The logic of Hiring Bad People – Or Maybe it’s Just HireLogic with Rich Mendis

Brian Fink: [00:00:00] Ladies and gentlemen, boys and girls, friends of all ages, welcome to the Ryan and Brian Show, a k a Sourcing School presented by the team at RecruitingDaily.

Rich Mendis: Ryan Leary. What’s going on, my friend? How are you?

Ryan Leary: What is up, man? I’m all good. We were just talking as we were coming in, getting back from SHRM, getting my my whereabouts back in and all my wires set up and I’m, I’ll tell you what, I’m excited not to be [00:01:00] on a plane because from Philly to Vegas, it just sucks.

Brian Fink: Okay, so real quick’s not good. Real quick, you’ve got all kinds of logistical issues in Philly right now, right? I like, I do. How did you get home from the airport? Because one of the main highways is not I was gonna say it’s down, but like that would be really legit. Down. Down.

That’s really, I know it’s legit down. Like I was like,

Rich Mendis: I don’t know if that’d be careless

Ryan Leary: or, yeah, no. So I go the opposite, so that’s north of me. I go south to go north. So coming outta the airport, I take the 95 South, not north, so I’m lucky. I’m lucky there, but it does, it still does affect traffic.

But dude, they, they have made a lot of headway. You can watch it live. It’s live stream on the repair. So they’re there now and they’ve got the entire thing cleared out. They’re just getting ready to put down the whatever they call it, that glass stone or something or other, where they’re gonna pay.

Hey, did you guys hear how they were gonna fix this thing?

Rich Mendis: This is the, this is 95, right?

Brian Fink: Oh, ladies and gentlemen, we are joined by Rich [00:02:00] Menis, the Chief Marketing Officer at Higher Logic. I guess that will be a good intro. Hey, rich, welcome to the program.

Rich Mendis: Thanks. Thanks. I was listening and I got in Enwrapped and then I realized I was, I actually was in New York for the day and when I took the train and when I came back my dad’s you thank God you weren’t driving.

And I was like, what happened? He’s talking about the 95, so I guess that’s what you guys are talking about.

Ryan Leary: Yeah. Yeah. And the way they’re gonna fix this thing is pretty wild. They’re so you can live stream the whole thing, just live to do, like live stream 95 or something like that. But they there’s a local company by us that creates like a glass stone or something or other.

And so they’re f they remove the entire road. It’s completely cleared out. Now they’re gonna fill it all the way up. And then they’re gonna pave over the stone. They doesn’t need to be compacted or whatever. They’re just going to blacktop the entire thing. And then they’re gonna build it from the outside in.

So they’re gonna start with the outside. They’ll have four or five lanes open, something like that. Once they get one lane, then they’ll move it and then expand to another [00:03:00] lane and move it again. It’s so wild. How are I gonna do it? I don’t know that I trust it. But it seems pretty I guess these people know what they’re doing,

Rich Mendis: It’s pretty cool. Ok.

Brian Fink: I wanna build off that five letter word you just dropped. Trust and trust it. Cause that’s why I wanted to have, that’s why I wanted to have Rich and the team from higher logic. On this call is that right now? So Rich, so real quick, for those of you who don’t know Rich and I’ll do a quick intro about hire logic.

They are really leaders in the space when it comes to their application, but also the thought leadership around AI solutions. And like you, you heard from the past episode, I’m totally obsessed with the AI and Mike Wilfred’s book. So go buy the book. But I’m obsessed with this. How are we gonna take inventory of ai?

How are we gonna use it in the HR organization? And I guess that’s a good segue for the word five letter word trust, because we want to trust this thing. And Rich I’m excited to have a conversation with you about listening, about job interviews, about the intersection of ai.[00:04:00] Without any further ado, maybe I should let Rich just introduce himself, rich,

Rich Mendis: and take it away.

Thanks, Brian. Appreciate it and thanks Ryan for having me on. It’s it’s a pleasure to be here and following what you guys are doing and love it. I’ve been in this enterprise software space my whole career. Started out in a company called Siebel Systems, and for those of you who know it, I’m probably dating myself.

That’s the pre to Salesforce, right? They invented C R M. From there I left and I did a startup in the carbon and energy management space, sold that to sap. Then went into mobile software just when mobile apps were taking off. And there’s some interesting parallels there between how mobile impacted HR and how AI is impacting hr.

We can talk a little bit about that. Sure. Sure. Did a company there, sold it to a integration company called tibco, and then found my way into hr. So I’ve been in enterprise software my whole career, but it’s the first time I’m really going deep into hr. And I think it’s an interesting time to be in the space.

I think, hr and recruiting is being transformed by [00:05:00] what’s happening in ai and yeah, happy to be on to talk about that. Let’s,

Brian Fink: There, there’s so many different places, so many different actions. We can take this in. Rich. I guess my first question is, should we be

Rich Mendis: threatened by the machines and the robots are coming to take our jobs?

Yeah. It’s a, it’s a great question. The honest answer is maybe right, but nobody can really tell for sure in which jobs. And so that’s a broader discussion, right? I think, we’ve been through. As human civilization, many of these kind of big changes right from the agrarian to industrial society and industrial to service.

And we’re going through one of those changes now. I truly believe we’re going through a change as significant as one of those. And we’re in this transitionary period to a post AI world. And I think, human, there’s this law called Amaras Law, right? And it states that.

Humans have a tendency to overestimate the impact [00:06:00] of technology in the near term, but underestimate the impact in the long term. And I think that’s what’s happening here, right? So people will freak out. Initially I’m like, oh my God, the world’s gonna end. We’re not, we’re gonna lose all our jobs. And it won’t happen.

