#072: The Future Digital Analyst...Redux...with Corry Prohens

Transcript

[music]

00:03 Announcer: Welcome to the Digital Analytics Power Hour. Tim, Michael, Moe and the occasional guest discussing digital analytics issues of the day. Find them on Facebook at Facebook.com/analyticshour and their website analyticshour.io. And now the digital analytics power hour!

00:27 Michael Helbling: Hi, everyone. Welcome to the Digital Analytics Power Hour. This is episode 72. Have you ever switched jobs on purpose? And do you spend time thinking about whether things will go well when you do switch jobs? I know I used to. In today’s fast paced technology world, you’ve gotta keep your skills current. Back in episode 35, we talked about whether data science was the future of all analysts and then more recently, Gigi Geiger sent us an email about whether or not analytics professionals need to become data scientist to stay current in the market. So it’s time to crack this one back open and spend a little more time, but this time we brought in some big career guns. There is this guy that you will eventually talk to in your career, about your career, and no, it is not Tim Wilson, but he is my co-host. Welcome, Tim!

01:23 Tim Wilson: Who’s it doing, Michael?

[laughter]

01:25 Moe Kiss: What was that?

01:25 MH: You know, I wasn’t even ready for that. [laughter] Why is it doing? Also Moe Kiss, who has a wildly successful career in analytics and is my other co-host, welcome, Moe!

01:39 MK: Hi, guys!

01:39 TW: Do it!

01:40 MH: See, Tim threw me off. I would’ve been ready for you, but then the what’s-ee what’s-it… Anyways.

01:46 TW: It’s what I’m here for.

01:48 MK: How’s it going? How’s it going?

01:50 MH: Is that really what you say?

01:52 MK: Yup. How’s it going?

01:53 MH: Oh, okay. That sounds not hard to figure out. Anyway. I am Michael Helbling. But who I was speaking of before. He’s a recruiter with kind of a golden touch. The man with the jobs and the knowledge to put the right people in them, I am talking about Corry Prohens. He has recruited in the IT world and in other industries, but he found his true calling about 11 years ago, when he opened IQ Workforce to solve the staffing needs of the analytics industry. He’s leading a recruiting and retaining analytics talent workshop at the DA Hub which is coming up in just a few days, and he also surfs regularly from his home in Long Island, New York. Welcome to the show, Corry!

02:38 Corry Prohens: Thank you so much, Michael, greatest intro ever.

02:41 MH: Hey, that’s all I do on the show. Now Tim will ask you some questions and we’ll get this whole thing wrapped up in a nice bundle. No, actually so this is something that actually is very current. I think a lot of analysts no matter where they’re at they’re asking this question of themselves like, “What is it that I need to be… Do I need to become a super technical data scientist? Can I focus on some of these other things?” So, what kinds of things have you noticed as trends in the marketplace and how do people position themselves to make the most out of their careers?

03:14 CP: Sure, well I think that there have been two big trends that have been really going on for years now, and I would say one is, the sort of convergence of the “business” becoming more analytical, while the analysts are moving towards the business. As you mentioned earlier, Gigi asked the question. Marketers are becoming more data savvy. So I think that’s one big trend and it seems to be causing some concern, and the other one is that digital analysts are working with multiple data sources. I don’t necessarily see a trend toward digital analysts becoming data scientists, I really think that those are two very different professions, if you define data scientist the way that I do. But I can see potentially, digital analysts in the future, 5-10 years from now, becoming a little bit more data science-y. So picking up some tech skills, picking up some quant skills, and sort of inching in that direction. But I don’t see everybody sort of growing up to become data scientists in 10 years.

04:37 TW: So when they’re working with multiple data sources, are they going into multiple tools, then exporting and then working in Excel? Or are companies looking at them and saying, “Because you’re working from multiple sources, either we’re feeding all our data into a data warehouse and you’re gonna work with the data there?” Or are they saying, “Maybe we don’t need you as a data scientist, but we need you comfortable with Python or R, or SPSS or something to pull the data out?” When it comes to the multiple data sources, how are companies expecting analysts to work with that data?

05:14 CP: Yeah, so I think the tools are different and the data set up is different obviously, company by company. And so is the structure of the team in terms of who does what in the analytics organization or multiple analytics teams. But I think at the end of the day, if you’re a digital analyst and you want to advance your career, you wanna be able to solve problems, answer questions about the customer across different touch points. So you need to be able to use digital data, offline data, customer data to synthesize an analysis. Yes, in some of those cases, a job will require somebody to have the technical skills to sort of pull all that together, but definitely not always.

06:05 CP: The usual paradigm is, in a larger organization there’ll be multiple people whose jobs it is to break that down to smaller pieces. Big company, small job. In a smaller shop, they might look for one person who can do all that, but obviously, the more things that you look for one person to do, the harder that job becomes to fill. And you tend to sacrifice some of the big skills when you’re looking for somebody who can accomplish all those technical goals.

06:41 MH: Yeah. So, I wanna jump back for a second. Near the beginning, you said, “The way I would define data science or data scientists… ”

[laughter]

06:49 TW: Now you’re gonna make him do it.

