Praneeth Patlola: I guess we are live. Hello everyone. Thanks a lot for joining us in a session. Sorry for the glitch in starting a bit five minutes late. We're still waiting for one more of our panelists, but we could get off and then proceed with it and we'll make it as interactive as we can. Thanks again for joining our session.
Praneeth Patlola: My name is Praneeth Patola and the CEO and co-founder for Willhire. A bit more background. I love HR tech, having been doing that for quite some time. Building products and solving talent acquisition or I sometimes call it social talent, acquisition engineering problems. Many at times recently did at Willhire and before that at Jobhuk., I'm so happy to be here and talk about this, but I have an amazing session panelist here. I'll start with Andrew. Andrew Karpie is an independent industry analyst. He has a long standing background in our industry as an analyst, writing some amazing articles. I'm a big fan of everything Andrew writes.
Praneeth Patlola: That is one thing I never miss. Not reading. Even shortlisting and bookmarking and reading all two weekends. Thanks Andrew for joining us. And Andrew does have a degree from CMU, which is huge. I know this is one of those things people would want to stay, especially if you're in the tech industry.
Praneeth Patlola: Great. I also have Kevin Akeroyd today who is a CEO of Pro Unlimited joining in here in a bit.
Praneeth Patlola: Awesome, right on time and a bit about the background about Kevin, you know, longstanding enterprise SAS, data solutions CEO building out things from ground up. All the way taking to IPO listing them in NYC is a huge standing effort and Kevin right now leads the vision and everything else from data, product intelligence, execution for pro unlimited super excited to have you give it.
Praneeth Patlola: And I have a privilege to work with Kevin. So everytimeI get to learn something new within the operations of technology and how HR tech is applied to customers to solve some real problems. So again, thanks a lot, everyone for joining us here. Today's topic for us is the "Future of Contingent Workforce with Platform, Data, and Intelligence".
Praneeth Patlola: This topic is super exciting to me as an engineer and pretty sure it is exciting to all of you and every officer on the panel. And the reason is as an industry, we have been doing so many things. And I think it's kind of, as Andrew indicates, it's a quantum leap that our industry is going through and many of early adopters and also customers are adopting some humongous changes into our industry that respect to all these three elements, which will rewrite many parts of our industry and how we move forward.
Praneeth Patlola: Super excited going through this Q/A session and thanks a lot Kevin and Andrew starting off.
Praneeth Patlola: While these three big topics are so humongous that we can actually talk about them all day long. Let's start with the basics I would say as data. Data as an engineer. It's too close to me because I do that and play with it all day long.
Praneeth Patlola: But I want to flip this over to the panel and start asking a lot, let's start defining the data for our industry. Like what does this data definition mean? Maybe you know, Andrew, take it out and.
Andrew Karpie: Yeah, sure. I mean, the first thing I want to mention though, is that I think that over the past five years, data and contingent workforce management has changed dramatically.
Andrew Karpie: You know, you just go back a short time and data was basically what you got out of the application you use, which is mainly the VMS and all of that has been changing. And and that, that. You know, what is particularly defining of that is the data sources and the multitude of data sources.
Andrew Karpie: And so once again you know, some years ago, just a short time ago, you got your data from your VMS and now data is coming from a number of different sources. It can be the, of course it's the VMS, but it's also a whole set of other applications that are used to manage contingent workforce today, which includes direct sourcing, sow services, procurement and so on and so forth.
Andrew Karpie: So there's a whole suite as it were of different applications that are collecting and producing data and operationally. Then you know, the next set of data is our next set. The next source of data is which we didn't have too much. You know, going back several years is a partner ecosystem of data and intelligence partners.
Andrew Karpie: And so I know with Pro Unlimited, for example, Pro Unlimited partners with Eightfold and with Glider, but other solutions that once again are also generating data or even intelligence, AI based intelligence into the process, the fourth area is basically external market data.
Andrew Karpie: It may be about rates. It may be about job employment and so forth, but there is now Rapidly increasing use of this type of data to feed into other analyses. And then lastly you know, I think global data and being able to command global data for some large enterprises is very important.
Andrew Karpie: It's really a whole different game in terms of different sources of data. And it all adds up to a massive amount of data that has to be dealt with and managed and from which value needs to be extracted.
Praneeth Patlola: Awesome. Awesome. That's a good start, Andrew. And in terms of data in defining, and especially from the source of start, and I want to extend this to Kevin too, from Kevin, from your point of view.
Praneeth Patlola: From being more of the data grew and building these companies and operating, and now in CWS best kind of redefining for us what that is. How do you see data and how do you see this definition may be more applicable towards our Industry?
Kevin Akeroyd: Praneeth sorry. I think I finally got my technical difficulties worked out. I apologize. I can go in and out. Sorry about that. I finally logged in. Hello, everybody. Sorry for being. Go ahead and just recap the question for me again.
