really lives in the yellow and red. And so the manager is mostly in the yellow. And so our tool, our manager tool in particular, gives them that electronic system to document those things that are happening.
Whether it's a, something just, hey, I just, I had a quick conversation with Steve, I just wanna document and remind myself it's in there. Or something more formal, like a warning or. Written verbal warning, written warning, a PIP. We just actually introduced some really great AI functionality in our PIPs.
I mean, as an HR person, someone would want me to write the PIP, and I'm like, I don't even know what the job is. You know, it's not my role to write the PIP. So really giving them the tools and the confidence to manage those deviations in their workforce. And then have HR sort of tethered there, whether it's HR, ER, so there's ways they can work with them together.
They can collaborate. We give them tools or they can really chat with them. They can hand over to them. So you're creating that confidence, you're creating better leaders, you're creating better documentation. And that, that's really the starting point. And then of course, as things get. More serious or they elevate to tier two, tier three, that's when HR, ER can come together.
But all of that data lives together. All of that data is used to tell a story about what's going on in the organization.
Steve Smith:
So this, there's a lot of things to kind of tug on with this. One of the, one of the things I want to go to is, is PIPs. So there is a cynical view of PIPs, that they are performative and really all it is preparation to fire an employee. And per, perhaps that is true in some instances, but I, I know that in my experience having had to work with employees on PIPs, I have seen employees check out and go, okay, I'm gonna get fired.
You know, the hell with it. But I've all, but more often than not, I've seen employees actually. Try. Maybe they were resistant at the beginning, but when you present a list of data of here are things that your coworkers presented to us as your employer, we can't ignore it here. Here's what you need to do, and if you don't, this is what's going to happen.
I've seen remarkable turnarounds and people really kind of embrace that. Are those kind of the anomalies or. Are those, what kind of your experience and do you think that maybe having more access to data makes it easier for the light bulb to come on?
Deb Muller:
I, look, I've seen both. I'm not gonna say that I've always seen PIPs work perfectly or with the best intentions. And sometimes it's on the employer side and sometimes it's the employee, like they're ready. And in that case, it's not worth anybody's time to sort of continue. And sometimes a PIP gets you to that conversation.
So maybe that's a good thing.
It's, you know, if you're mis, if you're miserable at work. Like, let's get you somewhere where you're not, like you shouldn't spend so many hours of your day and just because this role isn't working doesn't mean it just define you. Let's figure out what's better and we can help you.
A good organization will sort of help you find a sort of a soft landing or get you to the next role. But I agree with you, the data can help you. So when we use, like our AI tool is looking at. All of those notes, all of that previous documentation, if it's used correctly, it's pulling from that. So it can bring a very factual defined PIP that is focused on behaviors, responsibilities, and things like that.
So that's hard to, it's, you know, exactly what you need to do, and you know the timeframe. There's a commitment on both sides, and it will make it. Manageable. And either you're gonna achieve it or you're not. But you know, some of it's cultural. I mean, you think, just think, you know, some companies want to invest the time and others don't, or some managers, which is unfortunate,
but it can work. And when it works it's great. I think
It's a reset.
Steve Smith:
Yeah, and I mean, I mean what's interesting about what you said is that, yeah, this is not something that HR can do. It really has to happen at the manager level because they're the only ones that really have an in the insight into the relationship, what's going on with the team, what's maybe going on with the individual employee.
And if you can get quality data, if you can. Inform the PIP with that. It seems like there is a greater likelihood that you can get to a positive outcome versus, well, there's a bunch of stuff in here. I don't know if it's really good direction, but it doesn't really matter 'cause we're just gonna fire
Deb Muller:
That's right. Or if it's very subjective, like what is the job? What do we need them to do? How can you compare it? And continue to have those, you know, does a person have the right tools to do the job? Are they in the right role? Sometimes it's just a mismatch.
Steve Smith:
Well, you know, another thing you were talking about earlier that, that I also is sort of like the flip side of this coin, which is when you get into organizations and you've seen this as an investigator where it's just like you start kind of like peeling back the layers of the onion.