06:50 MH: No. I feel like it’s actually…

06:52 CP: I’m ready for it.

06:52 MH: It’s helpful, because, Corry, you talk to a lot of companies who are searching for those kinds of roles, so I feel like your definition is actually gonna be helpful in terms of understanding where the marketplace is. So, you’re probably a good authority to say what a data scientist is or isn’t. And, of course, Tim will disagree with you, but I’m here for you.

07:14 TW: No, to be fair, I say that it is defined many different ways by many different people.

07:19 MH: Well, sure. Anyways.

07:19 TW: Your obsession, Michael, is that they have to be building product.

07:23 MH: That is.

07:24 TW: That’s good. I’m glad, Corry’s gonna say it’s not building product.

07:26 MH: It’s fine. Go ahead, Corry. Don’t let Tim… [chuckle]

07:29 CP: Well, I’m not here to break any ties but…

07:30 MH: Influence you.

[chuckle]

07:32 CP: And we certainly have… If you look at our website right now, this is not a plug, but if you look at our website right now, there are a whole bunch of data science jobs up there, and they’re all different jobs. So, I think that sort of goes to Tim’s point a little bit. But I actually saw on a guy’s resume or rather on a…

07:49 TW: Wait, hold on. That’s score one for Tim? I know you’re not breaking ties, I just wanna make sure.

07:53 CP: If you are keeping score, then yes.

07:55 TW: Okay. [laughter]

07:57 CP: I saw… I’m gonna put one of Mike’s column in a second…

08:01 MH: There you go.

08:02 TW: Oh, God! Dammit. [chuckle]

08:03 MH: I live for this. [chuckle]

08:05 CP: I saw on this guy’s LinkedIn profile, just a couple of weeks ago, a line that says, “I program computers to solve math problems.” I was like, “That guy’s a data scientist.” So, to me, the definition of a data scientist, there’s a few different components to it. Number one, you’re working with real big data. Number two, you’ve got academic and professional quant chops, meaning you can seriously do modeling from scratch, or develop models. And number three, you’ve got some serious programming skills. And you put that all together with somebody who understands your business, and you’ve got yourself a data scientist. The reality of the situation is there are probably… I’ll make up a number and say under a thousand people in the world who fit that description. So, that’s the strict definition. And then, depending on how the company is structured, they’ll have a data science department that is made up of people who have those skills, and cumulatively, they form a data science team.

09:21 MK: Corry, I actually really like your definition. I think a lot of people kind of use the term a bit willy-nilly, but in Australia, my team is quite different. In the last two roles I’ve been in, we have digital analysts and we have data scientists, and we’re in one team, and for 95% of the time, we do exactly the same work. We put out a role, maybe a couple of months ago now, and there was an expectation. It was an analyst role, and they had to know SQL or Python. You had to know one of them. So, I definitely think we’re being, I guess, moving in that direction.

10:00 TW: That was a data science role or an analyst role or…

10:03 MK: No, that was an analyst role, and the analyst role said you had to know one of those languages. And everyone on our team knows one, if not a bit of two and that’s…

10:14 TW: So, when you guys go out to lunch as a team, do the data scientists have to buy because they’re getting paid more, even though they’re doing the same job?

10:21 MK: No. Well, that’s a whole another level of discussion.

[laughter]

10:26 MH: Uh-oh. Let’s just make a show about that.

10:28 S?: I hear you.

10:29 TW: I want Michael to make us back up and have Corry define “big data” [10:32] ____ and what it’s about,’cause that was…

[overlapping conversation]

10:36 MH: Yeah, that. We all want. The thought had crossed my mind, but no, I’ll stop short of that, ’cause I think there was a couple things. A, I think my definition is going through some nuances, so I don’t necessarily think that product is always an outcome of what a data science does, but I do think a data scientist takes that math skill that Corry was talking about, and statistical capabilities, the programming concepts, and is able to bring solutions to business problems to bear with them. And that is completing the data scientist, that third piece. And sometimes that means…

11:06 TW: The way the business…

11:08 MH: So, being able to have the business acumen. Sometimes that comes out in terms of things that the rest of the business can use to actually utilize data that they couldn’t have used before, until the data scientist created an effective model and system for it. But reality is, there could be analysis or insight generated against a specific problem, but that’s where the application of both the programming skill and the statistical and the modeling capability come to bear. ‘Cause a lot of people can program in Python, but can’t solve business problems.

11:42 TW: Well, but it’s interesting, and I know, Corry, you were just kinda rattling it off very well, but it was, working with big data, real quant skills and some level of programming. So, it was three things. And then it was like, “And able to bring it to bear to the business”? How core… If somebody can crank through on those three elements, do they also individually need… Are you saying the companies are expecting them to also be able to map to the business problem? Or are companies saying, “No, we have other roles,” maybe it’s a traditional analyst, maybe it’s a business analyst or something else, who will work with them on that alignment with the business that they’re actually solving the right problems? Or is that just the difference between a capable data scientist and a really effective data scientist?