Praneeth Patlola: So we were just trying to, starting to define the data there, and you were just giving a high level insight of what data means so that we can actually proceed the conversation around that.
Praneeth Patlola: But you know, from your point of view, Being a technology entrepreneur and building this companies and seeing multiple sets of this in large global enterprises applicable with respect to contingent and how do you see our defined data in your point of view?
Kevin Akeroyd: Yeah. Great and great question. And I think a couple responses to that are.
Kevin Akeroyd: One. It is, you know, kind of foundational and it is lifeblood. Andrew did a good job of alluding to that a little bit. We should consider it. The blood and the oxygen, right. That makes the rest of the body and the brain work and it makes everything highly functional. We need to actually do a great job of defining.
Kevin Akeroyd: And we're going to talk about this later in the panel. I know what data we do want to actually capture right. Because there's lots and lots of one of the great things about data is one of the things I love the most about it, there's all kinds of noise, but let's get out of the noise and get into the signal.
Kevin Akeroyd: Right. So what data do we want to capture for what reasons? Right? For what applications, what use cases, whether there's a risk mitigation, hiring intelligence, right? Benchmarking pay barriers. There's, let's make sure that we are aligning whatever the business outcome is that right. With what we need to actually acquire. And we are actually harvesting the right stuff is number Two.
Kevin Akeroyd: Number three, let's put some definitions around it. Right? Because data could be structured unstructured. It could be right, it could be quantitative. Right. Et cetera. And then, and I know we're going to talk about this, you know, let's make sure that we put the right governance and right controls and compliance and you know, all that stuff so that it can be this blood and oxygen and highly functioning. Business outcome driver intelligence, driver, decision driver prediction and analytics driver without running afoul. Right. Of any of the other things that we're actually trying to do right inside the organization from compliance or from a risk or from a privacy or from a GDPR from another.
Kevin Akeroyd: So even that does it a great disservice Praneeth, but because it's such a broad topic, but in my mind, those are some of the ways we think about that is. The foundational nature of the definition, what we do want to capture and what we, don't, how we want to use it, how we want to keep it safe and compliance, and then we can go put it to work, driving all these fabulous business outcomes that we're after, as the professionals in the field of where we are.
Praneeth Patlola: Awesome. Awesome. I never miss out on any meeting where Kevin does not have an analogy to refer to, which I can always remember all my life.
Kevin Akeroyd: I promise, Andrew. I promised Praneeth that I would not use the one that data age is like fish, not like wine.
Praneeth Patlola: Sure, thanks a lot for these amazing definitions. And you know, I want to close out that particular definition though from defining data from a new data sources point of view, like there is so much net new things, which is happening in our industry for an example today. We have data, which is coming out of interactions with a candidate where there is insight and interactions with the candidate that is a net new set of data from, you know, email insights or communication insights.
Praneeth Patlola: So SMS insights, that's new, how do you really communicate with talent and how do you repurpose that even right. And the second set of data is. Topic is again DE&I. Almost every organization is collecting this net new set of data. Maybe this was not in practice for several organizations, like maybe five to seven years ago, but now it's becoming a prime object.
Praneeth Patlola: You have, how do you collect what new set of data we are collecting and bringing in into this. And my favorite topic being direct sourcing and I live on the data side within that almost every day. I think without data. Direct sourcing as a net new side of rollout is building net new data for an organization, which they never even combined and even sought us, even for contingent talent applicable data.
Praneeth Patlola: How do you make decisions faster or even what kind of data is available which does not even exist. Right. You know, the candidate information, data never existed. The premium talent pools or redeployment many times organizations now have to go back and say, oh, I forgot this for 20 years. Not having this data in the right structure.
Praneeth Patlola: Right. Especially if I'm redeployment or human need. And there are so many systems, which is why the topic is. Communication of data platforms and intelligence. So with that I really wanted to go a little bit deeper into this and kind of start at a higher level. What are we seeing this data impact our industry?
Praneeth Patlola: And where is this applicable? Where is the highest impact we are seeing from industry from stake, all stakeholders from enterprise.
Kevin Akeroyd: Andrew are you gonna start?
Andrew Karpie: Yeah, I can start on that. You know, I think maybe the first first level of value is being able to now have a comprehensive view of the contingent or external workforce.
Andrew Karpie: Previously hasn't been possible. But now that all of this data is being collected from different sources and processed differently, not just as a part of the sourcing process, which was, you know, typically what was important at that time. But now that data is being held and managed in a different way so that organizations can actually have a very vivid.
Andrew Karpie: View and set of insights, AKA talent management of their contingent or non-employee workforce. So I think that's a very big and first step.
Praneeth Patlola: Awesome. Awesome. Kevin, I'll flip the same question to you back again, a bit more deeper in terms of like, how do you see data more applicable and user stakeholders?