It won’t happen right away. But what will happen is incrementally, you’ll start to see changes to jobs and changes to the way people do things. And over time, it will become unrecognizable from what it is today. Oh, because technology is not a linear change. It’s a, it’s like almost exponential and it has ups and downs, and humans are not good at forecasting what that means.

Oh,

Brian Fink: humans are horrible at forecasting everything, right? That the, I think of the interview question, where do you see yourself five years from now? God being asked in 2019? God, I don’t think anybody saw. I just

Rich Mendis: I hate that question.

Ryan Leary: Always, how do you answer that question?

I Rich, you’re on, you’re in a job interviewer now. Where do you see yourself in five years?

Rich Mendis: Like, how do you answer that? I’m a

Brian Fink: serial entrepreneur and I’ll go into another [00:07:00] entrepreneurial endeavor. I don’t know.

Rich Mendis: Sorry, rich, I didn’t mean to hijack for you. I’m sure there’s research out there right?

On what the average tenure is at companies now. It’s gotta be less than five years, right? Yeah. It’s insane. Sure. But yeah, so I, to answer your question, I think there’s gonna be massive change, but it’s going to be At a different pace than people think. And so you’ll start to see it slowly.

So I bet, like in certain job functions, if it’s not already happening today, qualifications will be things like, how do you use AI in your job? If you’re hiring a content marketer today, right? They gotta know how to use generative ai, right? So already we’ll start to see. Job descriptions and position roles change.

And over time, that’ll just become more and more to the point where a content marketer, three or five years from now, what they do is gonna be very different from what they did last year. Okay. That’s,

Brian Fink: go for it. Ryan. I’m googling this.

Ryan Leary: The the generative AI is interesting to me because I’m in marketing, right?

You’re in [00:08:00] marketing, rich. And I’m curious, leading a marketing function from a product or solution side. Yeah. How does your team, and I don’t know if you can go into detail about your team, but how do, how does one use generative AI to help with a B2C or a b2b b marketing solution?

Rich Mendis: Yeah, I look, it’s coming fast and furious in marketing.

It’s funny We use an agency and you know that one of the creatives on that company recently, left to start up an AI startup, right? And he was telling me that, look, it’s changing so fast that what marketing agencies look like today will be completely different, like in a few years from now.

And the reality is it’s impacting almost everything, right? So everything from written content, To so you can generate right, like blogs and things like that. And it’s not that it’s the final output, but it gets, you say 70, 80% of the way there, right? You can generate images, prototypes, things like that.

You can generate music now. I was making. Seeing a demo video and [00:09:00] I got, we have this really good salesperson who is actually trained in as a professional voiceover person, and I use the software to train her voice so I don’t have to bug her. I just type out what I want her to say and it says it in her voice.

So it’s impacting everything in marketing. Yeah.

Brian Fink: Okay. Alright. So real quick, I just did a quick search for generative AI in quotation marks, open parentheses, marketing or marketer, closed parentheses in the United States and look for jobs.

Rich Mendis: Across the us.

Brian Fink: There are 1,412 jobs that mention generative ai.

Everybody from Bertram Capital to TikTok. To a company called Lead Expert. Yep. Ogio, Amgen. Amgen Pharmaceuticals. So here, this is, okay. Yeah.

Rich Mendis: So here’s the, so here’s the thing. Brian bookmark this on your calendar. I’d love to come on, say, 90 days from now on your podcast and let’s look at the same query.

All right, I can do that. All right. Copy

Ryan Leary: that out. [00:10:00] Copy that out as we’re talking. Where we at June 16th

Rich Mendis: today you found about what? A thousand right. 1500. Yeah, 1500. Let’s look at the count. 90 days from now.

Ryan Leary: All right. June, July, August, September. So headed into, how about this? Let’s do it on we’ll do this prior to

Rich Mendis: HR Tech.

There you go. Okay. Prior to HR

Ryan Leary: Tech, we’ll get it, we’ll get it set up

Rich Mendis: and I gotta come, we gotta we gotta meet you guys there in person. Oh, a hundred

Ryan Leary: percent. A hundred percent.

Rich Mendis: Sounds

Brian Fink: like a plan. Now about that plan though, Richard, I gotta go back to this. Is that. We’re talking about a lot of things going on in marketing.

I wanna bring it back to HR and talent acquisition

Rich Mendis: recruiting. Absolutely.

Brian Fink: Should there be rules about how this is used? I know we have rules at McAfee where I am and I’m curious what are there best practices yet, or does somebody need to take leadership on this? What does this look like?

Rich Mendis: Yeah, so great. Great question. And I think, for anyone, and hopefully a lot of HR people will be listening to this, look, I think that this is the single [00:11:00] biggest opportunity for HR to take a strategic role in organizations as they adopt the use of ai. Let me rephra, let me rephrase repeat that.

I think that this AI transition. Is the biggest opportunity for HR leaders to become strategic across their organizations as they transform and adopt ai. Okay. And the reason I say that is, one, there is plenty of HR tech that’s gonna be AI enabled, and along with all of the questions about the ethics and things like that, but all of the regulation and there’s regulation at the municipal, state, and national levels happening all around the world.

We’ll be focused initially in two areas, privacy and discrimination, and that second one especially, who does that impact hr. So HR within a company, if you look at all the functions from sales, marketing, [00:12:00] back office, logistics, shipping, et cetera, HR will be one of the first functions in an organization that will be impacted by AI regulation.