12:39 CP: Yeah. So, I think that all of the jobs that we work on, whether you call it an analyst or a data scientist, I think of it in terms of just a simple Venn diagram with three circles, you’ve got the quant skills, you’ve got the tech skills, and you’ve got the business skills. So each job is going to overlap differently based on what the business needs, based on how the team is structured. So there is no one answer to that question, Tim. It really depends on the company and it depends on the situation.

13:14 TW: Well, I feel like, and I now wish I had the forethought to ask you to go check your system. Right now… And even I’m sure right now when we’re recording and right now when this actually launches, there are postings on your site that say data scientist. And you, I mean, to your credit, 11 years ago, you sort of staked out a spot in the digital analytics industry and have kind of been the go-to recruiter. Do you know or can you find out, I’ll throw it in the show notes afterwards, when the first time you actually posted a position that was data scientist? Because I feel like the technical, we say technical skills, but in my mind, five years ago, when we said technical skills, we were talking about JavaScript jQuery, the DOM, understanding the data collection, and in that definition of data scientist, the technical skills is much more about getting data out of systems, not data capture. So, I think I just blathered on and asked four questions, but would you like to respond?

14:25 CP: I would love to respond. First of all, no, I don’t know when was the first time we, I could guess, it’s probably about six or seven years ago, that was the first time we posted a job with that title. The other part, where you said, if you back up five or six years ago, when we talked about technical skills, we were talking about the implementation side of digital analytics, the integration of data. And I think when you’re talking about we, meaning the digital analytics community, yes, that’s what we were talking about. But still I think if you go back five or six years ago, when you’re talking about the data science community… What was his name, Davenport, help me out…

15:12 S?: Tom Davenport?

15:13 CP: Thank you. Tom Davenport. Definition of data science, I think that the same sort of paradigm existed. On the technical side, it was the programming skills. Have those programming languages evolved and become more prominent since then? Of course they have. But even then, let’s say it was SaaS instead of R and Python, but you still needed to be able to program. And then on the quant side, it was still very much statistical analysis or beyond statistical analysis, modeling. So I think with that the data scientist Venn diagram hasn’t really changed that much in that period.

15:55 MK: Just speaking as a current analyst, I do think that kind of merge between, I guess how much you have to learn, also depends a lot on the company. If you have, and in some of the companies I’ve worked in, where you have an amazing data team that have capacity to give you lots of hope as an analyst, you’re kind of not pushed towards learning those more technical things about how you actually extract data out of an API, and clean it, and prepare it, because you’ve got I guess, a team supporting you. But when you’re working in a smaller company or you have a team that, I guess, you don’t have a data team or you don’t have a data team that have the time to devote to, I guess resourcing analyst with the data that they need. You really are kind of on your own, and you have to learn those skills, because otherwise you can’t do your job.

16:45 CP: That’s absolutely… I couldn’t agree with that more. And I saw… Forgive me, I don’t remember exactly where I read it, but I should be giving credit to somebody. But I saw something the other day about data scientists, for every data scientist that’s hired, two engineers are hired that enable them to do their job. Basically, they get the data cleaned up and make it decent enough that you can either do some modeling and create some kind of work product out of it. See, I threw the word “product” in there for you, Michael.

17:17 MH: Perfect. Thank you.

17:18 CP: You can put that…

17:21 TW: Data engineering seems like one of those things that… Then again, it goes through the organization, they may want what data engineering, which I think is kind of a subset, or one part of a potential, the full scope of a data scientist or is data engineering…

17:35 MH: Well, data engineering has been around for a long time, ever since we’ve done ETL on data.

17:42 CP: Yup.

17:43 TW: Right. But I guess in the digital data world, are there companies that are looking for a data engineer, but they feel like they are gonna have more luck if they advertise for a data scientist? Or maybe flip it around, is that a role of a data scientist can be to do data engineering?

18:00 MH: There’s plenty of data scientists going around, doing the job of a data engineer. That’s happening, I think.

18:06 CP: No doubt and I think ideally, a company would love to hire a room full of fully skilled data scientists who can do both data engineering and the statistical side of the job, the quantitative side of the job. But in reality, you have to put together a team to accomplish these things. And you’re inevitably, once you get to a certain scale, you’re gonna need a bunch of data engineers to pull the data and clean it up. You’ve got a bunch of messy, unstructured data from lots of different places, they’ve gotta make it usable for the “data scientists.”

18:50 TW: So, if we put the… I feel like we kinda went right down where a lot of the conversations wind up being which is, what is big data? What is data science? What’s data engineering versus data science? What are those skills? How do they complement analysts? What do you see on the flip side and, I think this gets back a little bit to some of the exchange with Gigi, around the future and the opportunity for growth and advancement for, call it a traditional digital analyst, that says, “I’ve gotten to where I know the business really well, I can understand the business problems, I’m fine understanding what data science is and what can be done with it, but I don’t really wanna head down into that world.” Are you seeing roles for where that’s what companies are still looking for; is they want somebody who can actually really partner with marketing or the product team and not necessarily get into the world of modeling and the super technical side, but can still bridge that gap between guiding the business into asking smart questions and figuring how to communicate recommendations back to them?