Kevin Akeroyd: Yeah I think I think that one that. You talked about new sources of data and, you know, being a consumer data guy, like I was for 20 years where we had to deal with all of that stuff. Right. And Kevin Ackroyd had 17 different digital identities. He had social handles and he had email handles and he had browser cookies and he had the right hashtags.
Kevin Akeroyd: Kind of the same thing as now applying to worker right or employee, right. The employee data and as new and new sources come in, kind of that notion of a golden record or our system of record to say, Hey, that really is pretty nice. Right. And maybe I'm capturing new sets of data right. From 17 different new places.
Kevin Akeroyd: Right. And some of them may be different. The information I didn't capture before. Some of it might be skills that I didn't capture before. Some of them might be accreditation. Some of it might be right, whatever it is, I'm going to continue to capture more and more about him. How do I make sure that it all goes back to the golden record or the definitive Praneeth, right?
Kevin Akeroyd: That is kind of foundational. And then, you know, how do I put that to work? Cause if the dataset continues to get bigger and bigger, how do I apply that for. Say a staff log versus sow decision, or how do I deploy that for a direct sourcing decision? Or how do I look at that from, you know, skills versus cost or a location or whatever.
Kevin Akeroyd: So I think the notion of harnessing it, controlling it in tying it together and a system of record, so that the data that is there to enrich any individual use case or any individual decision or any additional benchmark or trend or, you know, anything like that. Really key. Right.
Kevin Akeroyd: Obviously you've gotta be careful with some of them because just like, and again, right. If I used personal information that I knew to edit an online ad, I could be in jail again, we're not quite there with employee data, but there are certain things that it's already frowned upon. And pretty soon it might be legislated where, Hey, it's okay to capture this kind of diversity information, but you can use it in these three ways.
Kevin Akeroyd: You absolutely do not use it right. In these four ways. So we're going to need to be really careful around right. Gather everything, use it, selectively the right. And in some cases, legally and you know, and then finally Praneeth and Andrew, you're not talking about this a lot. I would just say one of the things is kind of like derivative data or metadata.
Kevin Akeroyd: It's like, okay, what does the data itself tell us, like hiring complexity or hiring difficulty, right. For that skill set in that area or. Right. Who am I going to compete with? Or gosh, if I optimize for this kind of diversity, what does that do to my time to fill or my redeployment rate or my, you know, et cetera.
Kevin Akeroyd: So what are the downstream things that capture data, and then as you said Praneeth capturing more, let's look at some of those derivative data sets or what it can tell us rather than just the data set itself. So again, that's a mountain. But boy, there's a lot there. And if people are going to really harness these data assets and start putting them into use, I think those are a couple of things that all the audience should really start thinking about.
Kevin Akeroyd: Andrew. Please go ahead.
Andrew Karpie: Yeah, no, just to piggyback on some of what you said I think one of the areas of data and data management that is going to be very crucial and really has kind of exploded. In recent years, is that around roles and skills and being able to construct taxonomies and libraries of roles in skills.
Andrew Karpie: Once again, going back a fairly short period of time it was basically flat profiles and resumes and relatively stale data. And that was the way in which candidates and talent had to be viewed. And now it's possible to use these others. Categories to get a much more granular and vivid and valid view of the talent itself.
Kevin Akeroyd: Yep. And I, again not to sound too much like a geek, but as I'm going to start capturing 10 more things about Praneeth and then 15 more and then 20 more having that golden, that, that system of record. Right. And then the right taxonomy or the right ontology that applies to this around him as.
Kevin Akeroyd: Right. As a candidate for this type of thing or this as a candidate to go convert to her or this as somebody that I want to go redeploy, or this as a person I want to put here versus here, it's just going to be so important that we capture all that derivative stuff. And even probably our all right.
Kevin Akeroyd: Diversity inclusion could mean the military. It could mean ethnicity. It could mean gender. It could mean there's a lot of flavors of diversity, right. Which one's tied to my key? Business schools, whether it's to perm, whether it's redeployment, whether it is time to fill, whether it is expense, whether it is right.
Kevin Akeroyd: Let's make sure that we are doing the right data capture after the fact to make sure that, Hey, I didn't just optimize. Now, I've got this shiny new thing called I've got 10 fields of diversity data. Let's go put them to use to hit a diversity target. Right. Well, that's great. We should do that, but let's make sure that we are seeing how those decisions one, you know, manifests itself in the diverse talent pool themselves.
Kevin Akeroyd: Right. Do they want to come to work for me again? Do they feel more inclusive, right, etc. And as importantly, for us as the hiring enterprises, what did it do to all those other key metrics versus just treating it in a vacuum all by itself?
Andrew Karpie: Hey, Praneeth. I noticed that we have some questions coming in. If you want to have maybe fold some of them in here, keep it interactive.