So not only can they be the early adopters of technology, but they can be one of the first to figure out what it means to use technology under regulation and that best that experience. As one of the first organizations in a company to have that experience will enable them to advise other groups in the company on how to adopt AI solutions in a in a way that’s in compliant with regulation And part of an HR job.

HR function is learning and organizational development, right? Training is part of their core responsibility. So in addition to having the experience of using AI in a regulated way, you also have the function of training others in the organization to do it. So this is an opportunity for HR to step up as a strategic advisor to the organization and to the C e O and the [00:13:00] board to.

On how the organizations can adapt because they have to adapt. There’s no choice. If organizations don’t use AI to become more productive and efficient, it’s literally like them not using software or the internet. You just can’t imagine it. And so they have to use it. How do they use it? In a, in compliance with regulations and being good stewards of privacy and discrimination.

Ryan Leary: Rich, one of the, one of, one of the themes of, at least in a talent acquisition audience at SHRM, which wasn’t big by the way, but it was big enough for me to find, so that was good. Was recruiting how recruiters are going to evolve, recruit, the recruitment function is going to evolve and really get away from the.

Sourcing and identification of candidates. Yeah. And moving more into strategic partners around skill pathing, upskilling, internal mobility, and really facilitating and coaching employees through their journey in the organization. So I’m curious to get your thoughts [00:14:00] here. One, what is your thought on that and two, yeah.

How does all of this, what we’re talking about now, especially given that HR has this opportunity, where does talent acquisition fall within this?

Rich Mendis: Sure. So it’s an interesting thing, right? Like I’ve heard it thematically across a lot of the HR conferences this year about this movement to skill, right?

To becoming a skills-based organization. And you see it with a lot of talent management systems and methodologies and stuff. What’s interesting is that in order to do that, in order to become a skills-based organization, you have to start getting good at identifying people’s skills, right? And.

If you look at the, if you look at the hr, the human capital process, starting from when source, hire, evaluate, onboard, train. Mentor throughout their career and then end of life or exit interview that person throughout the whole life cycle. There’s, so there’s tons of technology across that spectrum, right?

Across those process. There’s tons of tech to, applicant tracking systems to help you post jobs and get candidates, right? [00:15:00] Onboarding technology, talent management, technology, et cetera. But there’s one area, there’s one area of the HR process that has been the same for the last, I would say 70, 80, maybe a hundred years or more.

And that’s the job interview. The job interview has not changed fundamentally.

Brian Fink: Whether you Oh, the interview, it’s the job interview, the thing that has nothing to do with whether or not you can do the in job.

Rich Mendis: Correct. The ar arguably the most important function, right? Yeah. And what’s interesting is if you think about that, HR is generally well trained to do that, but it often goes to a hiring manager at some point, right?

And those guys who’s not they’re not, they like there is no course in school about interviewing. There are very few people doing, like learning about how to interview in their career. So it’s by and large left up to, whether you aspire to be a good interviewer or not, or you’re in, instinctually a good interviewer or not.

Now as a result of this, The research, if you look back on, on [00:16:00] research, in, in on interviewing, we are extremely biased, right? So I read a stat from Harvard Business Review that said somewhere between 85 to 92% of all hiring is based on gut instinct. That means I

Ryan Leary: would not disagree with that.

Rich Mendis: That means we’re essentially interviewing people to confirm our biases. If you look at the things like resume screening, I read a stat that said if your race is revealed on your resume, minorities have a 50% lower chance of being called back. When you take the same resumes and change the name white male, Caucasian names have a higher chance of getting a callback.

So humans, in the absence of ai. Putting all AI aside for a moment, humans are extremely biased in the hiring process, right? And so what we’ve tried to do at Higher Logic, [00:17:00] as one solution is we are focusing purely on that problem. And you can’t change how people do interviews. It’s very difficult.

Or you’ve been doing it for the same way for so long and nobody takes training on it. So you can’t change how someone does an interview. So we built a solution leveraging machine learning and ai. That simply listens to the conversation, right? So if you’re familiar in the sales world with solutions like Gong and Chorus and a boma that use machine learning to extract sales intelligence from a conversation, hire logic is essentially that.

It’s the gong for hr, it’s extracting interview intelligence from a call with a candidate, whether you’re a recruiter or hr. And the things we, so we listen to the call and we’ll extract, we’ll listen for things like look. What are the candidates skills, job experience, functions, et cetera, but also what are they, their likes, dislikes, aspirations and go deeper than what’s in the resume.

Now, typically a resume is backwards looking, [00:18:00] right? What’s not in the resume is usually what they want to aspire to do or what they don’t like to do, right? Some of the cultural or soft aspects of a candidate, we try and listen for those elements and extract it. We also try and understand the quality of the interview.

And I don’t just mean, things like whether you said or ah, but how much of the job description did the interviewer actually cover? Did they just talk about the weather and baseball for 45 minutes or did they actually cover the topics and job functions that you’re looking for in the job interview?

And very importantly, did the interviewer ask any questions that could lead to potential discrimination or bias across the E O C categories, age, race, gender, et cetera. Rich, you bring. Yep. Go ahead.

Brian Fink: I wanna jump in there for a second. Is I was having a conversation with Faith the other day on H R T X.

She was one of the presenters and she was talking about neurodiversity.

Rich Mendis: And I want to know, and

Brian Fink: she made a comment that when

Rich Mendis: [00:19:00] you ask certain

Brian Fink: questions at the beginning of the interview to break the ice with a candidate, that you actually increase the bias of the interview for somebody who is narrative urgent because they don’t know.

They don’t know what you’re asking.