20:07 CP: Yes, I don’t think those jobs are going away, Tim. First of all, if you look at our business, we’re now filling fewer, sort of pure play digital analytics jobs than we used to. But at the same time, there are more people out there using digital analytics data and tools. Like I said before, that’s kinda of a big trend, but I think if you fast forward 5 or 10 years from now, I think the future digital analyst is going to be someone who is an expert on digital as a business, and is really good at answering questions with data and explaining it to executives. We still need the analyst to be the bridge between the business and the data. Yes, the business is becoming more data savvy, but that’s not an overnight thing. That’s a trend that’s gonna take a generation to come to fruition. That person’s gonna… The digital analyst of the future is a lot like the digital analyst of today. They need some level of analytics skills, some level of tech skills and they need to be experts in the business, and they need to be able to communicate to the business. They might work alongside data scientists in an analytics organization, but you don’t have to become a data scientist to work in analytics over the next 20 or 30 years.

21:31 MK: So, Corry, just on that. I’m just wondering, in the US, whether you’ve seen any evidence of organizations expecting more from their digital analyst? I guess with more rise and more positions in companies being data science-y, do you see any changes in the expectations of what people want out of their analysts?

21:54 CP: No doubt. I think that what you said before about your team is not unusual. When we work on digital analytics roles now, they’ll often have SQL as a requirement because they need people to not just get into their digital… Their Adobe tool stack, they want them to get into the SQL Data Warehouse with all the other data and be able to do some slightly more complex analysis. I think that some the… But that’s some of the jobs, that’s not of all the jobs. And part of this also has to do with where the company is on that analytics maturity curve. There’s still lots of companies out there that this is pretty new for. We’re still working with those companies who are just getting marketing analytics, digital analytics going or in the second, third, fourth inning of that, there’s still a big part of the market where that’s gonna be the case over the next decade.

22:57 TW: So, Moe, you’re asking the question about those roles, but you also said you feel like, at the iconic, that there’s kind of been a blurring and blending of the two different roles? Was that one where there was an ideal that, yes, there was a distribution of the analyst was gonna be the business savvy, the communication, the bridging, the requirements, and they’d be partnered with data science? What happened there? It seems like it’s almost a, “Oh, this was an ideal, but it hasn’t quite been achieved.” Moe, critique your company right know.

23:32 MK: I actually I…

23:33 S?: That’s not really the question.

[laughter]

23:35 MK: I think the opposite has happened, where we had a data scientist that worked with us, and now my current boss is a trained data scientist that I would describe as one of those weird unicorns that can do everything. And the environment has encouraged all of the analysts to learn more, do more. You kind of can’t rely on someone in the data team to get you the data that you need when you need it. So you sort of start hacking around to figure it out yourself, and then your skills just keep evolving. I have colleagues now who introduced themselves as data analysts rather than digital analysts, even though that’s been their professional career because really, they’re taking that digital analytics data and they’re melding it with all of the other data sets that we have. And I’m in that same boat, but I probably still wouldn’t call myself a data analyst. But we had this conversation last year at MeasureCamp, where one of the guys that I used to work with was like, “But Moe, you’re a data scientist.” And I’m like, “No, I’m not.” And he’s like, “We do the same thing.” And I’m like, “Still doesn’t make me a… ” I would never apply for a job that’s called a data scientist, but they would.

24:49 MH: I don’t disagree with that, Moe. I think, in the same way, I don’t think of myself as an analytics developer or technical person, even though I’ve done many implementations because there’s a higher level of skill that I see can be brought to bear by people who’d really have that. In the same way, I’m not ever gonna be a data scientist. Unless see Moe, comes through and actually does tutor me personally, I guess. And the other thing and I’m gonna steal something that I heard from Angela Bassa, she’s the Director of Data Science at iRobot, I heard in a talk she gave, she is like, “There’s no unicorns. Nobody has every little thing you’re looking for in a data scientist.” And so it’s kinda like, as you increase in your level of sophistication, we’ll all find our spots that we excel in.

25:40 CP: Yeah. I just wanted to chime in on the end of what Moe said there. I think that the trend of digital analysts becoming data analysts is much more significant than any trend of digital analysts becoming data scientists. Yes, it’s semantic, to some degree but I think that, essentially, that means that now, and more and more in the future, it doesn’t matter as much where the data came from and the designation of digital. Like I said before, it’s more about the business of digital than where the data came from. And data analyst is more likely what your title is going to be 10 years from now and data scientist.

26:31 MK: Corry, I just wanted to ask, on the side of, I guess, the managers, as these analysts are evolving and data scientists are coming more in the picture, are there challenges that you’re seeing when you’re putting managers in companies to be able actually lead these teams when… I mean, data scientist, probably not… Like I said, our team is hybrid. We have data scientists and analysts. And my last company was the same. Unless you’re one of those more technical data scientists… There are challenges there for managing teams when maybe you’re not as across what they’re doing now than you might have been in the past?