Praneeth Patlola: I thought of doing that too. So I think this actually ties to also one of my comments I was just going to make, but I'll read it through. So one of the questions from Chandini here asking for it looks like the DEI subject is very quantitative, but also is subjective in nature. But all she is agreeing with.
Praneeth Patlola: She's asking about how do we measure the level of inclusion or sense of belonging of an employee? Is there any system or technical tools to measure it? Anyone of you wants to take a stab at it, please feel free to go for it.
Andrew Karpie: Maybe Kevin,
Kevin Akeroyd: That's actually very kind and keen to what I was just saying.
Kevin Akeroyd: And that is, let's de-subjectify it a little bit, right by saying, okay. If I just want to say, okay, I got more diverse, well, great of a thousand candidates. I had diversity data on a hundred of them. Now I've gotten to diversity data on 300, you know? Yay. Congratulations. No, that's not really it that talk about subjective tend to your point.
Kevin Akeroyd: Well, big deal. I capture data for Y, right. But if we can actually say, okay, what am I capturing? And diversity is a good example. It's certainly not the only. What am I capturing? Why? Right. With what outcomes? And then I am measuring the outcomes, right. Again, whether those are costs or location or speed, or time to fill the redeployment or convert to perm or quality, you know, whatever it might measure the downstream impact for me is the enterprise.
Kevin Akeroyd: And I personally think, right. There's all kinds of simple data capture members in my actual, if I brought on 300 more diverse candidates out of a thousand. You know, last year I brought on a hundred out of a thousand. This year I'm bringing on 300, let's go capture some data from those incremental 200 people that are diverse.
Kevin Akeroyd: And did I get at the inclusion goals or the cultural goals or the, you know, whatever goals. So I think. More and more captures aimed at specific goals. And then tied back to right that Praneeth right. So that I'm not only connecting the diversity inclusion data around Praneeth, but I'm seeing how that did around his performance or his conversion or his deployment or his quality.
Kevin Akeroyd: And how does he feel about working for me as a contingent employer? Now we've actually started to capture derivative data. It's back to that master record called Praneeth. And now we have intelligence, right? Rather than just a set of data fields that were patting ourselves on the back, you know, for collecting.
Kevin Akeroyd: So I hope that didn't sound too esoteric to anyone. You asked the question, but to me, that is your question is exactly how we need to start thinking about it.
Andrew Karpie: Yeah. Yeah, I would agree. And I think you know, as you're really implying DEI and contingent workforce is a relatively new topic in terms of workers and and it's really, you know, it's kind fortunate that it is happening now at a time when all these new methods for organizing and making use of data are being created and shaped and put into applications.
Kevin Akeroyd: And I think, I mean, this applies to the same conversation, applies to direct sourcing. If I'm going to shift 25% of my workforce, Out of staffing and our sow and into direct same exact thing. What do I capture aimed at what is right? Cause it's more than just saving percentage. Same thing about converting sow, staff log.
Kevin Akeroyd: Same thing about going from domestic to right. Remote global. I mean, name the name, the part of what we do name the desired business outcome, right? A little civically, we just went around with the DE&I. It applies to all of them. So this should be our new methodology and our new framework with how to treat any kind of data right around any kind of business outcome and how we should think about actually constructing.
Kevin Akeroyd: It's super applicable to DE&I. I think we should recognize though that it's a good way to approach pretty much all the data sets around all the business outcomes you want to apply to them, not just the DEI.
Praneeth Patlola: Yeah. And just to extend that a little thought and to close that out as from a data point of view, I think as an industry we have groomed yourself always from a supply diversity and maybe have collected information to do only supply diversity data.
Praneeth Patlola: But the topic is not completely changed with direct sourcing. Being the center of that conversation is like, how do I build a diverse talent pool and that building of a dialect talent pool. Kind of a platform, but also a talent acquisition sourcing mechanism where you're now designing. Do you have a pool which you can actually measure assets to?
Praneeth Patlola: Is it diverse or not? And can you apply that from a sourcing point of view, audio channels from where you're doing this are diverse and from a tools point of view, there's so many platforms around us, but it has to be driven towards. Not just a single platform because it starts from sourcing starts from engagement, content engagement, which is personalized, but also post-hired or even in a full-time space, which can be extended where there are several employee resources, group pools and several aspects around that.
Praneeth Patlola: And and I think from a data point of view, the highest applicability for us is if you don't have redeployable. How do you even bring in that aluminate talent to redeploy and make sense of that redeployable talent pulses under the applicability of the way I look at it. It looks like Andrew has a bit of trouble with this network and is probably going to join back in, but I want to kick this off to the next question too, and then flip this over to our next topic, which actually I was there.
Praneeth Patlola: So Jason is asking. Looking at blockchain solutions to capture data and use as a data management solution. Now I promise myself, I won't talk about blockchain because I talk about all day long, which is getting me too excited about it about de-centralizing our whole industry, but let's not, that's an all day topic, but I think it's a blockchain is still in early stages as an applicable.
