HRExaminer Radio Executive Conversations Badge Podcast Logo

HRx Radio – Executive Conversations: On Friday mornings, John Sumser interviews key executives from around the industry. The conversation covers what makes the executive tick and what makes their company great.

HRx Radio – Executive Conversations

Guest: Jack Berkowitz, SVP Product Development for DataCloud, ADP
Episode: 327
Air Date: May 31, 2019




Important: Our transcripts at HRExaminer are AI-powered (and fairly accurate) but there are still instances where the robots get confused and make errors. Please expect some inaccuracies as you read through the text of this conversation. Thank you for your understanding.

Full Transcript with timecode

00:00:13:26 – 00:00:26:19
Good morning and welcome to HR Examiner’s Executive Conversations. Today we’re going to be talking with Jack Berkowitz who is new to ADP and you’re going to love this guy. Jack how are you?

00:00:27:24 – 00:00:29:11
Doing great John. How are you today?

00:00:30:00 – 00:00:47:23
OK so. So you’re the senior vice president of product development for the Datacloud. That’s exactly a big company job title. Why don’t you take a moment and introduce yourself tell us how you got there and tell those were the gobbledygook go with title mean.

00:00:47:23 – 00:01:20:27
Yeah that’s great. So you know I came here after a bit of a career in software and also in consulting. Most recently I was I was at a different software company. I was at Oracle working on some some interesting problems about introducing machine learning into business applications. But I came to ADP because of really two big things. First of all just the great reputation ADP has in really driving the economy. Now I was paid by through ADP for years and so were my parents and so far all my relatives.

00:01:21:07 – 00:01:55:11
You know it just was an attractive company to me from that perspective and the other thing was because of that the data that that ADP is able to. Show and deal and process was really really interesting to me as a technologist and we issue a thing called the National Employment Report every month who tells us. You know what’s happening with the economy and that’s all built on that data. So it’s really really interesting to me. What I do here is really three things. I run a really big H.R. analytics and reporting commercial product and we sell that or our customer subscribe to that.

00:01:55:12 – 00:02:25:24
Use it to interpret their people analytics in their people’s situation every day. I also work on on proving our overall product by work on building things like candidate relevancy scoring or chat bots or things like that that’s embedded inside of all of our H.R. and payroll applications to make the systems easier for people to use. That’s really what we’re doing. And the last thing is as I get to spend some time meeting customers which is actually the most important thing to me meeting the customers and meeting the people that work every day.

00:02:25:28 – 00:02:42:23
That’s what I spend a lot of time doing and that’s actually the most important part of my job. This is going to sound like a dumb and obvious question but you know ADP is so broad that I don’t think most of the people in the audience have an actual grasp of what ADP does to you

00:02:43:06 – 00:02:49:02
Package that up and term ADP into something easily intelligible.

00:02:49:03 – 00:03:19:05
Yeah I think it’s a great and a great point. You know ADP helps companies at the end of the day and it helps companies grow by helping them with their most important asset their Pete. And that can include payroll boys done that can crude include things like benefits and taxes. But we do H.R. systems we do things about compliance. We help people recruit and and we even for smaller companies they can even outsource their entire capability to us.

00:03:19:07 – 00:03:41:19
So we run the ability for companies to to even buy in bulk from us. So for benefits and things. So we even have things where they’re employees of a company. Also are able to participate in massive benefit plans. Really big firm but really about helping people. And that’s where about. What we’re all about

00:03:42:22 – 00:04:08:12
So you’ve got this job you’ve had working. I see and you were quite modest in your introduction. I see that you you spent time working with DARPA and the FAA on intelligence systems back at the time when the BEA so the BEA 7 7 7 airplane was being certified. You say you’ve been at this a long time. What’s changed.

00:04:08:12 – 00:04:09:00
What do you see

00:04:10:21 – 00:04:42:15
technology’s changed. Technology has accelerated well haven’t changed. And so you know I think there’s there’s you know interesting capabilities that that people are bringing to bear. You know like the iPhone is only like 10 12 years old and we wouldn’t think about not having it in our hands today. So technology’s changed but people haven’t changed. People still have that same motivation to you know do a good job wake up and see their families and everything else and so

00:04:43:01 – 00:04:55:14
I think the fusion of people in technology or the interaction between the two which is where I’ve spent my entire career continues to evolve but always always for the best.