27:13 CP: Those are brutal roles for companies to fill at this point because you’re talking about a skill set that’s pretty new, and yet you need to find somebody who’s… Generally our clients are looking for people who have already managed multidisciplinary analytics teams across multiple data sources and different skillsets, quant tech, etcetera. So yeah, it’s a huge challenge. Again if you look at our job postings, and it’s not a plug, [laughter] the director of analytics, the director of data science, the director of statistical analysis roles, very tough to find. For one person… The other thing that’s hard about it is, just because you’re a good data analyst doesn’t mean that you’re going to be a great manager of data analysts. We’re talking about a different skill set. And what’s happened a lot of times is, companies need to show career growth to people in order to retain them, so they turn them into managers and they have them leading people, and sometimes it turns into a bit of a mess because they’re not skilled in that way.

28:31 MH: But Corry, you and I both recruit successfully out of that scenario. So, everybody, keep doing what you’re doing.

[laughter]

28:41 CP: Yeah, I mean, I’m not saying nobody kicks a goal but…

28:45 MH: It is, absolutely I agree with you. It’s a big problem.

28:47 CP: It’s a problem nonetheless.

28:49 TW: And I’ll say, Corry, you’re saying, not a plug. But I would, for anybody who’s interested in this topic, it is worth going to iqworkforce.force.com/careers or just go to IQ Workforce and look at their… Their listings are pretty interesting whether you’re looking or not, I mean it’s a long list of titles, and that kind of sparked this a little bit, that it does range from data scientist roles to directors, to manager type level positions.

29:19 CP: Yeah, what’s kinda cool about it also, and germane to this conversation is that, in that list of jobs, and thank you for that Tim, by the way. In that list of jobs, sort of what I was talking about before, we have clients who are over on that left side of the analytic maturity curve and we’ve got some web analytics manager jobs right up there for companies that are just catching onto that and trying to get their web analytics up to cruising altitude. And then we’ve also got Wayfair and Instacart doing some really cool, heavy duty data science stuff. So there’s sort of a full spectrum of our conversation in there.

29:58 MK: So Corry, I’ve got one question for the digital analyst that’s listening. If there is one skill that he or she should learn in the next few years to make sure they have a job in 10 years, what do you think that… What do you think that is? That skill? What do you think they can’t live without?

30:16 CP: I think that depends very much on the individual and what they’re good at and what they’re interested in. Because there are few different directions that they can go in. If they’re a digital analyst right now who’s more inclined to move to the business side and be a data-driven marketer, data-driven somebody on the business side, the answer is gonna be very different than if they’re a digital analyst who wants to continue their career in analytics. So let’s go with the latter. If you’re a digital analyst and you find that you love analytics and you want to progress your career in that direction, there’s really two different things that I would suggest. Number one, definitely learn SQL, because inevitably, if not in this role, then in the next, you’re going to be expected to pull your own data and work with data from different channels that are stored in some kind of data warehouse type of situation. And the other is… Oh my God, I forgot the other one. That’s a cliffhanger. Sorry.

31:22 TW: I feel like a digital analyst who says, “I really do gravitate towards the business and now I’ve gotten a very, very thorough grounding in digital, so I can go into digital marketing or e-commerce or marketing, because I’ve got that digital understanding and the data understanding, I’ll be very successful.” I always die a little bit inside ’cause I’m like, “No! ‘Cause we still have a supply and demand gap. Do you have a… Do you see a bigger challenge on the supply or the demand when it comes to analyst and/or data science roles?

31:53 CP: Yeah, I wouldn’t say we lost one. I’d say we got one on the inside.

31:57 TW: Got one on…

[chuckle]

31:58 CP: We got a… We got a mole in there.

32:01 TW: There you go.

32:02 CP: I think the more that that happens, the better it is for everybody. I’m sorry, what was the question?

[chuckle]

32:08 TW: Well, is it… Where for years we’ve said that, or I have said, that the digital analytics has been a great business to be in because the gap between supply and demand is so big that I just have to be just marginally above mediocre and I can continue to draw a paycheck, is that closing with all the degree programs, and more interest, and just the profession maturing? Is there more balance between supply and demand for the analytics and data science roles?

32:39 CP: I think that there’s… So what I’ll say about that Tim, is that digital analytics is still a great way to break into analytics. It’s a relatively low bar to entry. You can get yourself a junior analyst role on a team somewhere with a degree and a certificate and some training. And you can develop skills that you can use in a lot of different career paths. I would say that the demand for digital analyst is certainly not where what it used to be. It’s not the fever pitch where you would just die to get anybody who knew anything about it, but it’s still really strong. It’s still really hard to find good ones, people who… It’s easier to find people who know the tools and who have digital analyst on their profile. It’s still really hard to find people who can move the needle on the business and understand the question and the problem and can find the data and answer the question, solve the problem. That’s still a very tough job to fill.