Praneeth Patlola: And my binder, if you still see that as identity management, as one of those things, which is solving for the long term, but that's so many aspects of applicability as a concept itself, I think it's like super early stage. It's in its infancy stage, probably not even in it's not even fully incubating.
Praneeth Patlola: Properly to see the real applicability of that in industry, wouldn't be surprising at least in the next 12 to 24 months, the real applicability is still coming through. There's a lot of marketing terminology. I would not go by that as an engineer. If we are solving a very specific problem from identity management.
Praneeth Patlola: Yes. The cough keeping or any other aspects around that from authenticity of a data for reconsultation around the data. I think we have, that is like, almost everything we do can be done on that. I'll take Kevin's.
Kevin Akeroyd: Yesh, Praneeth I'll pile on there. I think that's right. Will blockchain solutions capturing data become a big part of this as we advance through this decade.
Kevin Akeroyd: Absolutely. Blockchain data gathering will be like brushing our teeth in the morning by the end of the decade or the beginning of the decade. And I completely agree with what Praneeth said in it. It's kind of typical. You know, five years ago, machine-based learning and artificial intelligence were kind of hype cycle fairy dust, and they sounded great, but there weren't a lot of applicability use cases now.
Kevin Akeroyd: You know, there really are real machine based learning and real but I'll go back to what we said. So many people want to hop to the AI or hop to the blockchain. We're like, well, no, there's already 22 systems of current legacy data that you're not even capturing. You're not even keeping consistent.
Kevin Akeroyd: You're not even having hygiene. You're not even putting it into the factory. You're not, you know, there's so many more things and we kind of wrapped up. You know, we're at 1 0 1 level college here in the industry around this data management. Let's resist the urge to go hop to our second PhD called artificial intelligence or a third PhD called blockchain.
Kevin Akeroyd: Let's get through 1 0 1, 2 0 1, 3 0 1, 4 0 1, our masters. Right. And then let's get to the sexy PhD stuff like artificial intelligence and blockchain. And that's not to be too sarcastic at all. A lot of people want to write this. When in fact, let's go ahead and take the foundational steps because one there's so much value creation just with the existing legacy systems, existing datasets exist in use cases, let's go unlock that value.
Kevin Akeroyd: Let's get a lot more intelligent and a lot better at the data harvesting and management. And then we'll be ready to plug it into a machine based learning AI to start capturing a blockchain, et cetera. And for all those dynamo PhD experts that are in the audience. I'm not talking to you, but I am talking to about 95% of the industry where the comments I just made do apply.
Kevin Akeroyd: And we just need to think practically before we get too excited about the next best thing.
Praneeth Patlola: Absolutely. And in my personal experience of hands on doing the data, it was surprising to me when I walked into a 25 year old organization previously. And look, the first question I asked is how many resumes in context of this and where are they store?
Praneeth Patlola: How much data tracking do we have? Do we know every segment of information and nothing was even available, like for the 25 year old organization, 35,000 resumes was something like you don't even exist in the data world. And I think that is super, super important. Thanks for bringing it up.
Praneeth Patlola: And I also saw that you touched on AI, but I want to take this topic a little bit further also because it's. It's although in a few terms AI, I, in my world, I start defining that as machine learning and you know, parsing technology, language processing. How do you PA is the technology available today to parse and fill fields from a resume from a job?
Praneeth Patlola: Can it match candidates? There's so much of applicability around that, but I think to start to define AI when data is abundance available in, around in multiple fashions and faces and forms from schema point of view. But when you start building things on top of it, and that's where kind of the AI learning process comes in basically, building that on building blocks of machine learning, training those datasets, and migrating that into pure inferences to different definitions.
Praneeth Patlola: Kind of applying that what is intelligence application? AI is, as Kevin indicated for me also, it's like a big word for us, but applying that intelligence from the data intelligence is much more applicable for us as such. They want to touch base and see those big questions as such. How do you both see AI as part of our industry or data intelligence as part of our industry?
Andrew Karpie: Yeah. Well, I think you know, I think first of all a lot of the hype is that AI is going to eliminate so much labor and make humans unnecessary. And I think that's very hyped up. I think overtime, we're going to see that. AI is making humans more capable and you know, whether it's managers or whether it's folks within a service provider that are supporting their customers, AI is going to help to make things better.
Andrew Karpie: It's going to improve as opposed to radically disrupt the industry in any way. And then there are a lot of ways in which that can happen, which we'll probably talk about here.
Andrew Karpie: Awesome.
Praneeth Patlola: Kevin, you want to add something to that
Kevin Akeroyd: Practical applications? If AI can help us sort through a thousand potential candidates and get to the right 11 that are the perfect fit for that hiring manager.
