Episode Transcript
[00:00:00] Okay, everybody, I'm going to give you a tutorial on the talent intelligence data market, and then I'm going to talk to you about Lightcast and the acquisition of rhetoric and what it means. So, as most of you know, there is a massive, massive market for data about people. And this market includes multiple categories of data. People profile data, who you are, where you worked, your career, your job title, your educational background, your salary. Some of that data is confidential. Some of that Data is on LinkedIn in a public forum. But companies like People, Data Labs, Rhetoric, Cognizant, Signalhire, others, collect that data for a living. And they sell that data to many, many companies. They sell it to talent intelligence vendors, they sell it to companies that clean the data up and sell it to sales and marketing departments. They sell it to Dun and Bradstreet. They sell it to many companies. And so business is to find the data through surveys, through scraping, through access to various databases, hopefully mostly legal, but who knows, including Israeli spy software apparently, and provide it to the people who enrich it. So the second tier of data companies are people who enrich data. They take the source data and they do things like infer skills, compare it to the job market data to understand the potential pay of an individual job or person. Get information on the person's job history, which is what findom does to find patterns of individuals. Correlate the people data to the company data so they know who worked for who, when, and therefore how companies have changed and how people's careers have changed. Or correlate the data to education and degree and educational achievement so they can sell the data to educational institutions to understand who's getting jobs, how well their education is being used, and how well the university or the credentialing organization is delivering education. Or correlate the data to economic trends in the economy, gdp, inflation, other economic data so that the government can understand how the labor market is impacted by spending or economic trends or tariffs or other things. So there's this chain of data providers from the source identifiers and collectors of data to the many enrichers and users of data who sell it to different people in different use cases. One of the biggest use cases of data is for marketing, which many of you may know. If you want to get a list of people who have bought such and such a product, or a list of people in such and such a demographic, or you want to enrich your CRM database to get the names and the titles synced up, you can pay for products from companies like ZoomInfo and they will, and many others by the way. And they will enrich your CRM database so it stays up to date. And I'm leaving out a lot of other applications because the use cases for, for data about people are probably some of which are things you don't want to know about.
[00:03:07] And so in the world we live in and in HR and business, there are essentially these four big markets. There's the data for workforce planning, location planning, skills planning, and long term planning on where you want to hire, source, locate or develop various parts of your company. There's the data for recruiting on where we want to source people to hire them directly, the various issues on locations, demographic groups, pay per group, hundreds of things about sourcing. There's the data used for talent intelligence, inferring skills, building skills models, building career paths. By the way, you know, if you're going to build career paths for your employees, it would be nice to know if those careers are trending up or trending down. So even you as an employer gain benefit from these multiple factors of data. And then there's salary planning, salary benchmarking, salary analysis, both for sourcing, recruiting and internal benchmarking, doing analysis of pay equity, things like that. And then if you're growing or expanding or doing an M and A and you want to consolidate jobs and you know, simplify work or restructure around AI, you want to use all that people data for that too. So it becomes a very, very rich multi application industry of many sources and types of data, enriched and managed and organized for many, many applications. Now, until the last 15 or 20 years, most companies in their own corporate businesses didn't buy all this stuff. They didn't need it. It was certainly used by governments and educational institutions and marketing departments, but it wasn't really used by corporations as much unless it was part of their marketing department. Since talent intelligence started, which was about, I don't know, 12, 15 years ago. Most companies want to have this data themselves. They want to license it, they want to get it integrated into their HR, other HCM platforms, their ATSs and others. And they want to match the data from the outside world against the data inside their company for a whole bunch of reasons, for benchmarking, for enrichment and for analysis. And in order to do that, they need a company or a partner who does all the messy integration and enrichment and taxonomy development for them. Because frankly, most IT departments have many, many other things to worry about. They're more worried about product data, customer data, financial data, regulatory data. They don't really want to do all this kind of stu. Nor do most HR departments. So enter Lightcast. Lightcast, who we've been working with now for many years, which is really the merger of MZ and Burning Glass, but those companies are long gone. Is a PE funded very, very