00:04:55:18 – 00:05:07:24
I think you may know this I have a relative allergy to the term artificial intelligence but but it’s like being allergic to smoke in Las Vegas. You can’t

00:05:09:09 – 00:05:36:27
if you insist that big of a smoke free environments. So rather than spend a lot of time telling people that this is an artificial intelligence I generally let it pass but I’d be I’d be curious if you’ve seen anything that you would call artificial intelligence in a terror attack pure artificial intelligence I think now I think we’re using techniques to aid people and really really clever ways.

00:05:37:00 – 00:06:08:17
So there’s been an explosion for example if shepherds. Right. Well you know sometimes you just look at I mean like well wait a second. That’s just reading a frequently asked question list. But you’re starting to see some really interesting things where those bots can maintain the context of the conversation. Maybe within that conversation or across the conversation. And really helping employees get the help they need in the day. You know the the the candidate relevancy stuff it’s really not about finding the precise candidate

00:06:09:01 – 00:06:39:29
Right for a job. But what it does is help you. Eliminate all those candidates that aren’t relevant. And so you’re seeing an advantage of that. For the recruiter. It’s not replacing a recruiter. But it’s freeing up time for the recruiter to focus on really finding the right person really passing down the 70 percent or 80 percent that doesn’t matter in the flood of your estimates that a company may be getting. So. You know I don’t know that it’s it’s the type of thing where thinking on behalf of an H.R. department or thinking on behalf of

00:06:40:15 – 00:06:54:19
Of of an executive. But it’s tools. That are allowing those people. You know whether it’s to the H.R.. Department where there’s the CSIRO whether it’s an operational executive. It’s allow it’s giving them the tools so that they can make better decisions.

00:06:54:19 – 00:07:24:15
I heard a story once about about a recruiting leader who was just completely obsessed with his job but he just had twins and so he was in no position to quit just because he felt bad and so he decided to stick around and he decided that what he was going to do was take four years and make sure that everybody on the executive leadership team or company wore glasses

00:07:26:00 – 00:07:56:28
and so he just a recommendation he put forward only put forward people who wore glasses and I think that I tell the story because it’s a it’s a it’s an example of how powerful recruiting can be and the biases that you might not think about wanting to control that are part of that part and parcel of decision making. And so so the question here is you’ve been tossed around a couple of times.

00:07:57:02 – 00:08:08:01
I think you said Kim’s that relevancy is that the term now can relevant to you. And so the question is. How do you tell. How do you tell if your recommendations are right.

00:08:08:22 – 00:08:43:22
Yeah I think I think it’s an interesting question. Let me pivot it onto you. OK it is important to say if it’s right versus if it’s wrong. And so what we have to do is we actually have to measure both sides of that equation. And part of it is what is the right recommendation does that mean that that person should be definitely hired. Does that mean that that person should be considered. And I I tend to move towards that considered because it’s really about a partnership between you know we’re talking about people here we’re talking about behaviors and there’s a lot of sensitivities about that in the case.

00:08:44:06 – 00:09:22:11
And so we have to think about how thing how people should be considered and have that partnership between the recruiter. Or the hiring manager and the machine goes to 100 percent pure automation. I don’t think that we’re there yet. I don’t think any anybody even ready for that. The other part though that you brought up which is about that bias right to that recruiting manager with the glasses. Inserted. Their bias into into the capability. And so we have to be really sensitive to measuring bias. We have to be really sensitive to understanding both sides of the population both the people not recommended and people to be considered.

00:09:22:17 – 00:09:53:19
And we have to include information about. Long term job performance not just what they show up with because. You know I could put anything in my resume. It has to get verified but boy if I can actually understand long term job performance you know how long do people with this profile actually stay with the company and include H.R. data along with the recruiting information. Maybe I can start to see a better thing. It’d be interesting to go back to that story. To find out how well. All those people with classes performed in the company. That’s what I’m trying to get out there.

00:09:53:20 – 00:10:24:23
So there’s there’s an interesting interface question there which is which is I get that get the the story is that these are just recommendations but they’re inevitably scored and then they inevitably arrive on the recruiters desk scored and ranked. And so that means that the recruiter from the moment there is a recommendation from the machine will always have to answer the question why did you hire the top

00:10:26:08 – 00:11:07:18
. Why didn’t you get the top candidate and what you what I think is really the case is its candidates don’t really rank that way. That’s a convenience of interface rather than a statement of truths about people being rank ordered. You know that the rank ordering of people. Is out of favor everywhere but in recruiting that’s true. Also it’s an old fashioned way of thinking about people but we haven’t really spent much time working on the interface so that we understand that this person is two standard deviations out from the norm.