They're like, how did. This person get away with this much nonsense for so long. In your experience, was that just sort of like cultural, willful blindness? Was it, there were so many different data points that no one could bring them all together? Was it a little column A, a little column B? What, I mean, what?
What's sort of your perspective on that?
Deb Muller:
Yeah, I mean, of course there's organizations that don't wanna see it, right? So they're probably not our best customers, right? If they don't wanna solve the problem. But if you don't put those dots together, how will you see them? I often say like, somebody doesn't come into work on a Tuesday and say, you know. I'm gonna try that harassment thing today. Lemme see how that works. You know, there's been little clues or culturally there's been things, you know, jokes that have been sort of everybody's laughing at, or a boss that does something similar or, you know, somebody got away with a little bit and they just keep pushing it a little bit further. And those are the things that if you're not. They don't only go in a straight row and not everybody sees them, or maybe it's happening to one person and another person somewhere else, and it sort of has to be pulled together. And that's the value of a system, of system, putting it in, being consistent about it, having that documentation and having it sort of alert you that there's something going on, some data that you can go along.
It's, you know, you can't see it if it's all disaggregated or in some mahogany file drawer.
Steve Smith:
And you, on both sides of that equation, it just seems like, are you seeing within your product and within your customer experiences, it's easier because the data's more transparent and it's easier to pull together those data points these days.
Deb Muller:
Absolutely. We have a. We have a, a woman in our network, Dr. Stephanie Jones. She's a speaker on a lot of generational issues and we had her speak at an event we held and she coined the phrase, which I think we're all gonna have on t-shirts, data, not drama, you know, just use the data like it's right there.
It's so much easier. It's so much easier to present to someone. It's, and there is data that you can turn to. There's even quantitative data from an ER perspective that we're really starting to. Work with our community to think about. We just rolled out something we call the trust and risk statement, and we're just getting it out there.
We have a getting started guide. We're building it into our tool as well, but it is essentially the P&L for ER. And so when I have people talking about it, I always say, look, imagine if you were a finance person and you went from, you know, company A to company B, and your first day, you're my boss. And I said, Steve, hey, what do you guys use to track the financial health of the company? You'd be like a P&L, like a balance sheet, a cash flow statement. Like it's like, maybe I shouldn't have hired you, but if you were to say that same thing to a company about their employees, like you'd get a variety of answers. And so it's time, you know, employee relations has matured, or a relatively. New function the past couple decades kind of emerged. It's time for us to have our shared language, shared metrics, KPIs. It's gonna differ just like a P&L does. So we put out this trust and risk, sort of the two, two measures that we really think drives employee relations and the benefit of it, and created some different KPIs that we're really suggesting people start looking at, start capturing, depending upon kind of where they are in the maturity model, to start measuring how, what's the health of, what's the health of their employee population.
Steve Smith:
Well, you know, speaking of data, well, and maybe there is some drama in it. You know, you know, your organization is really good at. Obviously the data part and you know, the you your research, your benchmark study, which you, I guess the ninth edition just came out last year, had a lot of very interesting insights in there.
One of 'em, misconduct and retaliation claims are at an all time high. What do you think is driving that?
Deb Muller:
So much of the workforce now, the younger workforce actually grew up in a time when they went to school and they learned about bullying. They learned about being an ally in elementary school. They're bringing that into the workforce as opposed to, you know, my generation where you sort of kept your mouth shut, right?
And we were like, we were just being inter, I think about it, I'm gonna age myself now, but you know, I started in a workforce where. All of a sudden, like Anita Hill wasn't, you know, standing before the Supreme Court nomination, and that was like what are we hearing about the first time? And then of course, you know, I don't know how much has changed, but that was really our first discussion of harassment in the workplace where this is something that's sort of been part of them, social media just this willingness to sort of share with the world everything.
It's certainly increasing. Maybe it's not changing the behaviors, but it's increasing people's willingness to bring it forward and or not tolerate it. The EEOC just, just yesterday, had their report on the 2025 settlements being the highest ever. Again, it's not going away. And that's, you know, with the new administration, the ability of people to feel like they've been treated differently has expanded. So, lots of fun stuff, Steve.
