The vast amount of hype, speculation, opinion and hyperbole around generative AI can make it difficult to know how to start integrating it into the talent acquisition workflow.
The long-term implications of AI for talent acquisition are profound, and it is essential for TA teams to get fully up to speed with current capabilities and use cases. So how can AI become talent acquisition’s co-pilot, and where is the best place to start?
My guest this week is Siadhal Magos, Co-Founder and CEO at Metaview. I’m getting tremendous insight into AI’s potential by talking to super smart vendors who are baking generative AI into their product sets. Metaview is a perfect example of such a vendor. In our conversation, Siadhal shares his experiences on how AI can save vast amounts of time and resources in the interview process.
In the interview, we discuss:
• Current market challenges
• How should employers react to the AI revolution?
• Turbocharging productivity
• The top-down drive for efficiency
• How will AI transform TA?
• Detection and pattern recognition
• Making sense of unstructured data in the hiring process
• How is the interview changing
• Building more transparency into the recruiting process
• The importance of specialist tools
• The future of AI
Listen to this podcast on Apple Podcasts.
Transcript:
Matt Alder (0s):
Support for this podcast comes from Metaview, the platform that uses AI to automatically write your interview notes for you. Powered by GPT4. Recruiters and hiring managers at companies like Robinhood, Brex, and Genentech describe Metaview as a game changer for their efficiency and ability to have high-quality conversations with candidates. They can focus on the conversation rather than on taking notes. Metaview interview summaries are purpose-built for recruiting, so they’re 10 times more accurate and relevant than generic transcription tools. And, they work seamlessly with your recruiting stack, video conferencing tools, and even mobile calls so there s no need to change your existing workflows.
Matt Alder (49s):
You can see the magic for yourself for free on your first five interviews. Head over to metaview.ai/recruitingfuture. That’s metaview.ai/recruitingfuture to get started.
Matt Alder (Intro) (1m 22s):
Hi there. This is Matt Alder. Welcome to Episode 525 of The Recruiting Future Podcast The vast amount of hype, speculation, opinion, and hyperbole that surrounds generative AI can make it difficult to know how to start integrating it into the talent acquisition workflow. The long-term implications of AI for talent acquisition are profound, and it is essential for TA teams to get fully up to speed with current capabilities and use cases. So how can AI become talent acquisition’s co-pilot, and where is the best place to start?
Matt Alder (Intro) (2m 3s):
My guest this week is Siadhal Magos, Co-Founder, and CEO at Metaview. I’m getting tremendous insight into AI’s potential by talking to super smart vendors who are baking generative AI into their product sets. Metaview is a perfect example of such a vendor. In our conversation, Siadhal shares his experiences on how AI can save vast amounts of time and resources in the interview process.
Matt Alder (2m 30s):
Hi Siadhal, and welcome to the podcast.
Siadhal Magos (2m 32s):
Hey Matt, thanks so much for having me.
Matt Alder (2m 34s):
Always a pleasure to talk to you and great to have you on the show. For people who may not have come across you and your company before, could you just introduce yourself and tell us what you do?
Siadhal Magos (2m 47s):
Sure. So I’m style, I’m one of the co-founders and the CEO at Metaview. Metaview is the AI co-pilot for interviewing and for your interview process. So what we do is we completely eradicate the need for your recruiters or your interviewers, your hiring managers, to take any notes at all during the interview process because our AI is specifically designed to take world class interview notes for you. So happy to talk a bunch more about what that means for our customers and for the market as a whole. But yeah, that’s re really what we are about is eradicating that admin so that recruiters can truly focus on being, you know, can focus on building relationships with candidates.
Siadhal Magos (3m 28s):
They can focus on collaborating and calibrating with their hiring managers and really being present and curious and engaging during their interviews.
Matt Alder (3m 37s):
Absolutely, gonna dive into all things AI and recruiting in the conversation. Just before we do. It probably helps just to kind of set the scene a little bit, and this is something that I’m asking everyone who’s coming on the show at the moment because everything is so disruptive in terms of what’s going on in the market and technology and all these sort of things. With the clients that you work with, what are you seeing as their sort of biggest challenges in the talent market at the moment?
Siadhal Magos (4m 5s):
Yeah, so from what we are seeing, I think there’s two factors that are most sort of front of mind for a of our customers and a lot of our clients. One is, you know, we all know that recruiting teams have been especially affected by some of the larger, shots the larger layoffs we see especially in the tech sector. And so there’s this general sort of understanding and need to sort of do more with less. You know these recruiters A need to still sort of sometimes hit re pretty aggressive headcount goals. And B also need to sort of demonstrate that they are embracing best practice focused on the quality of the organization, focused on the sort of the talent density and the, the culture, the values of the organization that they’re bringing people into.