00:11:07:27 – 00:11:16:14
But the role requires somebody who’s a misnomer. That’s right. You get into a linear ranking.

00:11:16:18 – 00:11:46:21
Yeah. And I and I think that you know what you’re really touching on is also explanation. How do we build interfaces that can explain things to people. So that they understand those context right. People aren’t boxes people. You know I. I use the example people are like jigsaw puzzles except that the puzzle the shape of the puzzle is constantly changing. You know who you are today isn’t who you are tomorrow. But to describe you know how that jigsaw puzzle person is in the interface is a challenge right. And so part of it can be

00:11:46:28 – 00:12:18:17
Hey let me just see where everybody is it relative to each other. Right. On multiple dimensions. Or you know quickly at some recruiters under pressure for time. Quickly give you know in line the explanation as to why that person was here versus somewhere else and you’re starting to see some interface designs not just from ADP but in the market itself. That show some of these things. I think we have a long way to go. In in the people’s space. To provide those types of explanations. And to do it in an efficient and understandable way.

00:12:18:17 – 00:12:24:15
But certainly it’s the focus of of my team right now. And I think you’ll see some progress in the future.

00:12:24:17 – 00:12:27:26
That’s pretty exciting. So what are the big questions that you’re working on.

00:12:28:21 – 00:12:59:20
Well we’re going on a few interesting things right. You know I think one of the things that we’re working on ADP is what does it what does it mean to to generate a really engaged set of employees and then what’s the business benefit or the positive benefit either for the employees or the company is it. And as a result of that. So you know we have a group with us the markets Bunky ham company that really focuses on engagement and the acronym for that is team B.C. and we’re seeing some really neat result where we marry engagement.

00:12:59:20 – 00:13:38:26
And in this case engagement is about first line managers or me as a manager engaging with with my next level or first line and then seeing actually tying it into business benefits. Just the beginning to this but it’s a huge deal right. People have talked about engagement each year for years. Nobody really knew had a measurement. Nobody really knew how to tie it together into business results. And we’re seeing that and we’re able to do that. So that’s one big thing we’re working on. The other big thing that we’re working on a big challenge is you know people work by their org chart. Maybe they did in factories a hundred years ago but I’m always being grabbed into dynamic teams and it’s a bit of no overwrought phrase at times but

00:13:39:10 – 00:14:13:12
You know what it’s how people work. So. We’re working on you know how do you deal with teams but then how do you have those teams reflected into. Either metrics or recommendations or other types of problems. And that that’s really cool work. And then and we’re working on data itself one of know one of the interesting things about ADP is we have so many customers we have you know over 700000 customers and you know when we look at something simple like a job title. How many different ways 700000 customers describe a software engineer

00:14:13:23 – 00:14:52:17
Or a retail person. You’d be surprised. So how are those skills. Lined up and how do those things come together that allows us to do things like benchmarking so I can tell you what average salaries are. I can tell you. How long it takes to fill a certain type of job or I can tell you. You know what the turnover. Expected turnover is for that type of role and position aligning all that understanding. You know like that 20000 or 30000 different ways people describe benefit plans and then oh by the way these are all prescription plans and doing all that work which is really not outwardly interesting but to a data scientist or to an engineer or to an H.R.

00:14:52:17 – 00:14:58:04
person is really important. That’s a big challenge to us. And we really enjoy doing that type of work that’s interesting.

00:14:58:04 – 00:15:28:29
So so so a couple of things. I think you understand that it’s my view that the that the intelligence tools questions in H.R. are the most complex intelligence tools questions because because what we’re trying to do here is apply tools that are designed for games that have we understandable rule sets or widgets that can be counted.

00:15:28:29 – 00:15:57:21
We’re kind of part of those sort of two dimensional Lumia approaches to dynamic rapidly changing complex systems. And the people are complex just resolutions or complex systems. And the other section for starters is the sort of mapping that you’re talking about. But in the next layer of complexity it’s going to turn out that that there’s a reason that the roll call so different

00:15:59:28 – 00:16:16:18
that reason that girl calls so be different is is pretty dynamic. So the question here is where do you see this going. Where did Where where do you see this go. And do you buy the idea that this is most complex project.