33:51 TW: Well, I worry about those companies that are very early on the maturity curve, and people getting those junior analyst roles and basically getting trained with a very, very, very limited scope as to what the future should be. I’ve definitely seen companies that are still kind of defining what they need based on what digital analytics was. What the best digital analytics could be five or seven or eight years ago. I get nervous about people breaking into the field and not being… Not having the data science partner organization or not having the cross-data…

34:34 MH: Yeah, but I think that sometimes, the drum beat is too loud too. ‘Cause I had a conversation very recently with somebody on my team, and they’re fairly junior and they’re looking at their career progression and where they wanna go in their career and they’re pretty convinced that unless they become a data scientist with really strong technical skills, that’s it. Like there’s not gonna be good job opportunities in the future and I think that sent me into a lot of thinking. And so, I kind of categorized all the things that an analyst may do, and I’m gonna include data science and what an analyst can do, ’cause analyst, in the big term analyst, not entry level digital analyst, but there’s… People spend their entire careers in just one area of data collection and implementation, tools and systems, reporting a data visualization, analysis and insight generation, optimization, business strategy and governance, and leadership, and all those different things. All those areas could be a career. And so, it’s difficult, I think, for people to get a full picture sometimes.

35:43 TW: But they can be a career, but if you’re in a, and maybe we’ll put small and maybe medium sized businesses aside for this, but if you’re having that career inside a larger enterprise organization that doesn’t have the data science capability; and maybe it should be self-selecting. That the, winning on analytics, competing on analytics, back to Tom Davenport, that those companies will ultimately have to figure it out. So I think… I wanna… Between the role in the organization, that I think, where I’m starting to come down is, yes, there’s plenty of a future without going into the quant and modeling and the technical side. But I worry about being in organization where no one, there’s no organization that has that capability and that capacity that should becoming more emergent and the role the analyst shifts and that they should have that as a partner that they’re working with.

36:45 MK: As an analyst though, you definitely do feel the pressure. You feel that you have to keep up with this stuff that you kind of can’t. And I’m not speaking for all analysts, I’m speaking for myself. But I definitely don’t feel like I can go into a very niche area. I feel like I need to keep my skills pretty broad and keep learning. And yeah, I’m sure there’s lots of analysts that would probably share that sentiment.

37:10 MH: Yeah, I certainly felt that way earlier in my career and now that there’s no hope for me, well I’m just gonna cruise well But, totally felt this similar way. And it’s hard because you know eventually you can’t be as good at other areas. Like I’m just not ever gonna be a really strong developer or someone with a strong technical skill set. And I’ve… You know, I’ve come to peace with that but if someone’s new to the industry, or whatever, I don’t want that to force them out. I want them to find where they fit and that’s kind of the… The challenge I think for a lot of the people kind of coming in and trying to find their way today.

37:49 TW: Well so, Cory, where… That is making me… I gravitate towards and I think, everybody here gravitates towards learning and pushing and trying to understand the broader. So there’s a generalist component even though we hopefully are finding the specialization and the depth. When you’re talking to candidates, do you have little pieces that… I think anybody who’s going into… Is gonna work with digital data, doing any form of analytics, who says, “JavaScript terrifies me,” should go find another profession. And I would think the same thing, you know, “Looking at a SQL query terrifies me and I want to avoid it.” To me that’s a red flag. But that’s very different from saying, “I’m gonna go and learn SQL in depth” or “I’m gonna learn JavaScript in depth.” Is that a fair way to view anyone in the broadest brush, if they’re not at least interested in learning some basics, so that they can communicate and understand the mechanics behind things…

38:57 CP: Yeah.

38:58 TW: That maybe they… Okay.

39:00 CP: Yeah. I think that’s very fair. If you’re an analyst and you don’t particularly feel like learning SQL because you think it’s too technical for you or you don’t feel like, you know, or JavaScript scares you, then you should move to the business side, and you should let analysts do their work. You should listen to them and understand enough about what they do and respect the power of the data and go do that. You can go have a great career doing that. On the other hand you… You’d like to roll up your sleeves and get into the data and figure these things out and you’re curious about where the data comes from and how… All that stuff. And you wanna learn and expand, and… I think that there is a fantastic career for you in, what Moe said before, in data analytics. Not necessarily in data science.

39:58 MH: Not bad.

39:58 TW: Cool. That’s good. That’s what I thought. I’ve learned something from you over the years.

[laughter]

40:04 MK: I just… Tim, I was… Tim, I was just gonna say though, it’s okay to be scared as long as you’re still curious. And I think that’s kind of a differing point. You know, I’ve been terrified of some of the things that I now can do with quite a bit of ease. But as long as you have that curiosity and that desire to learn it, or learn something, I think it’s okay. But it’s okay to look at JavaScript and be like, “This scares me,” and then get your head around to it. It’s more when you’re just like, “I really don’t wanna do this,” that maybe you go to the business side.

40:38 CP: Yeah, I agree with that.

40:39 TW: I feel like I’d… That’s the kind of… I got to JavaScript to where I dove in and dabbled enough and then realized it was kinda the same as DBA and I’m like, “I don’t wanna do this, but I’ve tinkered with it enough than I’m comfortable.” I could’ve bailed on R a year ago and I still would’ve gotten… It would’ve been valuable to have done the exploration. The fact is I gravitated more towards it. So, yeah, I think that’s… Yeah. If you look at it and take the first glance and say, “Oh, hell no,” then I think that’s a problem.