well run, from my opinion company that has decided to really become the authoritative source of this business and labor market data. By the way, let me talk about the labor market data for a minute too. The labor market data includes hiring data scraped from all of the job postings all over the world. And what happens when you scrape job postings. You see job titles, you see job requirements, you see job skills. And now because of pay transparency laws, you see pay. So the labor market hiring data is becoming very, very valuable because every day you can go to Lightcast and see new data on the labor market and in a sense you can see the actual economy live and kicking, or not kicking from the Lightcast data. So, and that data is used to enrich and complement the profile data that's acquired by these profiling vendors. So the way Lightcast likes to call it supply and demand data, but that's a little more confusing, I think. I think of it as the economic data and then the people data. And by the way, there's a third source of data which Drop does and Talent Neuron does, and I'm going to talk about Talent Neuron in a minute. And that's company data. Company name, company facilities, what cities they do business in, what products they sell, what technologies they use, what's growing, what's shrinking in their business, what are their major initiatives. Because, you know, if you're trying to analyze the economy of a city, for example, or you're trying to sell to a large corporation, or you are a large corporation and you want to look at your competitors, you don't want to just know information about their people as a big list. You'd like to know where they are, what they're working on, what Drop would call a workload, which is in a sense, the type of work they're doing, not just the skills they have and therefore the experiences and the, maybe even the products and services that they're secretly working on. I mean, you could, you could look at talent intelligence data, for example, presumably about Apple, and you could probably guess what they were doing with that car, that electric car that we never heard about. And you can certainly do that in the biopharma industry and other industries. So that's the fourth source of data, which is the company data about companies, which, by the way, you can get from company press releases, you can get from company analyst reports and so forth. So if you look at all that together, this data industry has the capability, if the data is well managed, to be extremely valuable for strategic planning, operations, hiring, career development for educational institutions to plan their skills and what careers and courses they should teach for the government to study patterns of growth and shrinkage in the different parts of the economy and the labor market. And really for all of us for a whole variety of reasons. And we used a lot of this in Galileo because we have a very deep relationship with Lightcast and we have a large part of the Lightcast data set available to you in Galileo. Okay, so let me talk about Lightcast. Generally speaking, there are three or four big companies that sell talent intelligence data. One is Drop D R A U P. Drop was founded by the two founders of Talent Neuron. From a historic standpoint. You may not care about this, but let me tell you a little story. Talent Neuron was the first real demand labor market data system developed by the founders of Drop.
[00:09:05] And they built that product before the Internet, before the cloud.
[00:09:10] They found that the technology and the compute and horsepower needed to collect all that data from the labor market was so high they couldn't keep it up financially. They did get a bunch of customers and they sold that company to Corporate Executive Board which then got bought by Gartner. Gartner also bought a company called Wanted analytics which was doing the same kind of work, merged those companies together and branded them as Talent Neuron. Talent Neuron was later spun off to another company. I don't know if they're a PE company or investment company and is now a company selling this Data with about 700 customers. So they're a medium sized provider of this data. The two founders of Talent Neuron, Vijay and Vamsi, created Drop D R A U P. Drop is the next generation of Talent Neuron. Very sophisticated analysis of all the things I just talked about earlier in a parallel universe while that was going on, the EMSI EMSI was building a data set for educational institutions to study outcomes of education. So they were collecting all sorts of data on the job market and the patterns of mobility and burning glass, was collecting a whole bunch of data on the job market for various government clients and for skills analysis. Those companies were merged and became Lightcast. And Lightcast, who we work with a lot, is a very well run corporate focused data provider. They sell data and systems and solutions to education. They sell solutions to the government market, they solutions to vendors and providers and technology Providers, most of which use Lightcast to sort of fuel their backend systems and then they sell data to you guys as recruiters, workforce planning, company organizations and other parts of hr. They just acquired a company called Rhetoric. Rhetoric. By the way, there's all sorts of companies I'm not going to mention in this podcast that are doing this, but they're mostly smaller. Rhetoric was a startup that spun off from Wanted and tried to build a business oriented data collection business.
[00:11:21] I don't know when they started, maybe a dozen years ago or so, or maybe a little bit less.