Kevin Akeroyd: That's not star wars, sexy stuff. That's just super applicable. What used to take me a week now can take me an hour and it doesn't replace the hiring, managing human. It just makes her way more effective and way more accurate than way more productive. You know, some simple examples like that. But you know, they're really the needle movers that we should think about because they are very practical applications that drive real results.
Kevin Akeroyd: And they make humans better. They don't replace them. And you know, that's what I think is just the way we should think about that because those are starting to pop up, you know, all over the place are just literally munching on Praneeth. Right? You do this all day, you know, take the 211,000 skills that we've got and the 110,000 titles that we've got.
Kevin Akeroyd: And if we're going to go help the entire industry shift from a title economy, To a skills-based economy, right? How do we use the machines to go through what would have been 10,000 human hours of computation to figure out the best bridging? Let's let the machine based learning AI, say, ah, those skills now should replace what those titles used to be for those kinds of jobs.
Kevin Akeroyd: Right. And again, really practical benefits right. It has no hype, no fluff. Hardcore tangible, right? Economic impact and productivity impact that are real meaningful. And again, make the people better. They don't replace the people. So I'm in favor of how we streamline this incredibly intensive process, right?
Kevin Akeroyd: You think from sourcing right from 10,000 nodes in the supply chain to matching right to them, predictive vetting to this, et cetera, just these simple things can make the humans involved in staffing or in hiring managers or incubating wildly more productive and you know, let's start there rather than.
Kevin Akeroyd: Come up with science fiction stuff that we might be doing 10 years from now.
Praneeth Patlola: Yeah, totally. Yeah. Totally. Go ahead.
Andrew Karpie: I think that as Kevin, as you said earlier AI is very good at finding the signal in the noise. Right. You know, there are many applications where that principle holds. One of them is just in terms of program management and operational management.
Andrew Karpie: AI can be useful in filtering out a lot of data and detecting patterns and identifying patterns and informing of patterns or certain events that require a manager's tension. I think the filtering and presentation of data in the form of dynamic dashboards and alerts and so forth is something that can really support management decisions not just about talent acquisition but about operations as well.
Kevin Akeroyd: Yeah. Yeah. You know, Praneeth, maybe just one other thing, practicality for the audience. You know, machine-based learning is impossible unless you have very high quality and very high volume of data. You can't put enough training data through the machine for it to learn anything. It really is that simple.
Kevin Akeroyd: If you don't have high volume and high quality of training data, you can't have high quality machine based learning for the machines to learn. Right. One-on-one and then at the machines aren't learning, then the AI on top of it can start picking. Signal or making a decision or coming to a conclusion. So it's actually really simple.
Kevin Akeroyd: It starts with the data capture we're talking about, right. Pointing it at something for the machines to learn and having high quality, then go tell the machines what you want them to learn. So a hundred humans don't have to write. And then say, okay, what inferences? Right, Andrew, what business school do I want to actually have to, as I had to ask, it's actually pretty simple.
Kevin Akeroyd: Despite all the hype, in spite of the buzz, if the audience can think about it, that way, it actually demystifies. You know, you know, quite a bit and again, good discipline. Well, what decision do I want this crazy AI that has good machine based learning and has good training data underneath it? What decision do I want to help me make right ? I buy off what it could do.
Kevin Akeroyd: And then okay, let's go do that. And the answer to that will be no 90% of the time. It will be 10. Yes, 10% of the time. And let's go chase those practical applications.
Praneeth Patlola: Awesome. Awesome. Yeah, go ahead, Andrew. Yeah. Awesome. Great, amazing. I think we could talk on this topic all day long, but given the sense of time, I want to close that AI topic with a few other things, again, applicable to again, direct sourcing and how I think about it is also the pattern matching.
Praneeth Patlola: When we talk about culture as one of the biggest buzzwords used in HR tech and HR industry in the last decade. And I would say the next new one is basically the DE&I one, which is a big, bigger adoption into the programs. Those two can be applied using AI. When you apply that AI from a sourcing point of view, our from a pattern recognition.
Praneeth Patlola: Particular candidate with a potential candidate fitting traits from our diversity traits or from the thought process. You can, sometimes you would want to put a hiding strategy. I want to hide similar people under the time you would want to build a strategy of, I don't want similar people. Non-similar people create a vector graph around that, to make those decisions.
Praneeth Patlola: And at the same time, there are so many other things around screening processes, whether it's a volume based recruiting or whether it's going to be even a very niche high-tech skill. There's so many things today, which are available to us in enabling and to enable users. We need a platform.
Praneeth Patlola: This is where I think platforms become such a powerful thing for us. And if platforms are not available to us in the right fashion and a format, I don't think there is anything to deliver to a user in the right fashion. I mean, using that to the right platform. So I want to flip this back to both of you and start, like, how do you see platform play in our industry, a role where do you see this is going about where it stands today and how do you see that?
Praneeth Patlola: Definitely different defined and applicable in contingent workforce.