Siadhal Magos (4m 49s):
So, I think that’s one is like do more with less. So basically, you know, really I guess to input put in more basic words. Work out how to do a better job even if you are not having a tons more resources to do it is sort of one thing that we hear a lot of. If I was gonna paraphrase. And then the second thing, which is more of a sort of a, I think both of these things are opportunities really. But this, the second one that’s more clearly an opportunity, even though it may come dressed up as a challenge initially is what should we do given this AI revolution that is occurring around us? Like how do we react to that? Surely, we’ve gotta do something if we don’t, our competitors will and they’ll beat beat us in the sort o the competition for hiring great people. And this is true in every part of the organization.
Siadhal Magos (5m 30s):
It’s not just recruiting, but it’s sort of the sense we get is almost top down. There’s an expectation now that you’re gonna, you we should be using these productivity turbocharges. Of course as like the leader of a business you’re not gonna know how that should impact every corner of the business, but the leader of that function should and they’re trying to find answers to, how do we react, and adapt, and adopt the latest and greatest given this this AI revolution.
Matt Alder (5m 55s):
I think that is such an interesting point and it kind of reflects a number of things that I’m seeing in conversations that I’ve had. This normally when sort of technology comes into a recruiting, it’s sort of being pushed into the organization and people are trying to get that senior level buy-in because of everything that’s happening around AI and its potential effect on efficiencies and productivity in the way that companies do business. Yeah, I can imagine that lots of sort of TA and HR leaders are being asked, you know, what they’re doing to kind of embrace these tools. And I suppose that kind of leads nicely onto the next thing I wanted to talk about really, which is thinking about this sort of very broadly lots of conversation about AI and recruiting at the moment and some sort of very specific use cases.
Matt Alder (6m 40s):
And we’ll talk, obviously, talk more about interviewing in a second. But more generally speaking, what’s AI gonna do to talent acquisition? How’s it gonna transform it?
Siadhal Magos (6m 50s):
Yeah, and I think there’s a slight of it that is actually quite
Siadhal Magos (7m 32s):
The adoption curve on that has actually been like pretty good. A little bit behind marketing in sales maybe, but still pretty good on the adoption curve. But I think what’s happening now is there’s just a whole new bunch of activities that maybe up until about three or four months ago we thought were not sort of due to be automated anytime soon that actually can be automated. And so there’s this realization that a lot of people are having that, “Oh, that thing that I thought was a very human task.” Actually when I think about it, that’s pretty robotic and pretty repetitive in its essence and actually could be automated now that I had more intelligent tools at my disposal. So, in our case, and again, don’t wanna sort of over over index on this. But in our case that means, you know, note taking during a call, people always thought well actually to pull out the meaning from a conversation that feels like a pretty human thing to do actually turns out it’s not.
Siadhal Magos (8m 20s):
And the really human thing to do is to have the conversation, ask the right questions, be curious, be and give great responses to the cap. That’s still a very much a human and that’s what a great recruiter does. But actually pulling out the sort of the semantics and the sort of understand the meaning of what the replies you’re receiving. Actually now machines can do that better. So automation of some of these tasks that people have historically thought maybe would be further down the line. We’re now realizing that’s happening now. Like we are starting to automate those things now. And you see the same thing with like generating great blog posts. You know, same thing can of course can apply to like generating job descriptions. None of us thought, you know, let’s say a year ago that that would be something that that AI would be doing for us. But it’s obviously already starting. Still with prompting and engineering of like the actual output and final say on what that output is.
Siadhal Magos (9m 4s):
But in terms of going from like blank canvas to having something that’s, you know, 80% of the way there, that’s completely changed. So, that’s one is automation. The second bucket of impact that’s already starting to occur I would sort of define as detection. So detecting things that are happening in your workflow, in your process, whatever it might be that otherwise would’ve been essentially invisible or impossible for you to pick up. So, again, if I think about it in Metaview’s world, which is all about interviewing the ability to identify when one interview is like being run very differently to how other people are running the interview. The only way you could detect that previously is If you as a human being went and sat in on every interview and then realize, oh that one seemed to be very different to the others.