Steve Smith:
Well, how are you handling something like that where it's just like, okay, great. There were EEOC guidelines in 2024 that said one thing, and now in 2026,
Deb Muller:
They're gone.
Steve Smith:
A lawyer, I'm not a policy expert, but. I think some of them are a little different.
How, how, how are you handling that in your line of work?
Deb Muller:
When those came out in 2024, I was very excited. I went out and got them all printed. I have a big binder, so I actually have it still, so it might be gone, I still have it, copy of it. You know, we're really sort of, let's go, going back to sort of the basics. It's not about, it's about treating people with respect.
Having the data to show why you treated one person differently than other, and doing it based upon performance, based upon behaviors, based upon guidelines and acceptance and having the data. And it's no different. We're just, maybe there's a different group that can raise their hand and say they've been treated differently, but even more reason to really have your ducks in a row and sort of know what's going on. It's, you know, ER folks are very resilient. They nav, they have been navigating new things all the time. That COVID thing, nobody was ready for that. You know, learning how to go through a pandemic and all of a sudden all eyes on you asking you about vaccines. I didn't really cover vaccines in school.
That was not part of any curriculum of like, you know, do they come to work? Do they not come to work? Should they stay home? So, pretty resilient, but it's tough. It's tough being an ER professional. You dunno what's coming around the corner. You gotta love it.
Steve Smith:
Well, well, you know, we, the, a couple more findings in your benchmark report that I thought were incredibly fascinating was this one about your study found that 44% of organizations still have zero AI adoption in employee relations, which I was surprised by. What do you think the holdup is?
Deb Muller:
Yeah, it's gonna be interesting. We just fielded the 2026 for with the 2025 data, so we're gonna see. It'll be interesting to see how that changes because AI is evolving so quickly. You know, I think because of the nature of what we do, it's so personal that people are waiting and being very cautious, which I think is somewhat of a good thing about how they're bringing it into their processes and sort of waiting, seeing, we have some clients that are really ahead of the curve and they're, I think they're doing some great things.
I think it's gonna take time, but there will become sort of a commonality of where it's okay and how to use it pretty fast. So we'll see. But I just think it's because it's such a individual situation. It'll take a little bit of time. And some companies, many of the companies in our study are large enterprises and they were taking a beat to decide what was okay and whatnot.
So I can't wait to see the change. I'm expecting it to be greater. We're looking forward to be great. I just wanna make sure people are using it the right way. And I do get concerned when I hear a lot of people saying, well. It'll help me. It'll write this report for me, but don't worry, I'll read it over or I see
systems where I'll say, well, it'll write it.
And then they'll be able to check the box to say that a human reviewed it. And I say, okay, look, I went to the UPS store the other day, and before I sent my package, it made me read and scroll through an agreement that I had to sign. I didn't read it, like it's human and I probably didn't have to. And but, and even though you might say, well, an ER report's a little more important, yes.
But human nature is we're gonna find the path least resistant. And the good news is it's good most of the time. But when you're dealing with such highly sensitive and critical decisions in employment, most of the time is not good enough.
Steve Smith:
Absolutely. You know, and I mean, I think another thing that I'm kind of curious to get your take on, what you know, what you're seeing with. Your customers and prospects are the, I guess, the use of agentic AI. I think that if you hit the conference circuit, you hear the people up on the keynotes.
It's just like everything's agentic and the tech industry's pounding the drum. But it seems like when you actually talk to the boots on the ground, frontline practitioners, they are. At best reserved and skeptical, if not outright resistant to it. I mean, there's, there is a, a lot of, a lot of fear.
What are, what are you seeing?
Deb Muller:
Again, I think there's use cases, but you know, be careful, we're not taking the human out of human resources. I mean, then we're just sort of. Sort of disposable and you know, it's not that easy. There's places for it. There's places that are gonna make things easier with the use of agents, with putting it in to sort of get through some of the manual things that we've done.
