DUE TO SOME HEADACHES IN THE PAST, PLEASE NOTE LEGAL CONDITIONS:
David Yakobovitch owns the copyright in and to all content in and transcripts of The HumAIn Podcast, with all rights reserved, as well as his right of publicity.
WHAT YOU’RE WELCOME TO DO: You are welcome to share the below transcript (up to 500 words but not more) in media articles (e.g., The New York Times, LA Times, The Guardian), on your personal website, in a non-commercial article or blog post (e.g., Medium), and/or on a personal social media account for non-commercial purposes, provided that you include attribution to “The HumAIn Podcast” and link back to the humainpodcast.com URL. For the sake of clarity, media outlets with advertising models are permitted to use excerpts from the transcript per the above.
WHAT IS NOT ALLOWED: No one is authorized to copy any portion of the podcast content or use David Yakobovitch’s name, image or likeness for any commercial purpose or use, including without limitation inclusion in any books, e-books, book summaries or synopses, or on a commercial website or social media site (e.g., Facebook, Twitter, Instagram, etc.) that offers or promotes your or another’s products or services. For the sake of clarity, media outlets are permitted to use photos of David Yakobovitch from the media room on humainpodcast.com or (obviously) license photos of David Yakobovitch from Getty Images, etc.
Welcome to our newest season of HumAIn Podcast in 2021. HumAIn is your first look at the startups and industry titans that are leading and disrupting ML, AI, data science, developer tools, and technical education. I am your host, David Yakobovitch, and this is HumAIn. If you like this episode, remember to subscribe and leave a review. Now onto our show.
Welcome listeners to the HumAIn Podcast where we’re gonna deep dive today into all things, data, and data insights with Ben Zweig, who is the CEO and founder of Revelio Labs. Today’s episode features Ben who not only is a startup founder, but he’s also an adjunct professor at NYU Stern in New York City.
I’ve had the great chance and pleasure to meet Ben over the past year learning more about his startup participating during his capital raise and hearing about all the great insights that he and his team share with their insights about the workforce and the future of work. Ben, thanks so much for joining us on the show.
Thanks, David sounds good.
I’m really excited to learn not only about what you’ve done, and how we participate together but let’s rewind for our audience. Can you share a little bit about your career? And what brought you to today?
So, I started my career in academia, I was doing a Ph.D. in economics and really specialized in labor economics. So I was always very interested in labor data, and understanding occupational dynamics, social mobility, things like that. My first job was a data scientist, this was very early on at a hedge fund in New York. It was an emerging market hedge fund. I started that in 2012. That was interesting. I was like the lone data scientist on the desk. So that was interesting. And then went to work at IBM, in their internal data science team was called the Chief Analytics Office.
And they started working on all sorts of projects related to workforce analytics, and workforce optimization. So that was really interesting. And, merged these two worlds of data science and labor economics, which I was really passionate about. And that’s where my co-founder, and I started getting the idea behind Revelio Labs, which is really to create a universal HR database and help companies improve their performance and investors understand companies more deeply. So that was the impetus behind what we’re doing now. So we started Revelio Labs in early 2019. And it’s been an exciting ride.
It’s so fascinating, this whole field of workforce analytics myself, one of the early companies I worked with was ADP. And I worked in the PDO division with payroll and looking and seeing how that impacted jobs in the creation, the movement of workforces.
So I’m really passionate about what you and your team are doing, as you’ve been scaling up the company the last few years. Were there some of the insights that you’ve uncovered that you think are really fascinating, that you enjoy seeing in the market, or you’ve discovered?
It’s an interesting time to be analyzing this, because everyone cares about labor dynamics these days, especially with COVID and great resignation. So this is such a big topic. But, what we’re finding interesting is actually fundamentally basic things that are not known about a company. So even just, how many employees do they have across occupations?
How many engineers does this company have? How many salespeople? How’s that changing? What’s their attrition rate of key roles? What’s their hiring rate of key roles? How much are companies paying? You can really get a very deep view of what’s happening within a company just by tracking their talent dynamics.
