HR Tech Weekly Logo

Hosts Stacey Harris and John Sumser discuss important news and topics in recruiting and HR technology. Listen live every Thursday at 7AM Pacific – 10AM Eastern, or catch up on full episodes with transcriptions here.

HR Tech Weekly

Episode: 79
Air Date: July 14, 2016


This Week

This week John and Stacey discuss:

Erin Spencer sits in on this week’s show for Stacey Harris.

  • Thanks for participating in the Sierra-Cedar, we’re closed now and looking forward to sharing insights at HR Tech in October
  • Pokémon Go Link
  • Automation of Jobs, top down vs. bottom up?
  • Tesla and driverless cars
  • Workday acquires online learning company Zaption Link
  • The Power of Dirty Data
  • John live from Vegas – Ceridian conference updates

About HR Tech Weekly

Hosts Stacey Harris and John Sumser discuss important news and topics in recruiting and HR technology. Listen live every Thursday at 7AM Pacific – 10AM Eastern, or catch up on full episodes with transcriptions here.

HR Tech Weekly Episodes

Audio MP3





Begin transcript

John Sumser: Good morning and welcome to HR Tech Weekly. One step closer with Stacey Harris and John Sumser. As is becoming the norm, we have Erin Spencer sitting in like Kenny Loggins used to sit in with Jim Messina. Welcome, Erin. How are you?


Erin Spencer: I’m doing great. How are you doing this morning, John?


John Sumser: Okay. I hear there’s some rumor that Stacey maybe coming back to …


Erin Spencer: There is a rumor afoot. I’m hoping soon. I enjoy all of our conversations John but my life will be a little simpler once Stacey gets back in the office. We wish you all the best that she deals with her family situations and she’ll hopefully be back speaking with you guys shortly.


John Sumser: Fantastic. Although we may have to make her fight her way back here, it’s such a great experience.


Erin Spencer: Like in music, first chair, second chair. We have to battle in the marching band every week for your spot on the field.


John Sumser: Yeah. Let’s make her work for it. I have the mail bag, you threw some junk in the mail bag and we’re going to talk about it wouldn’t be this week if we didn’t talk about Pokémon GO.


Erin Spencer: Of course.


John Sumser: I’m starting to see an interesting trend which is that automation rather than being bottoms up in the organization is to start coming from the top-down and we’ll talk about that a little bit. I’ve been at the Ceridian Conference all week in Las Vegas. I’m sitting, looking out over the place where the desert meets the urban area in Las Vegas today. We’re going to talk about dirty data and Workday just bought a company called Zaption. Is there anything else that we need to cover today?


Erin Spencer: I think if we go through all that, we’ll be good but if we need to talk something else and at the end, we can. Let’s get started, John.


John Sumser: Okay. Are you playing Pokémon GO?


Erin Spencer: I am not playing Pokémon GO, which is strange because I have a nine year old who you think would be all over Pokémon GO but he’s not interested in Pokémon so we haven’t started it, but in my age demographics so close to 30 female, I am shocked at how many of my friends and compatriots are A, playing Pokémon GO and B, playing it themselves and not necessarily for their children.


Just completely surprised and I was actually out yesterday on a doctor’s appointment and my doctor was telling me that people were in her office on their phones with the apps and by the way there are a particular Pokémon in the parking lot, if I needed to know that, so you know that it has completely taken over.


John Sumser: What’s wrong with you? Why aren’t you playing?


Erin Spencer: I have enough to do. Well, I’ve been busy, John which is something that we haven’t talked about yet which is our slightly, remiss, we have officially closed the Sierra-Cedar HR System Survey for 2016, so I had been cleaning our data furiously trying to get ready for our debut at HR Tech the first week of October.


Even though it’s only the middle of July, the weeks between the middle of July and the beginning of October go very very quickly here at Sierra-Cedar with our data and the cleaning and analysis that we do before we can actually present in October.


John Sumser: Do you think the study is invalid this year because there were no Pokémon GO questions?