Siadhal Magos (9m 45s):
Like just not realistic because you got a bunch of other things to do. But now you have these essentially intelligent agents that can understand enough of what is happening in conversations such so they can detect where there might be inconsistencies. That’s a really sort of interesting area too. And I think that a place that applies to many of the parts of many parts of recruiting, many parts of business where just the fact that we have AI now that can pull out patterns in what is historically or is basically very unstructured data is a big change. So that’s I guess the mental model to have in mind. There’s lots of unstructured data in the hiring process. AI can now sort of see the forest for the trees and understand where there might be patterns that we might have actually been unable to identify without weeks and weeks of analysis.
Siadhal Magos (10m 33s):
So, that’s sort of bucket number two. So, so far automation and then detection, those are things that are happening now. And then I think the one that will, like we’ve essentially gotta prepare ourselves for and start to think about how we want to leverage it is intervention. So, at what point does AI start to sort of make recommendations or suggestions or whatever it might be for next steps within let’s say a recruiting or a hiring process based on the information that it has Because it’s seen how you’ve used that data in the past. So, an example of that would be, if our systems start to understand that when a candidate includes this, this, and this in their application process has this on their resume. And during the screening call mentioned, touched on these talking points, you know, 90% of the time they end up going through to the onsite stage, will we have tools that will start to pick and say, “Hey, you know, we should put this person straight through to the onsite because we already have enough confidence that that’s not gonna be a complete waste of time.”
Siadhal Magos (11m 27s):
So small things like this where actually they can make these really sort of specific recommendations based on their ability to automate and detect things and based on our confidence in their ability to automate and detect things is where I think the pucks going, let’s say in the next. Maybe months, but let’s say a year or two.
Matt Alder (11m 48s):
No, absolutely. And I think you really underlying there just how fast and somewhat even unexpectedly things have changed. To drill down and talk about interviewing. Again, I think it’s probably worth taking a step back and considering that it was only three years ago or three-ish years ago, pre-pandemic that the whole concept of technology in video and interviews, it wasn’t a niche activity but it was certainly a minority activity. And interview intelligence wasn’t something that people thought could exist. And we kind of fast forward three years to where we are now and what your organization does. Yeah, it’s kind of a massive leap forward and we’re about to take another one.
Matt Alder (12m 31s):
How is the interview itself evolving? What have you seen in the sort of the years that you’ve been doing this?
Siadhal Magos (12m 38s):
So, I think interviews as a whole are funnily enough. I wouldn’t say changing a ton in sort of like core structure. I think people have always understood that whether in intuitively like explicitly or inherently they’ve always understood that there’s a ton of signal that you pick up through conversing with another human being that is really hard to sort of have like pre-programmed into your application process or like introduced a couple of tick boxes for people to tell us some information. Or it’s very there’s just a ton of, I guess intangible sort of, well what felt like intangible information you pick up by having a conversation with someone that no one ever really got comfortable in sort of white collar creative work.
Siadhal Magos (13m 18s):
No one ever, which is what I can really speak to. No one ever really got comfortable with the idea that you wouldn’t have numerous conversations with someone before you committed to working with them, you know, potentially for the next couple of years of your life. And that was true on the hiring side and the candidate side. The candidate wanted to speak to someone. So, from an economics perspective, it always made sense to have these pretty in-depth. And so, sometimes what felt like pretty inefficient conversations. So, I think the thing that’s changed on the interviewing side is that bit is staying the same. It’s sort of, there’s like a, I don’t know if you came up with it, but Jeff Bezos is sort of famous for saying that when something changes in the world, it’s really important to seek out. What are the things that are gonna stay the same? So, my view is that the thing that’s gonna stay the same is that people are still gonna wanna talk to people.
Siadhal Magos (13m 59s):
That they’re gonna end up, they they’re gonna be collaborating with, you know, intensely for the next couple of years. And so that hasn’t actually changed too much the actual conversation itself. The thing that’s changed and is changing is an organization’s understa — just because it’s human driven doesn’t mean it has to be completely devoid of tooling, and devoid of data, and devoid of infrastructure essentially. So, the thing that’s changing is the amounts that we are actually pulling out of those conversations such that we can be much higher velocity in how we process the sort of — in our interview process to get people through it’s more quickly and much more reliable in the decisions we make because we actually have access to the insight that we’ve gathered from each of those candidates. So yeah, I think it’s more the thing that’s changing is the quality and the speed of those processes of the interview process that is improving more so than actually the like a fundamental change in the technique of assessment via interview, if you know what I mean?
Matt Alder (15m 3s):
Let’s get into the specifics of this. So, in terms of how AI is driving this forward even quicker. What are you seeing? What are you developing? How’s it pan out?