Rich Mendis: And so my question is, are you, how are you, and look

Brian Fink: like how are you changing a population that is resistant to, largely resistant to change? There’s a low barrier to entry to be in recruiting. In a lot of states you have to have a license to cut hair, but to be a recruiter and tell people what to do with their life,

Rich Mendis: you don’t.

Like I’m all

Brian Fink: in on this. I really am. I’m not just playing to the field here. Like I think this could be great to teach me how to do a better interview. Yep. How do you get people to get past that initial bias that they have?

Rich Mendis: So I think the first question, the first step, Brian, is we have to listen and understand how you’re doing it today.

That’s the very first step. There were [00:20:00] something like 11 million job openings in the United States at the beginning of the year. If you assume conservatively that. Each open position generates three interviews. Okay? That’s 33 million interviews happening a month, or 1 million interviews happening a day.

And to your point, there are no standards applied to whether those interviews are done well. Nobody’s monitoring whether they’re asking any bias questions. Nobody’s, neurodiversity is like a third or fourth order problem, right? Are you really asking good questions? That’s not gonna put the candidate off balance.

So the first are when we talk to customers and they get all excited with this tech, we say, look, baby steps. First of all, you have no idea what’s happening in the interviews. So just have us listen and let’s start to gather insights. Let’s start to gather insights, right? So we learn about the candidate and we learn what the questions are that you’re asking.

The next step is, hey are your recruiters or interviewers asking the right questions? Which interviews tend to get the best [00:21:00] responses or are there recruiters that have very high placement rates? And if so, why? What is it about their interviews that’s getting them high placement rates, right?

Once you understand that, then you can say, now, We figured out, Hey, this, these interviewers ask these really good questions. How do we get the other interviewers to ask those questions? Then we have something called an interview companion app, where it prompts the interviewer to ask those specific questions.

Now, interview guides are nothing new. Many ATSs have interview guides. The problem is, despite all the training and all the guides, you don’t know if somebody actually asked those questions. We’re listening, so now we can tell if the recruiters actually asked those questions and what the responses are.

How many webinars

Ryan Leary: have we all been on? Where people, it’s, marketing webinars specifically, or even recruiting messaging, webinars. Use this email, use this emoji, do this subject line, because it works. I get 90%, 80%. [00:22:00] That’s a guess, right? Some people really do track. But, and this and think, if you haven’t seen Higher Logic, and I know this isn’t a pitch for Higher Logic Rich, so I don’t want to make it that, but I was blown away at the stuff he was showing me, and I think if you haven’t seen it, you definitely need to go in and check it out because all of the, all, everything that Rich is talking about here is measured.

It’s measured with accuracy, and it’s very different than anecdotally saying, I get an 80% response rate. It you actually know what your what your your candidate or your interviewer is doing.

Rich Mendis: And by the way, Not a pitch, like honestly, like you can go to the site and sign up, it’s completely free.

There’s a free for life version and just try it and just see how AI can, do you think it’ll be helpful? And we tried to make this super affordable because if we can even help decrease bias by a small fraction of a percent, it can have tremendous impact. And Whether it’s Hy Logic or some other tool.

I think the interesting thing here is I [00:23:00] think we’re at a point now where AI and HR can actually help, because all the research shows that when we interview or screen candidates as humans, we are simply just biased. It’s just part of our nature. It’s unconscious despite our best efforts to be objective.

And I’m not saying AI is perfect and I absolutely think we should be very careful about. How AI is applied to HR to make sure we avoid bias and discrimination, hallucination, all that stuff. But the reality is it’s not one or the other. It’s some balance in between. What is the combination of AI and humans that can come together and avoid a, what today is a extremely biased process.

That’s the goal. That’s the challenge for all the HR people listening.

We’re talking about you silenced. No. I

Brian Fink: like I did my research on this beforehand because I was like, what are we gonna do? What are we gonna build? How are we gonna build it? What’s gonna be [00:24:00] the adoption? How is adoption gonna work? And this is not, we had a conversation with two mics about generative ai and we had generative AI in this conversation.

But what this is really doing is, if I understand this correctly, this is listening and providing data on whether or not bias exists. That’s part one, but part two, it’s also standardizing the interview process in a way that we make for a better candidate

Rich Mendis: experience, a better candidate experience, and better as let’s say less sub, less subjective.

Evaluation after the interview. So for example, and I know your audience can’t see this, but I’ll just, I’ll read a sentence. Okay? This is a sentence. So we listen to, let’s say a 45 or hour long interview, right? 60 minute interview. And the candidate is gonna say a bunch of stuff. Interview is gonna say a bunch of stuff, okay?

There’s also some alsos of machine learning, first of all to understand, hey, who’s the candidate versus [00:25:00] the interviewer? If you don’t want to have somebody to go pick it, we gotta listen and use machine learning to, to figure out who’s the candidate and who’s the interviewer. We can do that with a 98% accuracy just listening to the conversation.

Okay. Then if you think about what most interviewers or recruiters do, we immediately, after a call, they need to summarize something about that candidate, right? They need to create a short summary of a 60 minute interview that they can pass on to someone if they want the next person to interview this candidate.

Now you, that’s usually subject to the memory and the bias in your head. Okay, so this is, I’ll just read a sentence from the summary that higher logic generates from listening. It says, and this, these names have been changed to protect the innocent. Melanie is an electronics and communication engineer from the University of Arizona who has taken an interesting journey through consulting, sales, and product management.

Okay. Now, Melanie didn’t say these words, th this sentence verbatim, but she mentioned [00:26:00] someti throughout this 60 minute interview that she was an electronics and communication engineer. She went to the University of Arizona, and she’s a, got experience in sales consulting and product management in some linear timeline, right?