Erin Spencer: John, it’s strange that you mentioned that. It’s not that we did not ask specifically about Pokémon GO but we actually did ask about things like wearables and then when you talk about emerging technology, we’re asking about what organizations would you find valuable from their HR technology environments.


One of the things often asked about is gamification and badges, and it actually was out of the five things that we ask, it was right in the middle but I’m wondering if Pokémon GO had come out a few weeks earlier, people had seen the impact that this silly little video game could have on motivating people to get up and off the couch that gamification and badges might have been a little bit more popular.


I was seeing a little picture of … there’s little memes that come out on Facebook and they said that Pokémon GO has done more for childhood obesity in three days than Michelle Obama has done in eight years in office which is unfair and untrue but amusing the idea that this video game really is getting kids and families out together and moving to learn to play the game.


John Sumser: Well, I’m not sure that it’s unfair. I think that this phenomena which is in three days an app overtook incumbents who had built their audience over 10 years, like Twitter, in terms of utility as an app, that’s a phase of technical change that I think is becoming the norm. Imagine that the end of Facebook happens that fast.


Erin Spencer: It could.


John Sumser: The end of your job happens that fast, that instead of some long planned plotting corporate initiative to automate things like you do, in three days it’s done and you’re out of work. That’s what happen with Pokémon GO, the technology was installed overnight without much in the way of server compromise, and people change their behavior radically in three days.


Work is going to be subject to those rhythms, by your work, my work, it’s going to be subject to those sorts of rhythms, so Pokémon GO is an important cultural marker because it shows how rapidly technical change can happen. That’s cool. I bet that’s something that ends up in next year’s research.


Because if it’s possible to have change at that pace inside of organizations, a lot of what we know about HR technology is about to change.


Erin Spencer: Absolutely. We’ve seen just in the last 10 years mobile explode and organizations really learning how to use it and harness that power. What does that look like for the next 10 years. When we talk about job automation, John. You and I were talking about this a little bit earlier, but people were talking about the minimum wage a few weeks ago, and the idea that once machines become cheaper then essentially labor you won’t have people doing the same jobs that they used to do in the same way.


Your order-taker at McDonalds isn’t going to be a real person, it’s going to be a machine or a screen. We’ve been seeing that, but you actually had a different theory on who’s going to lose out in the automation game.


John Sumser: This has been dawning on me for the last week that the most routine and repetitive work that gets done in the organization is management, particularly micro-management. To the extent that a manager’s job resembles a foreman in the 19th century overseeing the work of people under them, giving goals, delegating stuff, following up on the delegation to make sure.


These are all things that are easy for machines to do. Really easy for machines to do. I’m starting to think that the wave of automation, all of the stories about automation had been about the loss of hourly jobs at McDonalds because robots are now flipping burgers.


I think that maybe the shift manager at McDonalds is could be the thing that really gets automated quickly. I was just talking to an HR tech department in a very large company, and the 40 or so people of that department, I said “You guys could count on being able to keep your jobs, but if I was your boss, I’d be worried.” Because what the boss does largely is monitor and control projects. That’s another thing that’s easy to automate.


Erin Spencer: Absolutely.


John Sumser: To the extent that management is a clipboard function where you wander around and check on people to make sure that they’re doing stuff, and eventually processes so that everybody in the entire world fits into a pipeline that chunks along, that’s easy to automate. Decision making about the actual work, that’s how it got automated.


It seems to me that the places that are safe from, safe is a weird word to use, but the places where automation will be slower to catch on are places where refined judgment is required. If you think about at least my experiences with the cable company or something like that.


I really am in the part of the world where you can’t automate human knowledge. The ground workers are the people who actually will be there and the scheduling system that makes sure that Sammy is the guy who answers the phone when I call, that’s where the automation goes.


If I was a line supervisor, committal manager in a Western company, I would start thinking about how to learn to do some actual work because … Yeah. This is sort of Dilbert’s dream.


Erin Spencer: It is.