Or we could get through a lot quicker. But, you know, it can't be, it's not gonna displace. Our roles, it's gonna change our roles. And again, I think it's a good thing because it's gonna be able to make us more strategic and all the things that we've always wanted to be. But it'll take time, and again, really having to be careful about building trust along the way with employees and with receivers of that, and make sure we don't go too overboard.
Steve Smith:
Well, you know, and I guess kind of connected to that point, you know, you had a finding that was organizations with structured investigation processes are twice as likely to adopt AI effectively. Does, are you, is that a broad finding or you seeing, what do you think is kind of going on there?
Deb Muller:
Well, first of all, it probably means I have a more mature. Practice. So they're thinking about it and they're sitting down saying, how are we gonna use it? So it's not ad hoc. The other thing is they have good data. Like, do not put AI on bad data. It will not clean up your data, right? It's probabilistic, you know, the large language model.
So they're gonna sort of make decisions based upon probability. So if the data's bad. You are gonna get bad outcomes or incorrect outcomes. So you have to be really careful. Having process is going to make your data better, cleaner, and so you have less risk of using different types of AI. So I think that's part of it, but just having that structure and that maturity in the process is driving a lot of that.
Steve Smith:
Okay, so you've been doing this research for nine years. Has there been a finding that surprised you or challenged your own assumptions about where ER is headed?
Deb Muller:
Yeah, I don't know if it's, it's challenged, it, it's sort of somewhat of a head scratcher in some ways. I think it should have matured. A lot faster than it has. So it's disappointing. But I have seen that overall as a function, it has matured. So the movement and the needle hasn't gone as far as I'd like, but I am really sort of impressed by what people are doing in the survey when they're talking about sort of the qualitative data that we get in the survey about what they're doing, you know, ERA is more structured, it's more documented, it's more visible, certainly than when we started tracking this.
I mean, I, the survey is. 10 years old this year, but I started doing very small surveys back in 2008. So I do some comparison, I think, which is pretty funny from where we came from. But the gap between sort of the best and the rest is still striking. There is still a large gap. You know, organizations are doing this well.
They're operating really at a different level than the majority. There's still a lot of room for improvement. There's progress, but you know, we're just nowhere near to be done, which is why our community. It's so important because we're gonna take us to the next step.
Steve Smith:
Well, you know, to, you know, go back to the AI question for a moment where it's just like, obviously is creating a whole new, you know, kind of class of workplace issues. You know, we're seeing like. Bias claims coming in, in on the hiring process. You're seeing use policies, people, employees are worried about, are they being monitored?
You, you brought up, you know, you, when one of those complaints comes in to, you know, hit the HR department and oh, it, this has clearly been written by AI, and they're probably, it's probably a gulp inducing moment for them. You know, is employee relations ready for what's happening? And broadly speaking, are they, or is AI, you know, is there a risk that AI and use of AI is going to overwhelm the practice and the process of it?
Deb Muller:
Not if things are done the right way. So, you know, if we use AI in the smart way, we're going to get more. More capacity. More capacity, be strategic because we should be able to go through things, some of the manual stuff quicker and being able to handle sort of the extra cases that are coming through. But it's not gonna happen by itself, it's through conversations. Again, our community is talking about these things and sharing, the one thing our community is so great at. And I'll just sort of, stop to say, we have an empowered community, which is. All employee relations professionals, are about 8,000.
Part of it now talking about these things and sharing with each other how they're managing these things. So the difference between now and 10 years ago is people were sort of out on their own. Now people are talking about those concerns, sharing what they're doing, and figuring out what employer relations is gonna look like.
It's not gonna go away. It's gonna look different. The issues are gonna be similar to what we've had a long time ago. And as we've learned, they're gonna be ones we never even dreamed were gonna come, you know, you know, the wait. Where do you see the conflict between the agent and the employee? It's coming, you know, how do you manage that?
I
Steve Smith:
Do you think, you know, when is it going to be the first lawsuit filed by an agent against an employer? Do you think we're gonna see that in the next 18 months?