[00:11:27] That was in the data collection business, the first part of the process. Their primary competitor is a company called People Data Labs PDL. LightCast acquired rhetoric. So Lightcast now has a business that collects personal profile data.
[00:11:48] They also buy data from People Data Labs, as do many of these other providers. And so Lightcast as a PE backed, pretty fast growing company, is expanding to really mature its People Profile data. By the way, you might ask yourself, where's LinkedIn? All this? LinkedIn is essentially doing the same thing in its own closed universe. LinkedIn doesn't do all this data enrichment. They just use their own data and they don't sell it. But it is available to these vendors if they scrape the public profiles of LinkedIn, because LinkedIn makes its public profiles available so you can search for people on the Internet through LinkedIn. So anyway, Lightcast acquires Rhetoric and so they are going to expand from the markets that they're in, which are government, education, corporate and vendors. Licensing to sales and marketing has nothing to do with us in hr, but that's great for them. What does this all mean to you? Well, I mean, there's a couple things that I think are important to consider here. First, when you decide to do Talent Intelligence, of course, that word has been abused and misused an awful lot. You're going to try to make decisions about the jobs and employees and skills and roles and cities and locations that are relevant to your company. It's not going to do you any good to compare your software engineers and your IT department or programming in COBOL to the software engineers in Microsoft that are programming in Visual Basic, doing completely different kinds of work, even though they might both have the title of Software Engineer. So what you want to do, and of course you all know this, those of you that have been doing Talent Intelligence, is you want to buy, license, acquire the data that's relevant to you, and then as you implement it in your HR systems, complement it and update it and enrich it so that it's relevant to what you do in hr. Who's ready for a promotion? Who's skilled for this? Who's capable of this? What is our skills gap in this area? If we want to move into this business, how ready are we for it? What's our location?
[00:13:57] Demographic distribution of skills? Where are we losing people? Why are we losing people? Why are we losing these kind of people? I mean, all of these business decisions that are really, really important to some degree rely on many, many parts of this data set. So the way this works today is you go out and you buy a piece of software from Workday or whoever and you want to turn it into your talent intelligence system through the use of a data set like Lightcast. Now vendors like Seekout or Eightfold or Bimaray and some others do this for you. They sell the talent intelligence layer with the data in it. So they've already integrated the data from all these sources. You don't have to do it yourself. And off the shelf you get access to their system and they've done the integration between the enriched data provided from say a LightCast or PDL into your system and then you pick it up and run with it. But what you usually find, and I don't mean this, you know, in a negative sense, but that these providers don't have all the data in their systems because. Because they won't load a bunch of data from an industry they don't do business in, nor will they spend a lot of time running AI against it. So if you're a regional rural hospital and have all sorts of strange job titles, I don't know what they may be, but unique to regional rural hospitals, I mean, you buy Eightfold, it may not know what the jobs are or the levels and skills of a regional rural hospital, it may, but it may not. And you know, if you're doing a bunch of supply chain work in Germany or you have technical skills in biopharmaceuticals or you're building electric cars, I mean, you know, you want to know that the decisions you're making on who to hire are relevant. I mean, just a good example, let's take the electric car example because I've talked to GM about this, your GM or whatever car company it is, you want to hire a bunch of engineers to help you build your electric car. The engineering managers are telling you exactly what skills they need. You're going to go into the job market and you're going to look for people that have those skills. But you can't just look for the people you're looking for, people who have job titles. And so you're going to be looking for job titles that may or may not infer those skills. But you're going to want to probably know what are they doing with these job titles because it's not going to be clear from their job title if they're really ready to work on the kind of stuff you're doing. So you going to want to know where they're working and what they worked on and when they worked there, who they worked for. Right.