So, one analogy I really like, especially for investment management, when they’re looking to understand companies is like, let’s say you’re buying a car, and you take it to the mechanic to see if it’s healthy. Imagine if that mechanic could only look at the speedometer. It would be ridiculous. You want the mechanic to actually look under the hood, look at the engine. And that’s sort of what we think we are. So investment management today is mostly, people spend a lot of time looking at indicators of performance, financial metrics.
And what we want to enable is a much deeper view of you under the hood of companies. And that is people, the engine of a company, the engine of every company is the talent. So, that’s kind of what we want to look at, and we’re really just scratching the surface now, but there are so many kinds of interesting insights about, where people are coming from when they get hired, where people leave when they quit, how satisfied the employees are. It’s really endless. There are countless possibilities of how this data can be analyzed.
Now, Ben, you mentioned the big topic that many of our listeners have seen, not only on LinkedIn insights but also throughout the news, this great resignation. And this is something that certain economists did see coming, that when the pandemic was going to end that a lot of people were just really needing to take a break.
And that’s what’s led to this great resignation of a lot of talent, both technical and non-technical, departing en masse from their jobs. And this is across the generations. We’re seeing this from the Gen Z’s all the way through the boomers. And what have you uncovered around the great resignation? Or how do you think that will impact technical hiring and pipelines for companies?
So, it’s a really fascinating phenomenon. I don’t think we’ve ever seen something like this in our lifetimes. But it’s exciting that workers are empowered more than they have been. But what we’re really trying to uncover is the heterogeneity across different subgroups of employees. So, we know that attrition rates are higher, quits are higher, people are leaving the labor force, people are hopping on to new jobs, much more than they have been.
But that’s not true across all occupational groups. It’s not true across all geographies. So, we’re really trying to find out where those shortages are and where that wage pressure is highest. And it’s typically more among blue-collar workers, which we had a paper that came out jointly with Barclays equity research, where we developed some trading strategies based on that phenomenon. So, that was interesting.
But even just in my personal anecdotal experience, we’re seeing it for high skilled workers too. So, we’re trying to hire quite a lot. And we’ve, data scientists, software engineers, data engineers, economists, and it’s really tough. It’s hard to find good people. And salaries are getting higher and higher. So, there’s definitely pressure on the employer’s side.
And a lot of that pressure we’re seeing here on the tail end of 2021, though, I am excited for where we’re going to go in 2022. Personally, as well, what I’ve seen working for multiple startups is that technical hiring and non-technical hiring has been very hard to fill during much of the pandemic because everything went remote and distributed. But now with this return to the hybrid workforce, and for many in person that does present that opportunity that there are a lot of openings, for careers, and for job seekers. That could be promising, especially if you’re searching now for a new career path.
The workers that were really hardest hit from remote work are really junior employees. They’re just getting started and they need that mentorship. And it’s much harder to feel like you’re developing and learning from others in a remote environment. But as we’re sort of going back, the more senior positions will probably not have that same benefit as junior employees.
So going back to the office, and that’s really because so much of the day-to-day for more management positions are on Zoom. And if you’re on Zoom meetings all day at home, that’s one thing. But if you’re on Zoom meetings all day from the office, that’s the worst of both worlds. So, that’s probably not going away anytime soon. So, it makes me a little bit cautiously optimistic. I’m more optimistic for more junior employees and more pessimistic, on what that means for management roles.
Now talking about the future of work and the future of cities, I know we’re both greater than the New York City area, the greatest city in the world. And New York is up and coming again. We’ve seen in the last few labor reports that building occupancies are up to a little bit north of 28%, which sounds like not that much. But at the height of the pandemic, it was down to only about 10%. So, we’re getting back up there. We’re almost to a third occupancy. Are you seeing any insights for those trends going into 2022 into the 30s 40s are 60 percent?