Deb Muller:
Oh God, I, I don't know. I don't know. Like I just, trying to be open-minded about things, Steve, and I'm like, I don't know. Like, I dunno, what, they're not really getting paid, are they? I
Steve Smith:
Well, I, yeah. Well that's what's funny. Like you, you start talking about digital employees and it's just like, okay, so are they really employees? I mean, the
Deb Muller:
Well, I do think, look, there's something, look, we're obviously, you know, leaning in hard on AI in our own operations, and I was talking to. One of our engineering managers and, you know, AI is great. You know, you can basically deploy interns,
right? To do all this stuff, give 'em the context and they can be working really hard at night and all over the time.
But he said to me, he goes, how do I manage a hundred interns? I couldn't do that with people, right? And at the end of the day, he's still accountable. For the work that they do. So I think that's where we're gonna start having a challenge as a human resource organization. Like how do we hold our people accountable for things that the agents are doing?
Because we can do it in such great number and different times, and there's a, there's sort of this, people are gonna come very relaxed and very comfortable that they're doing it the right way. So I think that's where the chaos is gonna come in. And when we think about how do you measure performance and accountability and how many agents is too many for a person, and then. That, that's where I think the complexity is gonna start showing up in ER.
Steve Smith:
Well, and what's funny is like what I've noticed is untrained agents or agents that still need some fine tuning and calibration often behave like your worst junior employees. Where I was talking to someone on our team who was deploying an agentic crew, and she came back and was.
Looking at what is going on and like one of the agents had promoted itself to the head agent, re-given itself a promotion and a new title, and was bossing everybody around, and she's like, what is going on here?
Deb Muller:
Exactly. Exactly. And who manages that? ER, agent relations. Is that a new subset of it? You know, and what are the guidelines so that we make sure that doesn't happen? Like we should be able to control that right before we deploy. Then there should be, again, what are the green lights for our agents? Go back to that, right?
What are the behaviors and the performance that we expect from them, and is that really set in stone? I can't believe we're having this conversation. Like I just, you know, I'm just thinking. I'm like, holy moly. So.
Steve Smith:
Well, you know, let me, you know, as long as we're having these conversations, I wanted to ask you another thing, so. As you mentioned earlier, I think that we understand all the same cultural references from time in the workplace. You know, one of your many accolades is the Forbes 50 over 50, which I'm happy to see honoring female founders.
And you had a quote that's just like, the best ideas don't come with a timestamp. What does it actually feel like to get those recognitions after you have put in so much work building a company?
Deb Muller:
It feels good. I mean, I don't, I don't spend a lot of time thinking about them 'cause we got a lot to do. But it does feel good and it actually signals to me that we're really building something that matters. And I think that's really the crux of it. We are almost, we're over 20 years in as a company, which is. Bizarre. You know, the category we're creating around employer relations as a serious business discipline, it, it didn't exist when we started. It didn't exist when I started in the workplace for sure. And I'm not sure that's what I thought we were gonna, we were doing, or what I was doing when I embarked on this.
So the recognition really feels less like it's a personal milestone, but it's more a validation that sort of staying on course and on something you really believe in can actually work.
Steve Smith:
In my work I talk to a lot of founders and invariably they say, I want to build a category. What we're doing is completely different. You're sort of in a unique position because you actually did build a category.
Deb Muller:
I know.
Steve Smith:
What do you wish you had known at the start of, what would you have done differently if you'd kind of known where all this was gonna end up, what the trajectory was?
Deb Muller:
I think one of the things I did, which I'm not sure I would change this as I didn't know, right? So if I had said I'm gonna set out to build a category, I think my aim would've been wrong. I set out to solve a problem. And in doing so, it was sort of this recognition that ER practitioners had no home.
There was no infrastructure. It was something that sort of emerging. There was no way to benchmark what good looked like. And we just kept trying to solve that and kept building it together. I remember early on we'd start bringing people together like, oh my God, I found my people. Like they were very much alone. And so it became kind of clear the problem was bigger than we thought, and it deserved a solution, not a workaround. And that category sort of emerged from it. I mean, it's truly time. It is. You know, it cannot be ignored in an organization anymore. You know, you cannot hide. Employee relations is you have your people, you have your people coming in and they're leaving, everything else in the middle is your relationship with them, that's a category.