00:16:16:27 – 00:16:54:21
I’ll start with your second question. I do buy that it’s most complex a I project I actually believe that it’s one of the most complex forget about a I think people in work and the future work is just the most complex problem that someone can try to attack. And and and it’s because of exactly what you were saying right people everybody’s different individual and they change over time and then companies or individuals. If you think about it that way or groups inside companies and they’re constantly changing and shifting and so it’s incredibly complex.

00:16:54:28 – 00:17:28:09
And where’s it going is is is down that line of what you are talking is. Is actually understanding companies themselves and in using that inner played with with the people. And so we actually we do a little bit of this today. We have the ability to look at the organization of a company and even looking at you know sort of the different percentages of people assigned to our India versus the people assigned to operations versus people assigned to sales and then cross that by you know about thirty five hundred industries that we manage inside of that

00:17:28:23 – 00:18:06:00
We can actually tell you by looking at that you know the nature of a company that that you may be in. Oh well you know you’re more on D centric you’ve got this sort of proportion of sales and this is that growth of employees. Okay you’re in a high growth company. OK. So therefore your your prescriptions for for recruiting are your prescriptions for managing your talent or your prescriptions for managing your compensation. Give me slut suddenly different than maybe you know a company. Even in the same industry but with a different profile and mix of people resources being deployed because they may be a later stage company.

00:18:06:00 – 00:18:45:06
So we can see the beginnings of those sort of organizational dynamics and the changes in them over time. I think harnessing that over the next few years is going gonna be really really interesting how that comes together and then how that turns into practical recommendations practical advice in H.R. that’ll be really really really cool. So as opposed to having to read a book and then try to figure out what your company is in in in terms of some book that you were at actually seeing yourself as a company you know in the concert of other companies in your business and then your people in concert to that really big challenge.

00:18:45:06 – 00:19:05:15
But so I think it’ll help make the H.R. team’s way more effective over time. A lot of education right. You and I have spoken about that before you know hard problems but it’s contingent on companies like us to also give the advice. If you hit a challenge hey maybe maybe this is some options that you have be able to deal with those challenges.

00:19:06:15 – 00:19:19:29
I’ll tell you what I would. I would love to have a really deep look at the at the evolving company data set. That’s the most interesting project I’ve heard about so far and I’ve been looking at this stuff pretty hard. What

00:19:19:29 – 00:19:57:19
What do you think the ethical issues are. Well I think we’ve had a few ethical issues right. We touched on bias already. That’s a big issue right. There’s an individual right about information I’m a strong believer on the one hand I deal with data about people and companies. That’s what I do for a living. But I’m a really strong believer in individual rights and privacy. Well in our company is as well. And so I think there’s there’s there is a responsibility there’s a responsibility whether it’s from ADP or any other company that’s touching data and that responsibility isn’t even something that needs to you know it’s not really a lawyer do interpret it.

00:19:57:21 – 00:20:33:17
It’s an individual in a company or not to be able to interpret. Is this right or wrong about what we’re doing with the information. And then at the end of the day it’s ethics and so there’s there’s things that people have asked us to look at that that we as a company just won’t do. All right. We’re not going to look at that type of information but then we also have sometimes have the inverse responsibility. Of pay equity is a wonderful example. I’ve got two daughters and a son I’m married to an attorney. Why should my daughters or my my spouse be paid less than somebody else doing similar work.

00:20:33:17 – 00:21:09:29
There’s no reason for it. And so we also responsibility at times to provide information that can allow a company to manage the pay equity situations that they’re dealing with. So you know lots of interesting ethical problems. We have a group here at ADP that we focus on those ethics we meet we review. We understand. We try to stay pace with personal and data rights both legislative and literally what’s right. And so our ethics group we we have these discussions and we try to apply them into product or into the services that the company offers

00:21:11:15 – 00:21:13:07
awesome and so.

00:21:13:08 – 00:21:18:21
So where are you headed next. What’s the what’s what’s right on your horizon line.

00:21:18:22 – 00:21:49:12
So we’ve got a bunch of product there we’re gonna be launching during the summer and then you know culminating in a chart tax. So can’t really get into the details of it all but H.R. tax is a big event for us every year. And you know we’ll be showing off some really interesting things. Our big thing is about customer success. If you really want to know what we’re focused on right now it’s meeting with customers daily working through. You know some use of the technologies and use of the data. To get some business results.