41:07 CP: Yeah.

41:09 TW: Or to Corry’s point, that is a, “You know what, go be an informed marketer.”

41:11 MH: Yeah. And go back to listen to episode one of this podcast and if it doesn’t resonate with you, and this show won’t, and you probably haven’t listened to all throughout.

41:22 TW: Well, it may resonate too ’cause the audio is… Or echo, or be staticky or, you know…

41:25 MH: Reverberate?

[laughter]

41:27 TW: Yeah, reverberate. Just ’cause it’s…

41:30 MH: That’s sort of the table stakes, right? Is the curiosity, the desire to innovate. And as you shape your career, I guess one of the things that’s actually pretty awesome about analytics and data science generally, is that as analysts, we still drive a lot of where our careers are gonna go and we’re not being dictated to by our jobs or industries as much as probably quite a few other ones. And that’s kind of a really pleasant place to be as a professional.

41:57 CP: Yeah, I think that’s true, Michael. I think it’s also important for people to be proactive about that. I think that a lot of people will learn what their job makes available to them, and then if you fast forward five or 10 years later, will regret not having gone in a different direction. I think it’s your job to… It’s kind of a cliche, but be the CEO of your career and figure out what’s going on, see the big picture, know yourself and what you’re good at, and what you enjoy, and have a plan. Know where you wanna be in five years. Obviously, you can deviate, based on opportunities or change your mind, but you should have a plan, and a learning plan, and move yourself in that direction proactively.

42:48 MH: Yeah, no, I love that. Well we’ve gotten to the point where we’re gonna have to wrap up ’cause we’re running out of time, but this conversation is excellent, and hopefully it’s not only helpful to the new analyst out there who’s considering where their career’s gonna go, but even to those of us who’ve been around a long time who have built up a big set of digital analytics skills, and they’re wondering, “Will these skills serve me well in the future?” Anyways, Corry, thank you so much for coming on the show. One of the things we like to do is, we like to go around the horn and share what we call a Last Call. It could be anything we find interesting, or something that’s going on in the near future that we’ve seen recently. Corry, you’re our guest. Do you have a Last Call you wanna share?

43:33 CP: Yes. So I’m actually gonna go back to the… It’s literally the five year anniversary of Tom Davenport’s article in HBR, Data Scientist: The Sexiest Job of the 21st Century. First of all, can’t believe it’s five years. It’s rude that time has moved that quickly. But I think if you go back and read it again, it’s held up extremely well. The main assertions have really played out, including all of the aspects that he focuses on on the talent side. Things like, he mentions there are important breakthroughs in tools and technologies. At least, as important are the people with the skill set to put them to use, and on this front, demand has raced ahead of supply. Could have written that today. That’s five years back. And so I’ll spare you my whole long list of quotes here, but I think that part of what I wanted to say about this is… Kinda goes back to what we were talking about before, that this is five years that’s gone in a blip, and I think a lot of our clients still have data science up on the whiteboard, haven’t even really gotten to real data science. So I think that what we’re looking at here is again a generational evolution. It’s not something that’s going to happen in mainstream business in the next five years.

45:11 TW: Nice.

45:11 MH: Very nice. Who’s next? Moe, you got one?

45:15 MK: Okay, I’ve got one, but I have to give a disclaimer that this will be the last time… I promise, listeners, it will be the last time that I talk about an R package. But I’ve spent the whole weekend hanging out with this R package, and it’s actually got me really excited, and had some big wins with it. And if anyone wants to tweet me and correct my pronunciation ’cause I can never seem to pronounce it right, but it’s a package called RPostgreSQL and it literally has changed my life. Because I keep dabbling in SQL, in R, and trying to pull things together, and this was the way that I could do both. So, I’m pulling in some big query data, and then I’m getting some data from our Redshift server using this SQL package, and if you haven’t had a try, it actually… Anyway, I had a very exciting weekend, as you can tell.

46:09 TW: Clearly, you need to… Once you’ve decided to drift over to the more business side of things then we’re gonna drop you from the podcast.

[laughter]

46:18 MH: Yeah, R packages Last Call’s always welcome here. Hopefully…

[laughter]

46:24 MK: I feel like I do it too much.

46:25 MH: Hopefully you stole Tim’s thunder.

46:29 MK: I was hoping I would steal his thunder actually.

46:29 MH: Tim, what’s your last call?

46:32 TW: Okay, I had one that was just totally whimsical, and then I saw another one that was… I’m gonna have to do two quick ones. One is from, courtesy of Mr. Simon Rumble, Moe’s buddy. But he had posted an article from hackernoon.com that is the AI Hierarchy of Needs, which is a short little article, but it’s nice. It’s reminiscent. You can bore your eyes and think back to digital analytics 10 years ago, where it’s basically laying out a little pyramid. It looks like a consultant saying, “Look, there are companies that are diving into AI wildly prematurely. They don’t have the data. They haven’t got collection nailed down. They don’t have storage. They haven’t got the blocking and tackling done.” So it’s a great little read of, “Hey, it’s somebody who’s just chasing what’s sexy, but they are woefully ill-equipped to actually act on it.”