Siadhal Magos (15m 15s):
Yeah, so we think about it in two ways. One is turbocharging personal productivity. So, how can you actually enable the people who are involved in recruiting actually predominantly. Recruiters be much more efficient when they interact with the interview process. So, what that means for us is really two things. One, no longer taking notes. So, Metaview can actually produce world class written interview notes. And I don’t just mean a transcript after the interview. I don’t just mean a summary of the meeting. I mean actually pulling out the pertinent information that relates to this candidate suitability for the role so that you don’t have to pass that in real time yourself during the conversation and note it down yourself.
Siadhal Magos (15m 58s):
Metaview does a better job than you so much so that you know, pretty much all of our users just stop taking notes manually, completely. And that sounds small, but actually the sort of the mental lift that affords you in order to actually be a human being that the candidate’s meeting and giving answers to their questions probing where you need to, reassessing how I got all the information I need out of this conversation when you’re not sort of splitting your focus between that and taking notes is huge. So, that personal productivity gain because you’re actually doing a better job In the interview is one. And the second part of that personal productivity gain is literally the time it takes to clean up and write up notes. Now, everyone knows the one of the key things that sort of, whether it’s a reflection of like the performance of a recruitment team or whether it’s just affects the sort of the emotional satisfaction of a recruiter on a team is the quality of the relationship with the hiring manager because that is, you know, who you’re collaborating with and serving in many ways.
Siadhal Magos (16m 54s):
When you can actually provide them with really high quality, concise, understandable notes about each candidate that you meet with. Whether that’s in your sort of written feedback or whether that’s in your weekly meeting with them. Again, the quality of your collaboration with the HM just goes up orders of magnitude. So, that productivity side of things too where you don’t have to spend a ton of time preparing your feedback, preparing your notes. You actually have it produced for you in a way that is basically 10 x better than you could it on yourself anyway. Is sort of, is that personal productivity bucket for us. The the second thing for us just to, sorry, finish off that is general understanding of what’s happening In the interview process.
Siadhal Magos (17m 35s):
So another sort of like tension point for recruiters with the teams that they hire for often is, you know, chasing up others for their feedback. And not really knowing what’s going on in these interviews. You know, I worked really hard to get this amazing candidate really pumped about this opportunity. I was excited about them after the recruiter, after I spoke to them as sort of part of that initial recruiter screen. I put them into the rest of the pipeline and they just get an, I get a bunch of nos with almost no clarity on why they’re getting a no. Like that does not have to happen anymore. That is not a way to run your interview process. You can understand what was discussed in these meetings, what was the most pertinent information that was gathered from these candidates. Does the information that was gathered match up against the rubric that you agreed with the hiring team?
Siadhal Magos (18m 17s):
And if so, do you need to change your screen to better reflect that? Or do you need to talk with the hiring team and say, “Hey, we’re assessing things we didn’t agree with here.” So, rather than just being this complete black box where you’re sort of just forced to, well go and source more people and let’s put them in the top of the funnel again and fingers crossed you can actually be much more advisory and much more consultative in how you are working with that parenting team. And that’s all enabled by the fact that you have reliably structured insights off the back of every single one of these conversations.
Matt Alder (18m 49s):
I think what’s really interesting is a lot of the conversation is about Generative AI and ChatGPT and Bard and all those kind of very general interfaces that people are using. But increasingly I’m seeing that the future is around very specific expert structured sort of use cases, which is, you know, very much what you are doing with interviewing. We’re seeing it in podcasting at the moment. There are various podcast AI tools that help you transcribe podcasts and make sense of what’s going on. And do you think it is that sort of specialist, you know, that specialist knowledge and approach that that’s really gonna kind of shape the future of how we use AI in this industry?
Siadhal Magos (19m 35s):
100%. And I think it’s a great example of how the sort of the human creativity and ingenuity in sort of harnessing AI but with particular personas and use cases in mind just results in a much better outcome. So, if we put that again through the lens of Metaview there are other tools out there that can transcribe and summarize meetings. But if you look at really basically. If you summarize an interview like a generically, if you just to summarize an interview then and during the recruiter screen, I as a recruiter spend half the time talking that means my notes are gonna be half. Half of it’s gonna be some me the summary of me talking. And actually as a recruiter I have no interest in that. I just wanna know what is the information I glean from the candidate, what did they react well to?