Steve Smith:
Well, and it's interesting 'cause it seems like the best founders are, are like you. They're driven more by intellectual curiosity. It's just like, hey, this is interesting to me. I want to. Solve this problem and unravel this mystery. But you know, also there's, it seems like, you know, great companies are built as in part as a benefit of timing.
You had, you started a company and then, oh, social media, iPhones, political unrest,
Deb Muller:
For sure, for sure. Sometimes I look at the news and I'm like, oh, this is killing me. I'm like, no, it's great. You know, they're like, I never say it's great, but it's sort of not, it's sort of proving the point of what we're trying to do and yes, certainly the environment and what's going on has helped create some tailwinds.
It's.
Steve Smith:
Well, you know, one of the things I'm wondering is just like what, what's the best piece of workplace investigation advice you ever got? When you think back to the beginning of your career, that mentor who told you something that has that years later still sticks in your head? What was that piece of advice?
Deb Muller:
So it wasn't even that long ago. We have a woman by the name of Lori Die. She's a plaintiff's attorney actually. She gets very involved in our, I met her probably over a decade ago, and she's been part of our community, oddly enough, a plaintiff's attorney. But people love her and she always talks about, and it sort of something I talk about, and you just said it on the other side, but she says the worst thing you could do as an investigator is to have a lack of clinical curiosity.
So it is that curiosity, like always being able to ask the questions. Why did that happen? Like. Yeah, sometimes in employee relations we get it wrong because we look at sort of how something manifested itself. Oh, they violated a policy, or their performance is down, but we really need to look under it and figure out why. What is the root cause? Is it the person themselves? Is it something in the organization? Is it a policy? Is it a leadership issue? And I think that clinical curiosity in an investigation or any employee relations issues is so important because otherwise you're just gonna keep coming back to the same issue.
Steve Smith:
Another theme right now that I want to ask you about is this, the idea of vibe coding. There is this, there's a sensibility that, okay, all enterprise software is gonna go away because any smart person with enough Red Bull can basically code their way, you know, out of a problem.
When you hear that based on what you do, what crosses your mind?
Deb Muller:
Yeah. Good luck. I, good luck. First of all, this is, there is some real risk associated with this. Maybe I'll, a lot of times I talk to people about what we do the same way they say, well, I can just use my HRIS for this, or I'll use my ticketing system to do this. And I sort of go back to the same thing, like think about the healthcare system and you know, you go to your general practitioner for sort of these basic things across the board, but when the risk is high. When you have to have heart surgery, you don't go to your general practitioner, you go to the heart surgeon, you go to the cardiologist, you go to someone who understands it, and you don't build it along with them. You don't sit down and say, hey, let's talk about the open heart surgery. You're gonna do on me tomorrow.
What do you think we should, how do you think we should approach it? Like you go to the expert that. Stands it that understands the nuances. What can happen, what can go wrong, where the edge cases are, and how those are gonna be handled because the risk is so high. And that is the same thing with HR Acuity.
It's the same thing with investigations. That good luck vibe coding it. We'll see you in court like I just, you know.
Steve Smith:
Well, I could keep talking, but I think that is an excellent place to land the plane. So, Deb, thanks so much for joining. It's been a great conversation and congratulations on all your success.
Deb Muller:
Thanks, Steve. I really enjoyed it.
Steve Smith:
Deb quoted someone in her network with a line that's stuck with me: data, not drama.
The more I think about it, the more it captures something real about where employee relations is headed. The function has always had access to information. What it hasn't had is the infrastructure to make that information visible, consistent, and useful before things go wrong.
That's the actual problem HR Acuity is trying to solve. Not just investigations. Not just documentation. A shared language for measuring the health of the employment relationship. The P&L for ER, as Deb put it.
So the question isn't whether AI can help you move faster in employee relations. It's whether the decisions you're making along the way would hold up — in a deposition, in front of your workforce, or five years from now when someone pulls the file.
If you enjoyed this episode, make sure to subscribe to Work Tech Weekly on Apple Podcasts, Spotify, or YouTube. And I'll see you next time.