John Sumser: That guy with the tps reports. He’s the one who’s going to get automated out of a job. I think it’s fascinating and I am really sure that almost no line supervisors and middle managers are going “Yeah, I’m the next one to get automated.” They’re all thinking that its their employees who are going to lose their jobs.


Erin Spencer: Absolutely, well, and nobody wants to think that it’s their job that’s going. John, we were talking about a month ago about the cornerstone event that we attended. A lot of what they were doing was predicting jobs and who should take over what and doing succession planning.


They were using the software to do that and again to go back to the survey, predictive analytics is something that a lot of the HR technology would find extremely valuable in their technology tools. Those are decisions that individual managers used to make or obviously continued to make at current companies.


If you’ve automated that and you have the machine learning making those decisions for you to show you what they think is the best option, that definitely takes out the human component which some people could argue it could be bad, and then other people could argue that you got obviously some diversity issues in organizations if you take out some bias in hiring, that could change.


John Sumser: Yeah. It’s so interesting to live today because so much is changing and we, I think the place where I am most likely to encounter my biases is in my imagining of what the future is going to be. That’s where do you see the stuff that I really hope will happen. That’s where my biases show, in my hopes and dreams and fears of the places where I put bias into the system.


It means that I am in general less likely to see what’s about to happen that somebody who’s stepping right next to me, it’s an interesting phenomena, and we’re going to get really tested as a species on that one.


Erin Spencer: It’s interesting you talk about biases, my son who’s nine years old is fascinating by the idea of automation and what the future is going to hold, and the idea of flying, everybody having their own airplane and flying cars and things. He says “Mommy, I don’t think that they should make driverless cars.”


I’m thinking to myself, well you’re not the kid who have to drive around the carpool, I think the driverless cars would be a great idea. It all depends on the idea of where you’re coming from. [crosstalk 00:14:26].


John Sumser: You should tell him that, it might be right. I’ve been following the Tesla accident where the car didn’t anticipate the situation where the truck that it ran into would blend in with the sky and that the camera wouldn’t be able to tell there was a truck there.


That problem which is that the human driving the car trusting the car to do what it seem like it would do is a design problem and in some of the articles that I’ve read, they’re championing the Mercedes approach to that which is to never let you lose your attention to driving the car even though it’s got the same amount of intelligence as the Tesla does.


In the Tesla, there’s a web browser with cell-based web access that you can play with in the dashboard, on a 12-inch screen. When you put it on autopilot, there’s a thousand distractions, you can go read the newspaper on your web browser in the car when it’s on autopilot. In the Mercedes, they won’t let you do that, the Mercedes interrupts its autopilot to make sure that you’re paying attention.


Erin Spencer: One of the appeals though would be that you wouldn’t have to pay attention while you’re driving.


John Sumser: The Mercedes people don’t think that’s safe. What’s fascinating about all of this is that we’re about to enter into a set of conversations about what is and isn’t safe in HR software. I don’t think that conversation about safety in HR software has ever been had before.


I don’t know anybody who’s talked about the safety of HR software, but there are some extraordinary issues popping up like what if the recommendation that your HR software makes for my next interaction with you causes me to do something that you should help me for or that causes you to flip out in a way that causes damage to other people.


Where’s the responsibility for that decision line? It’s an autopilot kind of question. If the system says if you just punch Erin in the nose, she’d be a better worker, and I punch you in the nose because that’s the recommendation. Who holds the liability for that? That’s the question I think that we’ll have to spend a lot of time on.


Erin Spencer: Well, that’s frankly if people had been making unethical-ish decisions like that for years, “I don’t want to hire Erin because she’s a 30 year old female and she doesn’t have any children and she’s been married for two years, and guess what’s going to happen and I don’t want to pay for maternity leave.”


John Sumser: Right. Yup. If you get a machine to recommend that decision then the liability probably extends to the machine provider.


Erin Spencer: If that will be an interesting conversation in the next few years is liability and the lawsuits that come from that.


John Sumser: Is there anything exciting that you’re noticing in the research that’s coming out of, you’ve got something like 25% more responses this year?