[00:16:22] So you can see that any competitive hiring strategy or growth strategy is dependent on this. I mean, if you look at what's going on in AI right now, I mean just in the tectonic shifts of the large vendors, with, you know, Microsoft targeting the people from OpenAI, I mean, that's a small list of a couple of hundred PhDs. But you know, when you want to hire an AI implementation person or somebody who's really got an experience with AI and government regulatory compliance and banking or healthcare or whatever it may be, pharma, you know, you're going to need an enriched data set to find these people because job titles are becoming more and more meaningless. It used to be that a job title meant a lot about what you did and no matter where you worked, that job title was similar. That's no longer true. In fact, the job titles are sort of almost. They're not completely meaningless. They give you the general job family but they don't tell you what this person has done. So you're going to be using talent, intelligence, technology vendors and data more and more and more. Now this is going to get even better with AI because if you do a good job of amassing or collecting the data you need in your company and that some of that vendor's data sources, but it's also your technology stack, then you will be able to use AI to do all sorts of amazing things. And by the way, you can do this with Galileo today because the bulk of the lightcast data set, at least the corporate part, is in Galileo. You can go to the Galileo or any AI you have that has access to this data. And you can say things like, I'm trying to hire a chemical engineer to build polymer plastics in the Midwest and I'd like to know the three cities that have the most number of engineers that have these kinds of skills. And by the way, please list all the skills in polymer plastic manufacturing or polymer plastic research, whatever it is you're doing that I need to look for. What are the salaries they're making in those cities and who's doing the hiring. And you won't have to do query, query, query, query to try to find that stuff, which is actually just hard to do with query tools. You'll just ask it that question and the AI will go get you the answer. And you can, you know, do that more or less today in Galileo. And you're going to, that's going to be true all over the Internet as you get access to some more data. So this, this marketplace of strangely named companies that are all doing things that sound like they're doing the same thing under the covers is really kind of a big strategic issue in how plan your HR strategy, your HR tech strategy and your company's growth strategy. Now let me talk about Lightcast. I don't know the size of Lightcast in terms of revenue, but from what I know about Lightcast, they are probably the most authoritative source of this combined data set of information.
[00:19:09] They are a really well run company. They really focus on taxonomies and also organizing the data. By the way, it isn't just enough to collect it and sell it. If it's not organized well, you can't use it into the job families, the government job titles around the world organized by a taxonomy of skills that's rational and not wacky. You know, they do all of that extremely well. They're very well run company. And so the acquisition of rhetoric is an indication to me, and I know them pretty well, that they are going after the bigger part of this market. They're going to go into the marketing side and that you really can't ignore them. One more sort of distracting sideline on this, there are vendors like techwolf and a few others that believe that they don't want to license any data is that they're going to get it themselves. They're going to do their own scraping, their own analysis of the data inside of your company and they're going to use your emails and your meetings and your documents inside of your company and they're going to figure out what skills you have. I mean, that's somewhat of a good idea, but it is limited. I don't think it's going to be the total solution. I love the guys at Tech Wolf, they're very smart people and they've done a lot of great integration things. But it's pretty hard to keep up on this market without being very, very invested in the bigger changing data market in the outside world. Every single day, there's thousands and thousands of new jobs being posted, thousands and thousands of people changing jobs, changing roles, moving, going into new companies, changing their careers, companies reorganizing. And that data doesn't all come to, to fruition easily, but these companies, like Lightcast and all the others I mentioned, make it their lives to go get it and clean it up and make it available to you. So I'm not, I don't have anything against techwolf, but I think you have to consider the fact that without access to the outside world of data, your inside data isn't going to be that useful. I used to joke all the time that when Workday first introduced the Skills Cloud and were a little bit confused and didn't really understand what was going on in Talent Intelligence, I did a lot of workshops for different Workday customers. What I basically it was telling them, you guys, you got a nice piece of software, but it's empty, you don't have any data in it, so it's not going to be very useful until you fill it with data. And that's the whole world of AI, right? The, the algorithms of AI are important, but I think in some sense they're less important than the data that's trained it and the data that it has access to. So that's sort of the story of what's going on. The reason I wanted to do this podcast today was that the Rhetoric acquisition was just announced and I figured it would be a good chance to explain all this to those of you that are in the HR space. In particular, we're very good friends with most of the providers in this market, so if you want to call us up and get some explanation, we'll be happy to walk you through what's going on and help you with your Talent Intelligence strategy as well. Thanks a lot.