So, New York has had a spectacular recovery. Earlier on there was a lot of movement from tech hubs to Texas, Florida. That was a big trend. That’s something that we’ve seen a lot of that has slowed recently. That phenomenon will probably slow down. But we’ve seen a lot of this return to New York City. I don’t know exactly. We don’t have any additional insight on that. I don’t know if we have a specific forecast but it seems like a positive trend.
And beyond these trends in the data that we’re seeing, you’re also now running a new course, at NYU Stern, called The Future of Work. I’m so fascinated by this because everyone’s been talking about the future of work for the past two decades. COVID happens, future works here. It’s here now. And now you have this new course. Can you share with us? What is the future of work in the classroom?
It’s really exciting. I’ve been trying to push this class for a few years but didn’t quite brand it properly. So, I’ve been teaching classes in data science and econometrics for a while and trying to transition to something a little bit more related to labor economics and human capital. So, I’ve been proposing classes around the economics of human capital and things like that.
And there wasn’t really a tremendous amount of interest. But then, I had this moment where I was like, oh, maybe it should be called The Future of Work. And then, the next day, it was just like, the Dean’s excited, everyone wants to make this course happen. So, it’s a little bit of a branding trick.
But I’m really excited about how it will go because there is so much interest in it. So the way we’re organizing is, basically the first chunk of it will be around the foundations of the labor market, what are occupations? How do we assess activities and skills, and think through the dynamics of employment, contract labor, and hiring all the other kinds of foundations of the labor market?
And then really, using those foundations to understand the hot topics of the day. So automation, remote work, the gig economy, getting into these more topical areas. And then in the last chunk, thinking through what are the applications for businesses, for workers, for governments, for entrepreneurs. So, thinking through the applications is also a really exciting part of it.
One area from the syllabus I’d like to highlight and perhaps you can tease for the audience, some of which is around the gig economy. And the gig economy has been continuing to grow in the last few years, of course, without so much as some controversy, as we’ve seen in California with some of the companies like Lyft and Uber, and DoorDash, you’ve had different opinions on classification of gig versus not gig or, a lot of tech companies do we go gig plus-plus everything go gig, everything go freelance has been a lot of opinions, a lot of knowledge, a lot of thoughts around this, what’s the take that you’re seeing, or whether some of the insights you’ve uncovered around the gig economy?
One phenomenon that we see quite a lot is that companies have a huge contingent workforce that is not reported on their financial statements. So, for example, I mentioned I used to run this workforce analytics team at IBM. And at IBM, we had 330,000 employees, that was like the number that’s in their HR database, but you go to their LinkedIn page, and it looks like 550,000 people say that they work at IBM. So, what’s going on here? Why are there so many more people that claim to work at a company, then the company claims to work there? And that, of course, is just a sample; only a sample of people actually have online profiles.
So, the real situation is that a huge fraction of their workforce is contingent labor, in some cases, up to two-thirds. So that is just a legal artifact, it doesn’t really matter when you’re assessing a company’s workforce, who’s classified as a W two employee, and who’s classified as a contingent worker. So, there’s real limitations to how companies categorize their own workforce. And that creates all sorts of incentive misalignment. Companies really only need to report their diversity metrics on their official employees. So, that can lead to some greenwashing. They only need to report geographic distributions based on their official employees.
So, that might lead them to understate the extent to which they’re offshoring. So, you really don’t get an accurate view of a company’s workforce from just looking at their official employees. And companies do have incentive to expand to contingent workers and also gig workers and freelancers, and that is just like the shadow workforce in a way because they are not required to disclose that information. So, that worries me, just from an accounting perspective. So, it’s just unknown and in some ways unknowable.
So, we’re trying to get back into that as much as possible. And what we’re seeing more of, but it’s a little hard to say whether this is a real trend is that companies will expand more geographically. It’s up in the air, whether this latest COVID crisis will cause more globalization or more nationalism. So, we’re hoping that this creates a more integrated labor market, across geographies. And the freelance economy is very global. So, we’re hoping that expands and it has been expanding, but it’s a little too soon to tell whether this will persist.