Erin Spencer: We do. I really just want to say thank you to John for promoting the survey again on your site and making sure that people heard about it every week. We’ve got an extra 300 responses more than we have last year. We had about 1200 responses last year, we’re going to have a little over 1500 responses this year.


Between you and the outpouring of the entire HR community, this has just been such a fabulous year for us for data collection and I’m currently cleaning the data and working on just some of the basic stuff that we do with our data, so I don’t have any really great conclusions yet, but I can promise the vendor data this year is going to be very entertaining. Be on the lookout for that one so you start to share it. I’m sure Stacey will be sharing some more highlights in the next few weeks as far as interesting stuff that’s coming out of the data.


John Sumser: On the acquisition front, Workday bought a company called Zaption. Did you see that?


Erin Spencer: I did. My first thought was what on earth is a Zaption? The name doesn’t really tell you anything. Reading some of the articles. It’s a website where teachers and companies can host videos or use existing videos for teaching purposes. Basically you can put contents on top of other videos of YouTube and such and capture responses and run analytics.


It’s an education [E thing 00:20:13]. Workday is definitely doing something within the learning space, so John, what do you think about that?


John Sumser: Well, I’ll tell you what, from what I can tell. I had never heard of Zaption and part of what that goes to show you is that it’s impossible to know everything about this marketplace that we’re in, it’s exploding with investment and ideas. It’s an exciting frame.


As I learned this learning about Zaption, it seems to me that it’s similar to a company called GuideSpark. In that the core of the offering is a tool for rapidly developing course materials in video. Zaption allows you to edit and augment a video in order to tailor it for learning purposes, and GuideSpark has the same sort of capacity plus actual video development tools and animation development tools so that you can quickly get to polish output.


It seems to me that this has two things. One is that there is a curriculum development facility that Workday needed to acquire. Workday is trying to build a single platform and they are pretty committed to having one code base, so if they buy something like Zaption, they’re buying the expertise, buying the technology and making a commitment to code it into their platform.


This is a great hint about what the capabilities of the learning system are going to be and one would expect that the learning product at Workday which is I believe in the market and undergoing fast fast development is going to be accelerate by this. It’s kind of an exciting announcement.


Erin Spencer: That is exciting. We’ve been seeing a lot of change going on in the HR marketplace of course. Obviously one of the things that you need for change is data and to be able to make good decisions, but John, you and I were talking a little bit about data, especially as we were closing the survey, and you had a theory about dirty data. Would you like to share it with our audience?


John Sumser: Great. Thanks. Thanks. I happened to be in a conversation with some people who were working with data, and I said “Let me see your data. Let me see your data.” And they got all sheepish. They said “It’s not really very elegant. We’re embarrassed about it because our data is captured in spreadsheets. And it’s snippets of this and snippets of that, and tags, and it’s the rawest form you can imagine.” I said “Yipee.”


They lit up and then I said “The thing about data is that clean data is for executives. Clean data, what clean data means is that you standardized on certain terms and you standardized on certain answers and you narrow the expressiveness of the source data so that it’s easier to sort. By standardizing and throwing out crap and adding little bits here and there, you make the data easier to sort so you can get it into nicer tables and PowerPoint presentations and [reference 00:24:28] and stuff like that.”


Those things are well and good, but they’re all about communicating conclusions. The real insight in data comes from consuming data that’s not been filtered in that way. This is kind of like, why do you drink raw milk? It’s because the pasteurization process. I wonder if we can talk about the pasteurization of data.


The pasteurization process while it gets all the bugs and dirt out of the milk, also destroys all sorts of really important crystal stuff, and there’s this balance between healthy and dangerous that you thread, and the closer you get to dangerous, the higher the nutritional value.


It’s one of those paradoxes of modern society. The cleaner something is, the less interesting it is. When it comes to data, it seems to me that what one does is immerses oneself in dirty data till you can’t bear it any longer, and then you let your mind work and you let your mind work by taking a walk or taking a nap or having a meal or playing with your kids or doing something that decidedly doesn’t look like work while your brain goes off and actually does the work. Then at somewhat hard to predict later time insight immerses.