Siadhal Magos (20m 16s):
What their motivations concerns? What’s their experience? So even at the most basic level, the fact that you can build a tool that specifically is designed to A, output notes that are essentially exactly what you want out the back of that conversation. The notes are exactly what you want. Not a summary of the whole conversation, but exactly what you want as a recruiter to help you do your job and you take your next steps is only gonna happen if you take a very specific view. And then the second thing about it is, of course you want it to fit into your workflow, right? So, within recruiting that means, well I still have an ATS where, you know, a lot of this feedback or this feedback is gonna have to end up. So, how can I really make sure that the interaction between the notes I get and what gets submitted into the ATS is as seamless as possible.
Siadhal Magos (20m 56s):
I also know that different people have different, sort of are stakeholders in this role. So, if I’m working with a hiring manager, I don’t wanna maybe always have to proactively ask them, “Hey, can you share that transcript from that meeting you had with the candidate? Because I’d like to see what happened too.” No, it’s just baked into the system. It’s part of the workflow that of course the recruiter can access the downstream conversations that their candidates had. Those are just two like very high level examples of what is a massively long tail of use case and domain specific sympathy. That means I think we will end up with as you said, these domain specific tools rather than these very generic ones like you know, a world where everyone’s using chat to sort of solve their recruiting queries.
Siadhal Magos (21m 41s):
That’s not realistic.
Matt Alder (21m 45s):
So, final question for you, and I suppose this is really a summary of what we’ve been talking about. Because we’ve been talking about what’s gonna happen in the future throughout the whole conversation. But how does this all come together? What do you think the future of talent acquisition is gonna look like a few years down the line?
Siadhal Magos (22m 1s):
Yeah, so as with many other professions, recruiters are going to have a co-pilot. And there’s many different ways that people have talked about this. People think about it saying, “Hey, just imagine you have like 10 interns at your disposal. You can tell them whatever you wanna do. They’ll go and do a great job. They never get tired. Or think about it as this assistant who’s very good at these specific set of tasks. And guess what, you don’t even have to ask them to do it. They just do it automatically.” There’s like various ways and different sort of blends that I guess people will come up with. But essentially that is the future that is happening now. You know, we are all starting to get our assistance for automation and sort of improvement of decision on our most high leverage tasks. So, I think that’s the future. In many ways it’s the present to be honest with you. Then I think the future becomes more about, well how do those, you know, the the sort of the AI assistance that we all have?
Siadhal Magos (22m 45s):
How do they interact with each other? And other is there like, how do they operate as a buffer between different parts of the organization? And I think one part that’s really interesting is how do they start to lubricate some of the handoff points between different — whether it’s people or different parts of the organization. So, what I mean by that is if for example, right now something that would be super helpful for every recruiter to do every single time is get a debrief on every interview, inform the next interviewer about the things that were flagged or missed during that interview so they can include that in their interview, right? It would be great if we did it every time. It’s not possible every time because I can’t always get the debrief.
Siadhal Magos (23m 26s):
If I do have time to get the debrief, I might actually get the insight I need because the person might be rushed anyway and they don’t gimme the key info. But actually in a world where you have this always present smart assistant that’s essentially part of all of these conversations in some way that becomes something that. That handoff just happens all the time. You don’t rely on the sort of slowdown of information or the sort of the human inspired slowdown of passing information between one team to the other. It’s just all automatic. So, I think those sort of like those current points in organization where there’s a little bit of friction where information does not actually go, the right information does not always get passed at the right time to the right people. That’s what will start to change as well because we all have this improved productivity and this improved ability to take on information and have it served to us at just the right time.
Siadhal Magos (24m 11s):
So, I think that’s a really exciting thing. Again, I think this is all near future though. I don’t think we’re not talking like five years out. We’re talking, you know, six months, 12 months out. How are we gonna be 30 to 100% more productive.
Matt Alder (24m 26s):
Siadhal, thank you very much for talking to me.
Siadhal Magos (24m 27s):
Thanks so much Matt. Really enjoyed it. Thanks for having me.
Matt Alder (24m 31s):
My thanks to Siadhal. And if you want to experience Metaview interview summaries firsthand, they’re offering five free interviews to Recruiting Future listeners. Just head to metaview.ai/recruitingfuture to get signed up. That’s metaview.ai/recruitingfuture. You can subscribe to this podcast, in Apple Podcasts, on Spotify or via your podcasting app of choice. Please also follow the show on Instagram. You can find us by searching for Recruiting Future. You can search all the past episodes at recruitingfuture.com. On that site, you can also subscribe to our monthly newsletter, Recruiting Future Feast, and get the inside track about everything that’s coming up on the show.
Matt Alder (25m 19s):
Thanks very much for listening. I’ll be back next time and I hope you’ll join me.