Because it depends where in the economy look at some of the early signals when we look at startups, companies that are raising precede seeds, Series A, no longer do, they need to be co-located in one city, like New York City or Boston. But you could have a distributed team, different cities in the United States, different cities in both the United States and Europe, and even Asia and South America and Africa Continental. So, that expansion is continuing.
Because talent is global and distributed, the team could work from anywhere, which is really fascinating. And we’ve seen that with some of our portfolio companies as well, both through data frame and data power, where that movement is happening. Have you particularly seen more of that shift happening quicker with startups being the innovation versus the big IBM’s of the world, which actually already are global, so to speak?
Absolutely. So, we didn’t really segment this by startup or established companies, but we’re certainly seeing this pressure in tech. So, within tech, there’s significant wage pressure, like wages are converging across, the more productive high wage cities and the lower wage cities. So, that gap has been shrinking quite a lot in the past year or so.
That’s so fascinating. And so we’ll continue to see where that occurs, especially as the world continues to open up and we move back into this hybrid modality that we’re at. Let’s change gears of the conversation to Revelio Labs. You’re the founder and CEO of Revelio Labs, focusing on workforce intelligence in New York City. And earlier this year, you had a raised round of capital. Can you tell us about your company being venture-backed and what you’re doing to continue to scale the workforce intelligence?
It’s been very exciting. And being venture-backed has really allowed us to grow very rapidly. So, I’ll put in a shameless plug, if you’re a talented data scientist, or data engineer, or economist, give us a call because we’re looking. Basically, the foundation of Revelio Labs is to create a ubiquitous source of workforce intelligence. We sometimes model ourselves a little bit after Bloomberg, and companies like that, who have created a ubiquitous source of financial intelligence. And, if anyone’s looking for financial data, they know they can find it from Bloomberg, Thomson Reuters, FactSet, S&P. And these are enormous companies.
But when it comes to human capital data, and employment data, that really does not exist, it’s not even really close to that. There’s so much data that’s siloed in internal HR databases, which like I mentioned before, really only include a fraction of the overall workforce of a company. But what’s cool about this is that when an employee is stored in an HR database, that information is mirrored in the public domain.
So, you and I might be employed by a company, and we’re in that company’s HR database. But we also report that information on our resume or LinkedIn profile or something like that. So, there’s a tremendous amount of information out there on profiles and individuals’ career paths and jobs and skills and how they describe their jobs. So, that’s just an incredibly rich database that we’re continuously collecting and updating.
There’s also job listings, which can be a great proxy for the demand for labor. You can also get salaries from job postings, which is really very rich and changes every day. So, you can see day-to-day what’s happening with wages. There’s also sentiment out there. So, sites that track employee reviews. There’s also layoff notices.
There’s an act in the US called the WARN Act since we’re worker advanced retraining notification. So, companies have to report layoff 60 days in advance. There’s immigration filings, there’s so much data out there. And what we’re doing, we’re bringing it all together into one place. And we’re constructing this universal HR database.
But, the challenges there are pretty enormous. On the one hand, it’s a huge set of data. There’s 10s of terabytes of data that come in every month. And also there’s just problems with the data in its rawest form. So, first of all, people use job titles and seniority levels that are completely idiosyncratic. So, different companies have different conventions for occupations and seniority levels.
So, we really have to create a taxonomy that updates that changes with an evolving occupational landscape and the changing economy. We also really need to infer the activities that people do, because those are the building blocks of a job, or the job is a bundle of activities. So, we really need to understand that when one person says lawyer and another person says, attorney, those are probably the same occupation, but when one person says Product Manager in Facebook versus a Product Manager at JPMorgan, those might be totally different occupations.
So, we really need to do all that natural language processing, to categorize this data properly. We’re also creating enrichments, and new features predicting salaries and models of prestige and suitability to remote work, things like that. But we’re also removing biases in the underlying data. So, there’s sampling bias, not everyone’s got an online profile, there’s also lags and when people report a transition, so we really like the data to be point in time, that’s really important for a lot of a lot of end-users.