It’s less easy to do that with data that’s been standardized because you get the feeling when you sort data easily, that sorting the data is the same as doing analysis, and it isn’t.


Erin Spencer: It is not.


John Sumser: This area which is how do you actually do analysis or what does it mean, has not been explored very well and the people who’ve been the dominant talkers in the conversation all make the case that you got to clean the data before you could use it, and I think it’s non-sense.


I think it’s non-sense. I think eventually you have to clean the data, you have to be able to help executive C. As we said earlier, the [shoulder 00:26:57] jobs are all going to be automated so you won’t have to do that for much longer.


Erin Spencer: You think that’ll work with me, John. That it runs the next month, I’m just going to run to Europe and not bother cleaning the data, and we’ll just throw up something on the state of HR tech. You think people are going to go for that?


John Sumser: Yeah. Do that. You’ve got a lot of savings, right?


Erin Spencer: Not enough for that.


John Sumser: Yeah. There’s a balance, but really cleaning data is a second thing, immersing yourself with the data is the first thing. Cleaning the data can be done in the cyclical ways so that you extract the maximum amount of data dirty value as you’re going through the process of cleaning it.


In fact, one way of thinking about this is that the act of cleaning the data is where insight comes from. Another way of saying, immersing yourself in the data is looking at it so closely that you can see where the anomalies are that require correction and standardization so that you can deliver it to an executive. I believe that’s where a lot of the novelty insight come from.


Erin Spencer: Yes. Yeah. Of course. John, I know you’re at Ceridian this week, you want to mention anything that’s come out of that so far in our last two minutes?


John Sumser: I just want to say one thing. We’ll maybe talk about them some more next week. Ceridian appears to have perfected something that is blowing my mind. This week, I went to two sessions in this. This is a massive conference with all sorts of people.


I went to two sessions. One was a presentation by a VP of HR at Ceridian and one was a presentation by the CIO of Ceridian. Usually, at these big software development companies. The people who present the product are the people from the product development team or from marketing.


Here, the presenters told the story from the perspective of being an HR manager using Ceridian software even though she worked for Ceridian and it was drinking your own champagne. She was so steep in the software, that she could explain the entire business of HR in the company by referring to the functionality in the software. It was a much more engaging story than some product management person trying to regurgitate the functionality.


When the CIO, this product management people and research and development people over in product management who usually are the ones talking about market trends, the CIO gave a briefing on human capital management technology market trends.


Because he has to solve this problem on a day to day basis, the presentation was way more interesting than the presentation from an analyst or product management person would be because his view of the future had to do with real problems that he was solving right now.


Across the board at Ceridian, they’re taking this view that if they’re not the experts in their company about the product by using it on themselves, then it’s got no business selling it to other people. If they’re that good, they’re key mod would be briefing it. That’s astonishing. It’s just astonishing.


I don’t know anybody else that’s doing it at this level. It’s really exciting to see. We are over time and over budget, and I hope we get to have a couple weeks where we do some transitions, so please don’t go away if Stacey comes back. She may be so in out of shape that it’ll take a while for us to show her how to do it again.


Erin Spencer: Well, we’ll get back her into shape, John, we’ll get her back.


John Sumser: There you go.


Erin Spencer: Have a lovely time in Vegas.


John Sumser: Okay.


Erin Spencer: We’ll talk to everyone else next week.


John Sumser: Yeah. As usual, great conversations. Thanks so much, Erin. You’ve been listening to HR Tech Weekly One Step Closer with Stacey Harris and John Sumser, Erin Spencer sitting in doing a great job. Thanks for listening. Have a great day.

End transcript


Tagged with:  
Read previous post:
2016 07 20 hrexaminer tell your story sumser photo img cc0 via unsplash saint andrews beach australia photo by alex wigan
Tell Your Story

Last week, John Sumser spent several days at the Ceridian annual users' conference. He came away with the view that...