Kevin Akeroyd: Yep. All right. Yeah. I'll cut me off in three minutes. Okay. Cause Andrew knows to pull string and pull the string back and I'll go on platform for, you know, three hours. So cut me off. But I think here's a couple ways that I think we should think about this.
Kevin Akeroyd: We use the whole industry's thing about this is, you know, what our platforms are basically right? When you tie the sum of the parts, right. That the whole is greater than some of the parts we as an industry have lived in. I use 17 different things. I use this for sow and this for screening and this for background and this for VMS and this analytics and this for data and this for sow and this recipe and this for payroll and this, I mean the whole industry lives in a world where we as companies spend years doing process people in systems integration of 15, 20, 25 disparate little pieces, right. That we have to stitch all together. And the inefficiency is just insane and that's just in the US and there's seven more in Europe and there's two more in APAC and there's three more Latin-America and there's six more in India. We as an industry, never move forward if we don't solve for that. So by bringing things together right, where it can be kind of a system of record that does all of that, it is all integrated. It is all together and facilitates the seamless end to end. Right. So I think about them as gosh.
Kevin Akeroyd: Do I really want to have four different SAS containers and six different service providers in seven different data providers and two more employer records. Or do I just want one system of record? That's an enabler, right? The way I haven't a CRM or the way I have an ERP or the way I have an HCM, you know, or any other area of the business, the answer is I want a platform.
Kevin Akeroyd: So platforms are holistic platforms, service, most of an end-to-end workflow, right. From sourcing all the way to offboarding right in every in-between I can get it done. And then they're interoperable with my other systems, like an HCM or ERP or a procurement to me, that's the right way to think about it.
Kevin Akeroyd: And then, you know, the vendor community. Is responsible for pulling together massive amounts of these tiny little point solutions that just do one little piece in one geo, putting them into that system of record or platform that can let me do almost everything across all my business units, across all the Globes, across 90% of the workflow and the functions in one place, rather than logging into 17 different UIs and vendor managing and integrating 17 different vendors.
Kevin Akeroyd: That's the very definition of the platform and the benefits are pretty inherent speed. It's quality, it's efficiency, it's cost, it's time, you know, it's all that good stuff. And it benefits the hiring manager to experience the enterprise and the candidates, because now there is that one system of record that everybody's connected to rather than 90, 90 versions of it living in and constantly trying to sync everything up.
Kevin Akeroyd: So that's what we think the big deal is around platform and why we've been pretty flag-waving about it's important in the industry.
Andrew Karpie: I think just to emphasize that all of what we've been talking about in terms of data and intelligence, you really can't achieve that without a platform for doing that is designed to organize all of this.
Andrew Karpie: Integrated for many different sources, process it in different ways. This is all heavy lifting by the way, this is not just something that happens. And you know, creating these, whether, you know, whether using artificial intelligence or other forms of analytics or just providing data and being able to feed that back into applications that people are using to make decisions and support.
Andrew Karpie: And yeah. Platforms are really an essential part of this. And there's a lot of talk about data platforms and data stacks and data ecosystems as well. And as a new world and also software is just not software anymore. It's a software that is driven partly with data and analytics that comes from sources other than the software application itself.
Andrew Karpie: You really need a platform to pull all this together when it comes to data.
Praneeth Patlola: Awesome. Awesome. And I think just to add a little bit to that, And then my mind just keeps evolving. I don't direct sourcing. I dunno why he does that, but when it comes to like all those matching engines, which are built on machine learning, which is built on a huge set of data accumulation, which is bringing in order to meet. If you look at the candidate experience, right? That candidate experience is delivered through this platform, which is why most of the direct sourcing solutions have a candidate portal side of it because the platform through that software element, but also engaging that particular candidate, not as just in one medium, we have multiple mediums, which is an ecosystem of communication of candidate, relationship management, through.
Praneeth Patlola: Email is primary SMS and, you know, multiple communication apps have become more extension. And also, you know, now able to see how a candidate responds to something. Use such important, intelligent insights to a recruiter or curator aspect to make faster decisions. Some new applications we have seen are like the probability for a candidate to change the job, right? You have 50 candidates. You're sorting. Like who, whom to, how do you prioritize that? So that's where the matching is done. But delivering that as part of like, again, the platform where you have to do multiple integrations, where from a sourcing integration, from a sourcing platform, from a curation platform, from assessments, from you know, reporting capabilities around that.
Praneeth Patlola: But one of the key insights, also one, one delivery on the platform side was about. The data where you have economic indicators from the market. And on the left side, you have the talent demographics in an area, your platform for an enterprise it's like, do I, can I generate this job here? Should I fill this role?
Praneeth Patlola: Here is becoming more like an immediate delivery process. When you go into a system of saying, I want this new role to be created or a need, should I go into 10 99? Should I go in? And , should I go into this, but also, is this role supposed to be created here or is it better if I create this in this particular region and geography for this particular rate, which is again, driven by that rate of intelligence, which is another model of it, right?