So the AI listened and pulled together this sentence as a summary because it felt that these were pertinent pieces of information. Now, When I show this to somebody, there was a very interesting HR person. She says, Hey interesting, according to who, right? It says she took an interesting journey according to who, and I said that’s a great question.

The abstraction that we applied to this, the machine learning algorithm felt that this was an interesting journey. And it said it, but that’s subjective. But what’s not subjective is that she was in consulting, sales and product management, and she’s a communications engineer from the University of Arizona.

So it’s an interesting, it’s interesting to see how these things, it, it would take a a recruiter, probably a good. 10, 15, 20 minutes maybe to synthesize [00:27:00] all this information across all the notes, assuming they even took the notes down correctly and remembered it into a succinct summary like this.

Whereas here, it’s done immediately for you. I remember. So I wanna

Brian Fink: Go ahead. No. You go for it. Cause Yeah.

Ryan Leary: I, what I was gonna say I remember agency days when you had to pitch your candidate to the hiring manager after you spoke with him. This is it right here. That’s it.

And every, I tried to be different, but every summary I put together had the same damn thing. And God, God bless the hiring manager’s Got it. Cuz I thought they were good. The hiring manager probably is he just have a template here, like what’s going on in, in this?

Yeah. Something like this. Rich, you said something that, That Mamie perk up, you said it felt like for their journey. When people who are in the know you’re like, you work with this this is your baby, this is what you do. They talk that way about it, almost like it’s a companion.

It’s their friend, right? No it [00:28:00] felt this way and this is why it said it. And they defend it and talk about it as if it were a human sitting next to them. That’s really interesting to me. Yeah.

Rich Mendis: And it’s not right. So I think we need to be careful of this. It’s not, what’s interesting is what large language models are, cuz that’s what we’re ultimately talking about are basically probability engines.

They’re probability engines that have been trained on the sum or average of knowledge from where it gets the data to predict, right? The next word in a sequence of tokens. And in this case, the large language model felt. That this journey that this person took, not felt, but the large language model calculated the probability of the word to insert here as being an interesting word, right?

And so ultimately, pick that. Nobody can truly define why pick that because it’s 175 billion parameters, for example, in GPTs large language model that with weights and all the things that prob, that probabilistically [00:29:00] calculate the word. But ultimately it’s a probabilistic calculation of what to write here based on everything that’s been trained on.

And therein lies potential for bias, hallucination, et cetera, because it’s only as good as the data that it’s been trained on. And it’s impossible to audit all that data, and it can sometimes output things that don’t really make sense because it’s just a mathematical probability that’s outputting something.

It doesn’t really understand what it’s saying. And so sometimes you might get what’s called these hallucinations, but these are the things that as software developers using this tech, we have to make sure we minimize right, the negative impacts of those and minimize hallucinations bias, et cetera.

So the question that I wanted to ask, question that I wanted to jump in here about

Brian Fink: actually ties in real nicely with what you just shared.

One of the things that I am guilty of doing in an interview is that a candidate will say to me, [00:30:00] Hey Brian I built this using JavaScript and on the front end and I built this on the back end and I scripted Python and I will say, Without question, and if you’re candid of mind listening to this, I’m sorry.

I will say, wow, that sounds hard, or, wow, that sounds difficult, or,

Rich Mendis: wow,

Brian Fink: can you tell me more? And I don’t, I, I don’t that there’s, that’s nowhere on an interview guy, right? Yeah. It’s just that I’m caught up in the story. Yeah. How do we operationalize that answer to remove my bias

Rich Mendis: from being,

Brian Fink: enthralled with the candidate?

Rich Mendis: Yeah, look I don’t necessarily think that’s a bad, th this is somewhat of a philosophical debate, right? And in terms of what’s the quote unquote best interview style, I think that, that’s a great question to ask, especially if it’s germane to the position, right? Because you’re trying to get them to expand on and answer, and you might learn a little bit more about their thought process that goes beyond the resume, right?

Yes, they have JavaScript skill, but [00:31:00] how they think about solving the problem as a developer and approach it. It could be much more interesting than whether or not they know the syntax of JavaScript. And so I don’t necessarily think that’s a bad question. I think it makes for after the interview, when you’re looking at it trying to create notes, it makes for an interesting challenge because if you’re trying to look for the answer of what, what the level of skill is of this job, a developer, the results might be all over the interview.

You, it may be at the beginning and then you went somewhere else off a tangent and then you came back to it. And so one of the things that, that you know we do is we’ll try and listen for a question and then we’ll try and put the response regardless of wherever it, it was said in the interview.

So you may have veered off a track on another question and then come back to JavaScript and we’ll bring that JavaScripts thing back up to, to, where you asked the question. So if you’re trying to digest it afterwards, it’s morning. It’s more easily digestible. Interesting. How do we tune

Brian Fink: AI though?

For the core values of an organization, of a company, right? Yeah. I like to think that when we’re [00:32:00] asking questions, behavioral based interview questions, star interview format questions. Yeah. That we’re asking about those behaviors and those are in the past and this is predicting the future

Rich Mendis: as DDI has talked about before.

How do we do that? Yeah. So I think the first step is, can you. Train the model, right? To pick up on, on things like not just listening for keywords and skills, but actual functions. And then how do you translate that function into a, let’s say, a characteristic or attribute that’s more of a soft skill.

So I’ll give you an example, right? So in this interview, I think there’s one here. Yeah. So she said I apart from helping the marketing, right? Let me turn off some of these things so we can see it. Apart from also helping marketing I’m leading a team of two to three technical writers.