47:29 TW: So it actually seemed relevant to the topic of the podcast. But then I also feel like, I realized, just terribly, I don’t think I’ve ever given a shoutout to my favorite analytics song, but every time there’s one of our group listen Turntable.fm or whatever the one is we’re using now, I’m always the guy who comes in and puts on Todd Snider’s Statistician’s Blues. So if you’re at all into humorous music, it is definitely poking fun at statistics, but it’s a good little listen. So I feel like that song doesn’t get enough play in the industry.

48:04 MH: Very nice.

48:05 TW: There’s my twofer.

48:06 MH: And my Last Call is actually… I am speaking today at Auburn University, to their Business Analytics Department about careers in analytics and I’m stealing all the content from this episode for that talk.

[laughter]

48:23 MH: So, great job. You played right into my diabolical hands. [laughter] No, but it’s actually funny… Well, it’s just… Economies of scale. It’s just always be… But hopefully in doing this, we can leverage some of that. And actually, if you are a student and you’re thinking of a university, Auburn University seems like a pretty quality one and I’m really excited to get a better understanding of their business analytics program as I go talk to them today.

48:55 MH: Alright. If you’ve been listening and you’re thinking, “Man! That is exactly what my career’s going through,” or, ” I need more information about what I just heard about what Corry just said”, please don’t hesitate to reach out to us. We’d love to hear from you. Both through our Facebook page, our Twitter page and through the Measure Slack. And Corry and his team would probably be happy to talk to you at IQ Workforce. If you’re in the middle of a job search or you’re thinking about your job opportunities, I’ve referred numerous people to Corry throughout my career and I’ve always been very happy with how he’s been able to help and identify great opportunities for those people. So personally…

49:37 TW: Likewise. And to be clear, if you’re looking to hire or looking… Yeah, if you’re looking to find somebody or looking to be a found person, he obviously has to cover both of those.

49:46 MH: Exactly.

49:48 TW: He’s too nice to say it himself, but…

49:49 MH: Yeah, yeah. He doesn’t say so. I say so.

49:51 CP: It doesn’t mean I don’t appreciate it though.

49:53 MH: There you go.

[laughter]

49:54 MH: Anyways, thank you so much, Corry. We really appreciate having you on the show.

49:58 CP: Thanks for having me, guys.

50:00 MH: For my two co-hosts, Moe and Tim, keep learning data science out there, I guess.

50:08 Announcer: Thanks for listening and don’t forget to join the conversation on Facebook, Twitter or Measure Slack group. We welcome your comments and questions. Visit us on the web at analyticshour.io, Facebook.com/analyticshour or at Analytics Hour on Twitter.

50:28 S?: So smart guys want to fit in, so they’ve made up a term called analytic. Analytics don’t work.

50:37 TW: Yeah. While you were struggling with Skype and Moe was running late, Corry’s video wasn’t working. So it was… It was kinda the perfect storm of… No one felt like they were…

50:47 MH: Alright. But now, we are ready.

50:51 MK: Oh my God! That is the funniest thing I’ve ever seen.

[laughter]

50:57 MH: That’s one reaction.

50:58 TW: That’s…

51:00 MH: Yeah.

[laughter]

51:03 MH: You know, if you’re concerned, Corry has tons of job opportunities.

[laughter]

51:09 MK: Actually, I was…

51:11 TW: What are you drinking?

51:13 MK: I’m drinking this… Oh! People in Australia don’t drink coffee out of this. This is water. When you’re drinking coffee, the cup’s like yay big.

51:20 TW: Not a jumbo diluted Americano.

51:22 MK: In fairness, my sister probably drinks coffee out of this. So you know…

51:27 CP: Could be programming a satellite or something.

51:29 MH: Yeah, yeah. We’re just redirecting the missiles now. You just have to connect them up with AWS or Hive or PIG or other words I’ve learned.

51:41 MK: What?

[laughter]

51:48 MH: Oh boy! Hey now, at least Moe is doing it to you. You’re in fine company, let me tell ya.

[laughter]

51:56 MK: Although, Tim, check out my T-shirt today. Ugh! I’m trying really hard.

52:02 TW: Oh, good!

52:03 MH: Oh! [52:03] ____ Nice.

52:04 MK: Oh wow!

52:05 TW: Look at that.

52:06 MH: Oh my gosh!

52:07 MK: Oh, we’re nerding out hard.

52:09 MH: Oh my gosh.

52:09 MK: We’re nerding out hard.

52:12 CP: I got distracted by Michael pouring that a little bit… We’re gonna have to edit that out.

[laughter]

52:20 MH: Are you thirsty?

52:21 TW: Corry’s drooling on his keyboard, you can see that he’s trying to save the electronics.

52:24 MH: This is a two-drink kinda show.

52:28 CP: That looks delicious.

[laughter]

52:33 S?: Rock flag and career advancement.

[music]

Podcast: Download | Embed

Subscribe: RSS