Praneeth Patlola: So much applicability around the data and you know, on the an, a closing note, I think we have like a book, like 10 minutes. What do you see as the top three trends in your own words, assets, you know, the list could be much longer, I can imagine, but the top three trends you're seeing are heavily applicable for the contingent industry.
Praneeth Patlola: Whether it's going to be enterprise staffing, providers, candidates are as an industry as also.
Andrew Karpie: Sure. Of course, you know, it's interesting because suddenly it's it, the top three trends is no longer an easy question to answer, because it just seems like there are so many different smaller trends.
Andrew Karpie: You know, so I've had to kind of pick and choose To, to come up with my top three lists. And they're really very close to home. The first is I think we're at the beginning of this data and intelligence journey and it's happening very quickly. But it's going to roll out over several years.
Andrew Karpie: And so I think we're going to see the continuation of this. I think most people, a lot of people in the industry. Are not aware of how quickly this is going to happen, but it's going to become more and more evident. The other you know, think a lot about direct sourcing or, you know, you wonder why you're obsessing about it.
Andrew Karpie: Well, it's not only because that's what you do, but it's also because it's actually happening out there. And I think more and more organizations are going to be embracing direct sourcing, which is something that, you know, it's not the same as what it was. Five years ago, it's actually becoming viable because of the data and analytics that ensure that talent acquisition can occur without hiring a hundred recruiters somewhere offshore or whatever.
Andrew Karpie: So I think that's the second one. And I think we're going to be seeing more and more smart socks. Solutions where more and more data and intelligence are being threaded into them at the application architecture and application level.
Kevin Akeroyd: So those are my three. Yup totally. And so to pile on Andrew you got three of my top three, so I'm going to have to go somewhere else, but I, you know, I agree blindly with what Andrew just said, and I'd be happy with those being the top three.
Kevin Akeroyd: If I'm going to add incrementality, I think I'm going to, I'm going to stress again. We're in a sellers market. Talent has power. That's right, Andrew, that talented white collar contractor. He's got nine companies willing to pay him top dollar. Right. And that'll be 19. By this time, next year, we are in a talent scarcity.
Kevin Akeroyd: The talent has the power. If we do not get focused on worker experience, worker engagement, relevance worker, feedback worker. If we don't do that, We're all screwed. Right? So that's, I think trend number one is it's a talent marketplace. It's not an enterprise. The enterprise doesn't have the power of talent has a power.
Kevin Akeroyd: Let's all just get ourselves around our heads around that because it's gonna be the case to, to is. And again, we touched base on this a little bit that we are going to see this continue just, you know, the walls are going to break down right. The professional services firm used to be a silo.
Kevin Akeroyd: And the staffing agency was a silo and the enterprise was a silo, and the talent was a silo. And the background checker was a silo. And the, you know, all of that is just walls crashing down transparency, seamlessness, and that's why things like platforms that facilitate, you know, all of this, but that disintermediation going away, right?
Kevin Akeroyd: The opacity going away, it's all transparent. The seamless. Everybody is coming together in a much faster, more and more transparent way. And that means a lot in including the importance of platforms. And then third is you know, I do think that all these new things like remote global, like diversity inclusion, like right.
Kevin Akeroyd: Think of all the things that are things that we think about you know, that are kind of. As the last couple of years, I think those are the new table stakes. And we will be thinking about those as table stakes, you know, going forward and there'll be something else that's a new, you know, a couple of years from now, but those are the three things that I think we need to kind of pay attention to as well as the new normal you know, and we'll figure out what the sexy trends are for next year later.
Kevin Akeroyd: But those are no longer sexy trends. They're now the normal kind of. Yeah. Awesome.
Praneeth Patlola: Well, this is huge. I think, you know, data platforms play remote and several aspects of bringing all the with intelligence. I think this is the most fun session as an engineer.
Andrew Karpie: As an engineer.
Praneeth Patlola: Really enjoy that terminology because you don't get enough time to speak up.
Kevin Akeroyd: It's great to produce, can admit what he finds fun.
Praneeth Patlola: Other than solving customer problems, but really thanks a lot, Andrew and Kevin amazing insights. And I'm pretty sure. Audience has some more questions and they'll answer them as they come to us here. But we still have five minutes, which is so hard in this remote wall setting in front of PCs and video calls, managing calendars.
Praneeth Patlola: We won't get with that. I think we can close out and thank everyone for this. Opportunity for listening again, Kevin, Andrew really amazing as always talking to you and thanks a lot for educating our session audience here
Kevin Akeroyd: with your insights. Thanks for Praneeth. And thank you audience, everybody to tune in to listen.
Kevin Akeroyd: We've got a valuable thank you very much for your hour today. Awesome. Thank you.
Praneeth Patlola: Enjoy your four minutes back in your way. Awesome. Thank you.