Now, leading a team of two to three technical writers was flagged as a job function, and if we turn on this other option here, it was also flagged as a leadership skill. Or leadership quality and other things. [00:33:00] If you train the model enough, okay, it can start to detect other job functions and other things like aspirations, likes, dislikes, et cetera, that can help you, go a little bit deeper than the resume, right?

And then you can summarize it and say, show me all of this candidate’s aspirations, right? Or all of this candidate’s interests, right? And it may not be a hundred percent. But it gives you very quickly without you having to go through, or remember or go through a copious amount of notes what might be some of their aspirations in these other things, which then helps you perhaps match to culture.

Now, what’s interesting, Brian, is where we’re going with this is the first step we have is collecting this data right? Whatever you’re comfortable with. The second step is now that you’ve been collecting data, and think about like you, you mentioned you’re at mc, was it McCaffey? I think it was, yeah.

So I don’t know how many interviews you guys do every month across all your interviewers, but I’m assuming it’s a fairly large amount. And what goes back into your ATS normally is whether or not, the recruiter [00:34:00] took good notes, but imagine all this information, which, and everything I’m showing you here on the screen was generated without taking.

Any notes just listening all of this, going back into an ATS and now imagine a Google-like interface on the ATS that’s using natural language processing where you can say, can you show me candidates I’ve spoken with in the past that no Java have experienced working with offshore developers and are looking to do something more in the field of x.

And having machine learning go through, and now you have all this data you gathered about somebody that’s beyond the resume and pull up candidates and tell you why it feels like this candidate might be a good match for that natural language query. That’s where we’re going. The holy grail, right?

The holy grail of this is. Nobody’s gathering any data during interviews. But imagine if you do, if you do gather this data and you’ve got in your talent management system, right? The performance of those [00:35:00] people, right? Whether they’re, how long they’ve stayed in the job how, what their performance is.

Imagine, over time you can start to look at, hey, are there questions asked during an interview or responses given or things said during an interview for a position that tends to predict better performance in that position or longer tenure?

Ryan Leary: Rich, I think you literally just broke his mind. Oh

Rich Mendis: dude.

Dude, we are,

Ryan Leary: I know the audience can’t see you right now, Fink, but you literally look like you just have brains gushing out of your ears. So

Rich Mendis: If you look at, if you look at this from, look, there are, if you think about value that, that, AI can bring to a company. Okay. There’s, and I know the audience can see this, but imagine a two-dimensional chart, right?

Tactical on the bottom, strategic on the top so what is the tactical value it brings versus strategic, and then immediate value on the left and longer term value on the right. You can look at your [00:36:00] portfolio of AI products in different organizations as to where it brings value, right? So for example, with a tool like Hire Logic, the day you turn it on, you’ll immediately get some productivity benefits, right?

Less note taking, faster to get interview feedback, et cetera, right? That’s tactical. Immediate value. There’s more strategic, immediate value, which is, Hey, I can now get, use these insights perhaps to be more objective about my evaluation of the candidates. Then there’s longer term value. Am I improving my interview process, right?

Can I use this to coach my people? Can I make sure they’re being compliant? But the strategic longer term value is if I’m capturing all this great data when I’m doing interviews. And I’ve got a talent management system, which today we know is not integrated with an ATS that has performance and tenure.

What if I take all the insights from a solution like urologic, drop it into a data lake, bring just tenure and performance in, and now if I’ve interviewed and hired dozens of people across a [00:37:00] position, I can start to potentially find correlations between things interviewers have asked or things candidates have said, and whether they actually perform well in that job or have long tenure.

And that Rich again. Yeah.

Brian Fink: No. Rich. Again, I know that our audience can’t see what we’re looking at right now, but there’s one part that jumps out to me is enriching the ATS and CRM with insights for sourcing. Yeah. Okay. As a sourcer goes recruiter goes sourcer back and forth. I’m just wondering, does that mean that I can boo a conversation between me and an, between a candidate and a recruiter that has long since

Rich Mendis: left the company?

Sorry, repeat that again. Does it mean that you can, what

Brian Fink: Can I, so let’s say I’m looking

Rich Mendis: for Java developers. Yep.

Brian Fink: Okay. Our JavaScript developers as the example was earlier, and I type in that I want to find somebody who knows how to do higher order components. You’re nodding your head.

So this means that I [00:38:00] can go back and I can search my entire database for my silver medalist who knew higher order components and I’ve got short list before I ever post the role as to who I need to have a

Rich Mendis: conversation with. Correct. Because something like short order com, that may not be on a resume.

And everybody’s struggling to create a one page resume, which means you take out all the details, right? That’s the type of stuff that would surface during an interview, right? Or something you’re talking about a lot of technical people, so it could be somebody who says, look, yeah, I’m an AI engineer, but I’m really specializing in VI vision, right?

And I really want to do vision. And that’s not something that might be on a resume. You may have a need for that type of a candidate later. And if we take that data from higher logic and save it into your ATS, you now have access to that. Okay, so

Brian Fink: let me do a different one is the other day I was recruiting

Rich Mendis: for,

Brian Fink: Revenue accountants revenue, sorry sorry.

State and local tax revenue [00:39:00] accountants. Yep. And I didn’t know what the term salt meant, S a l T, but yet they were using it. And I would’ve glossed over that in a resume. But this is an, a perfect instance of where I could have just searched the conversation and found out, oh, that’s

Rich Mendis: exactly what’s going on there.