So, we need to produce an outcast basically predicting what will be retroactively revealed, since a lot of updates don’t get reported immediately. So, there’s just a tremendous amount of challenges to make the data usable. And, that’s what we’re doing. Somebody’s gotta do it. So, we’re trying to tackle those hard problems, and it’s fun.
And from where you are in the team of Revelio Labs have been growing, you’re scaling in the data science and data engineering arena. So there’s a massive amount of these databases and datasets that you’re seeing each day? What’s Calm? Can you share with us a little about the product roadmap that your team is building out? Or you’re getting requests from your clients, they’re saying we want to see more of this?
So first of all, in the alternative data market, which we’ve been playing in, there’s a really pretty wide distribution of sophistication. So, there are some clients who just want a data feed, they just say, ‘okay, we’re fine’. Just give me terabytes of data, every week, or every month. And we’ll just have fun with it and talk to you never, so that’s one set of fines, but then there are others that really, maybe don’t have teams of data scientists or data analysts, to explore this on their own.
So, they’re really looking for a little bit more insight. So, a little bit more of that curation on our end. And so what we’ve done there is we’ve created a dashboard. And this dashboard is really popular now. So it’s just an easy way to navigate with this complex set of data and get insights really quickly.
So, we built a few new features recently, typically, we have just tracking the workforce dynamics, you might want to, compare the composition of companies today, for a set of companies, there’s also being able to track those changes over time, let’s say the share of the company that is, that has skills in TensorFlow, that might be a useful metric. Also tracking transitions, where do people come from?
Where do they go, that is a very popular feature. We also recently put together a screener, which is very useful for private equity firms who are looking to source new deals. And for us, we’re lucky in that the data set is not specific to public companies, private companies, big or small, doesn’t really matter. As long as the company’s got employees, that’s part of the database.
So, being able to source companies through a screener, can be very powerful for certain clients. And then for HR, also, we have a talent acquisition tool. So, you can discover new pools of talent, and also be able to use the taxonomy that we’ve created. So, you can navigate the job titles that are similar to the skills that are similar to the activities, and be able to, be able to impose a lot of structure on data that they might even have internally. So, all of those things are new and getting some traction. It’s a very new product, but it’s really starting to take shape. And we’re always changing it, every new client gives us something that we can learn from.
This is all very exciting and can’t wait to see more of that in the product that you and the team are building at Revelio Labs. And if we’re doing one more forward-looking insight for the audience, are there any trends, or items that we should be seeing as signals for the future work that should be on our radar for the next year into 2022 and beyond?
It’s a great question. We do weekly newsletters where we try to report on some of the interesting things we see in the data. So, we have a team of economists who work on some of these newsletters, and very often will collaborate with Bloomberg or Business Insider, or some media publication, to kind of get these insights out there. We’ve done a piece recently. Let me check out if there’s any that we did, most recently. We did one last week on the future of work.
So, what are the HR tech companies that are really dominating, and, then it gets even specific, who’s dominating the self-driving car market, how benefits help retention of women in the workforce, that’s something that we’ve seen some changes in the past couple of years. We did a piece that I really liked, which was tracking the rise and fall of hustle culture.
That’s something we could see, people, having side hustles that’s been declining dramatically, which we thought was really pretty fascinating. So, on our site, you can check out Reveliolabs.com/news. And, we really like doing these, it’s generally really short. It’s like a paragraph in a couple of charts. But it’s always data-driven, and it’s always something non-obvious that we’d like to share
Well, looking forward to more of these insights and trends, and all of that today and moving forward on HumAIn Podcast. Thank you to Ben Zweig, the CEO of Revelio Labs, and adjunct professor at NYU Stern. Ben, thanks so much for joining us on the show.
Thanks, David, it was a lot of fun.
Thank you for listening to this episode of the HumAIn Podcast. Did the episode measure up to your thoughts and ML and AI data science, developer tools, and technical education? Share your thoughts with me at humainpodcast.com/contact. Remember to share this episode with a friend, subscribe, and leave a review, and listen for more episodes of HumAIn.