00:21:49:12 – 00:22:20:11
And so we have some great stories. About customers saving. You know not just one hundred thousand here one hundred thousand there but million to dollars for example. In overtime costs simply by providing information to their managers. So H.R. instead of H.R. just doing spreadsheets and sending out monthly reports actually enabling some capabilities to put metrics about turnover and overtime right into the hands of their operational first line managers and take millions of dollars in savings. So one thing that we’re going to be talking about a lot over the next

00:22:20:23 – 00:22:39:11
Year are not just those results but how companies start those programs how to get going. So that you can you can see the H.R. business function you know really being part of of the overall company’s success. So that’s one thing that that’s one of the big things for us is focused on that customer success.

00:22:40:09 – 00:22:58:29
So last question. You know one of the things I do is track the emergence of new vendors who are offering some kind of intelligent tool and I can count seven hundred and fifty small Venture Finance startups

00:23:01:01 – 00:23:10:14
who are who are bringing some sort of was swearing at the problem. How do you think you deal with that volume of supply

00:23:11:28 – 00:23:42:15
Yeah and you know having done a few startups myself you know you always see my startup is going to be better than yours. I think I think we we at ADP think about things as an ecosystem. We we even have a thing called the ADP marketplace where some of these startup companies. Can be integrated along with our H.R. systems. So that you can try them out if there’s a specific niche value you can get out of them. Go for it. I don’t believe any one vendor is going to have the full solution overall

00:23:43:00 – 00:24:16:02
And that’s why we started the marketplace to provide a place for these vendors to play. I think though that there are some interesting trend. You know when you see things and I’ll plug a company that I’ve got no interest in but I think they’re cool. There’s this little tiny company in New York called Work Olympics. Where they’re looking at how people use tools like Jira and. Confluence and slack. And seeing productivity across those types of tools. So there’s there’s always going to be a new idea coming in a startup.

00:24:16:06 – 00:24:57:15
You know the question is whether or not to use that startup is is do you need that benefit. And you know certainly a question for H.R. people to say is I think is stable enough. Because the one thing is is if they don’t have enough funding or runway or momentum to last a couple of years. You might get burnt with your employee base. And so you know balance but innovation with some stability or some backing. You know having there having a vendor like a into a something like our marketplace or one of the other marketplaces out there helps give them a little bit of strength that they may not have on their own because you know the sharing of data and the connection to systems is really important.

00:24:57:15 – 00:25:07:09
So I’d I’d look at one of those startups to make sure that they’re somewhat aligned into a partner with one of the more major vendors. Before you dive in too deep.

00:25:07:09 – 00:25:13:25
Cool. Do me a favor introduce me to somebody at work with. The first I’ve heard about them and there’s some there’s some for me.

00:25:14:11 – 00:25:45:26
Yeah I just met them one or two times. It’s fascinating stuff and certainly I have my eye on it. Yeah yeah okay. So we’re we’re through our allotted time servicing that you want to be sure that somebody takes away from this conversation. You know I think the thing is that whether it’s here at ADT or other other vendors in the space people are working to make the data and the information more consumable and easier to use. And we’re finding huge successfully when people just lean in a little bit to the information particularly in H.R.

00:25:45:26 – 00:26:02:02
department. H.R. people can be you know apprehensive about it but I think if you if you go into with your eyes open you can get some great benefits. We’re here to help for for people that are in shit and talking but but I think overall I love the way the industry is heading. I’m just excited to be part of it.

00:26:02:03 – 00:26:06:03
So take a moment and reintroduce yourself and tell people how they might get a hold of you.

00:26:06:21 – 00:26:18:24
Yeah my name is Jack Berkowitz SVP of data cloud. You can get in touch with me really easily just drop me an email Jack dot Berkowitz at ADP dot com and I’d be happy to talk with you.

00:26:18:24 – 00:26:41:06
Great. Thanks so much. This was a great conversation. You’ve been listening to HR Examiner’s Executive Cnversations and we’ve been talking with Jack Berkowitz ADP’s president of all things data is how I will send you off Jack a shorter title. Thanks again for doing this, appreciate it. And thanks everybody for listening in. Talk to you soon.

00:26:41:06 – 00:26:48:27
Bye bye now.


Read previous post:
Keeping your Eyes on the Prize: A Parable for Leading AI / System Innovations

“Any large system, especially as it ramps up, is going to be operating much of the time in failure mode....