Correct. And by the way, that’s a really good one because in, in many, in industries there are these esoteric acronyms, right? Or certifications that don’t always, These appear on resumes or, they may have, maybe you don’t have the resume, but you have the interview and you’ve as a recruiter, you just didn’t enter that in, you didn’t enter that into your notes. Ha, catching those things is a great example, right? Of, of skills and industries that may not be in the resume, not be in your notes, but now you know, it’s been recognized by machine learning and put into that ATS, right? And by the way you guys have seen this, right?

This is fascinating. Like I haven’t been in the HR space for that long, but I’ve talked to now in the last year, tons of HR people. [00:40:00] I am shocked at how many companies are always migrating from one ATS to another. And coming from the, do you guys concur like that? That happens a lot right

Ryan Leary: now.

It depends on how good the salesperson

Rich Mendis: is, depending on how, now what’s interesting is if you think about enterprise software, let’s step back a little because there’s E R P C R M, HR Tech, all this stuff, right? What is it that makes switching costs easy or hard between software? It’s how tightly it’s integrated into your process or how deeply it’s integrated with how much proprietary data is in the system.

Okay, my thesis. And so if you look at, integration companies like what tibco and those companies do or SAP and E R P company, very high switching costs because it’s tightly integrated with a custom process or you have a lot of proprietary data in it. One of the reasons I think it’s very easy for HR to switch ATSs is because they don’t have enough proprietary data in the systems.

It’s just resumes. Migrating from one [00:41:00] ATS to another is as simple as exporting the resumes from one and uploading it into another, right? However, if you have deep information, okay, about your candidates leadership aspirations, likes, dislikes in various different parts, maybe custom objects in a bullhorn system, right?

Or integrated into a success factors data lake so that you can search for it. The value of that ATS becomes much bigger and much harder to move away from. That’s a thesis. I may be wrong, but that’s one of my, one of my feelings that we, the reason we see a high, low switching cost between ATS is there’s not a ton of proprietary data in it.

Rich, you

Brian Fink: mentioned the word bullhorn. You know I am an agency

Rich Mendis: recruiter turned Corporate recruiter.

Brian Fink: Yeah. So I’ve used Bullhorn and I loved using the system. I know there were people who had problems with it, what have you. Yeah I’m just thinking this application that this is just so [00:42:00] inspiring to me that we could have.

An agency tomorrow that could port all this data with resumes of can Sam in their database and they could act that much faster. That is my wt

Rich Mendis: a f. Yeah. So imagine this. I’ll give you a better one. I’ll give you a better one. Okay. Let’s take recruiters for minute. We’re talking about hr.

Let’s switch to recruiting and staffing today. If you were to ask a VP of operations for a recruiting company who has, let’s say, a hundred recruiters, right? Fairly decent sized, but not super big recruiting company. Hey, can you give me at a glance how the performances of your recruiters across interviewing and whether any of them are asking non-compliant questions can, do you think they can answer that they can’t?

No, they can’t unless they audit. Unless they audit or sit in on calls, they can’t do it right? So I showed you what we gather from One Call. Now imagine aggregating that across all your recruiters, right? So here are all the interviews that happened. These are [00:43:00] all the interviews with potential compliance risks.

And if you drill into the team right here, is all of the team members doing interviews? What the average interview length, Hey, Rachel only does seven minute long average interviews. Why is that? Let me drill down. Or Jen and Vivek. Have, greater than a 3% compliance risk, I may need to do some coaching with them.

Instant up to date, as of the last call, visibility into the performance of your recruiting team. Alright, different question. What,

Brian Fink: so that we’re talking about team, I wanna focus on, I don’t wanna focus on team, I wanna switch and focus on the individual. You have a lot of people who have been. Laid off that rather than looking for a new role, they’ve started a new recruiting agency.

They’ve gotten set up with a bullhorn. You said that your tool hasn’t always free version of it. Yep. That I imagine would be great for an individual practitioner. Am I wrong? Is it built for somebody who’s running an individual practice that wants to hone their skills, become

Rich Mendis: a better [00:44:00] golfer?

Yep. So if you’re a product, if product marketing or product management or enterprise software, we’ve taken a go-to-market approach that’s called product-led growth, which means you don’t need to talk to one of our sales reps. You don’t need to put your credit card in. You just go sign [email protected] and you can start using it.

And if you’re a s, individual recruiter maybe you do, five, 10, interviews a week, something like that. It, you’re covered and the data is yours. If you get, if you start to grow it, we price it at $49 a user a month. It’s not expensive. And so yeah, you can use it, we have small recruiters.

We just closed the deal with the a single recruiter company. And it’s interesting, he actually wants to use it for intake calls. He said, yeah, this is interesting for interviews. But I wanted to join my intake calls, which we hadn’t really thought about. So he wants it to listen to the requirements that the customer is talking about to help him create the right job position, then go interview the candidates and find the match.

So now we’re looking at, okay, what does [00:45:00] an insights view look like for intake call, where the output is not a candidate, but perhaps a job position.

Brian Fink: That’s an interesting, that’s an interesting use case. That’s a very interesting use case to make sure that you heard everything the hiring manager was laying down, making sure that they’re, if you will, to almost have somebody read the order back to you.

And I mean that with the most reverence possible because I don’t think that the intake session

Rich Mendis: should

Brian Fink: be where you’re ordering takeout from a

Rich Mendis: recruiter. Yeah. That is, that’s awesome. I was at the Bullhorn Engage Conference, I think it was, what, two weeks ago maybe? And you hear all these KPIs, right?

Redeployment rate, candidate placement rate database utilization, big one, database utilization, right? Hey, if you want to find existing candidates in your database what’s the number one thing that needs to be there? It’s rich information about the candidate. And today, most of the time it’s the resume.

And [00:46:00] if you’re lucky, some interview notes, right? But what if every time you talk to that candidate, and we have big recruiting companies, Dion is one of our customers, formerly DySIS, right? Thousands of recruiters. They want this information every time you talk to that candidate, not just when you’re interviewing them for the first time.

So every time you’re touching base, you’re updating the ATS cuz they’re not updating their resume, but you’re updating the ATS with their latest experience.

Brian Fink: W with that latest experience. That kind of gives me pause here, is that rich, like when I call a candidate, Do you mean that I should be using this tool also when I’m prepping candidates, when I’m like, Hey, you’re gonna meet with rich guy’s vivacious.

He has a good personality, very intelligent, gonna ask you a lot of questions. Do you think that I should have this also listening to me when I’m giving that advice to

Rich Mendis: the candidate? That’s up to you. I think if the an, the answer is if you feel like a discussion. Will surface some insights that you feel [00:47:00] are useful to have in an ATS, then?

Yes. Okay. All right.

Brian Fink: Because cuz you know when I have that conversation with the candidate and I tell them, oh, you’re gonna meet with this hiring manager, you’re gonna do this, you’re gonna do this, Hey, there’s no if you have a question about benefits don’t ask the hiring manager come to me. I’m the pro.

Also, just real quick, rich, I want to make sure that. That I got this right. You said that you were looking for one 60 with a total upside of three 20. Does that sentence, does that still sound competitive? And you say yes, right? Yeah, I’m trying to do the pre-close

Rich Mendis: on every call. Yeah.

No, I think it’s an inter, we listen for things like KPIs and action items and things like that too, right? So it, I would say your default should be just invited because. You never know where it could come in handy, for things like that. And you can decide what gets pushed to an ATS or not, right?

But like action items for example. If you’re one of those people that says, I just want, something to just tell me what my action items are from every call I took, you [00:48:00] can use it for that if you want.

Brian Fink: Yeah. That’s, it’s fascinating. That’s, it’s ridiculous. And ridiculous.

In a good way. Yeah.

Ryan Leary: And Rich, this has been a ridiculous conversation. I know we, we took a ton of your time here. We really appreciate this, but before we wrap up here and let you go let’s dumb it down for the, and they’ve gotten this far, they’ve gotten the NBA and everything we’ve been talking about, but let’s dumb it down for the listeners out there who aren’t dumb.

Yep. Saying that they’re not dumb. But what are three, four things? Something? Yeah. Three, three to four things that they truly need to understand in the next two to five years in their career as a recruiter recruiting manager in within embracing, yeah. Everything that’s happening

Rich Mendis: today. Great. Great question.

I think, look People. I think generally, the phrase software is eating the world and, data is king, right? Content is king. Those apply to the [00:49:00] recruiting space. I think more, pretty strongly, right? And so if you’re a recruiter, I think you, it’s it’s, you shouldn’t be afraid of what’s coming with ai.

I think you should actually. Really be pretty excited about it because a lot of the mundane parts and aspects of the job, like taking notes and who actually likes going into an ATS and entering data. Honestly, like I, I hardly think that’s like the highlight of a recruiter’s day, right? So imagine if you can have solutions that just listen to great conversations that you’re having, lets you engage with the candidate.

Okay, lets you actually engage with the candidate instead of looking down and writing notes, extracts that in those insights, lets you add your own notes and puts it into the ATS for you automatically, and then lets you go on to the next one and the next one and not, you not have to worry about losing the fidelity of that memory coming back and being able to find candidates who fit the, the new positions or openings more, most accurately.

I can’t think of a better use of [00:50:00] tech. It’s making your life easier. It’s helping, you evaluate candidates more objectively. It’s helping you place hopefully people in the right jobs, at a higher rate, reducing the turnover, things like that. I think there’s a lot of promise for tech like this, right?

I think, if there’s, again, one piece of advice to hr, look, you guys are leaders in your organization in terms of, how human capital is managed. Human capital is basically employees like mobile apps when they first came out. And people were trying to figure out whether this is gonna have an impact on the workforce.

It absolutely did. And everybody uses mobile apps today, right? AI is gonna be used by every employee in the company, and it’s an opportunity for HR to determine. How the policies and governance of how they use those AI applications in a fair, productive way. And in order for HR to be good at that, they need to adopt and use these tools so that they’re speaking from experience, right?

And we could go on and on [00:51:00] forever about regulations and all those things, but I think that’s what it comes down to is, learn, embrace and, be an advocate. Okay. Okay. So I was gonna

Brian Fink: say there’s so many questions that I have written down on the, it’s like that I have written down on the pad.

I am glad that we are going to have you back on the program in the next 90 days because beyond finding out about generative AI and its effect on marketers or marketing Yeah I, this is. This is growing up real fast. I don’t mean to compare this to the internet or to the iPhone. The iPhone CH in the app store changed everything.

But this is revolutionary. This is a really good use case for how AI is gonna create a better

Rich Mendis: candidate, a better candidate experience,

Brian Fink: a better recruiting experience, and provide insights to our

Rich Mendis: organization.

Brian Fink: Rich, I really appreciate you being here. Thank you for making time for us today.

Ryan, read us out, buddy. What

Ryan Leary: do you got? Yeah, man, rich, there’s nothing to read out here, man. Mike Trout, walk off stage, you’ve done it. Thank you, man. Appreciate it.

Rich Mendis: It was my [00:52:00] pleasure. We could, I’m sure, talk for much longer and maybe next time, the regulations coming fast and furious. EU is about to pass the AI Act, so maybe next time, we can talk a little bit about about the regulation, both domestic and abroad as well.

Awesome sauce.