Today I’m chatting with Kyle Winterbottom, who is the owner of Orbition Group and an advisor/recruiter for companies who are hiring top talent in the data industry. Kyle and I discuss whether the concept of data products has meaningful value to companies, or if it’s in a hype cycle of sorts. Kyle then shares his views on what sets the idea of data products apart from other trends, the well-paid opportunities he sees opening up for product leaders in the data industry, and why he feels being able to increase user adoption and quantify the business impact of your work is also relevant in a candidate’s ability to negotiate higher pay. Kyle and I also discuss the strange tendency for companies to mistakenly prioritize technical skills for these roles, the overall job market for data product leaders, average compensation numbers, and what companies can do to attract this talent.

Highlights/ Skip to:

• Kyle introduces himself and his company, Orbition Group (01:02)
• Why Brian invited Kyle on the show to discuss the recruitment of technical talent for data & analytics teams (02:00)
• Kyle shares what’s causing companies to build out data product teams (04:49)
• The reason why viewing data as a product seems to be driving better adoption in Kyle’s view (07:22)
• Does Kyle feel that the concept of data products is mostly hype or meaningful? (11:26)
• The different levels of maturity Kyle sees in organizations that are approaching him for help hiring data product talent, and how soft skills are often overlooked (15:37)
• Kyle’s views on who is successfully landing data product manager roles and how that’s starting to change (23:20)
• What Kyle’s observations are on the salary bands for data product manager roles and the type of money people can make in this space (25:41)
• Brian and Kyle discuss how the skills of DPMs can help these leaders improve earning potential (30:30)
• Kyle’s observations and advice to companies seeking to improve the data product talent they attract (38:12)
• How listeners can learn more about Kyle and Orbition Group (47:55)

Quotes from Today’s Episode

• “I think data products, obviously, there’s starting to get a bit of hype around it, which I’ve got no doubt will start to lead organizations to look down that route, just because they see and hear about other organizations doing it. ... [but] what it’s helping organizations to do is to drive adoption.” — Kyle Winterbottom (05:45)

• “I think we’re at a point now where it’s becoming more and more clear, day by day, week by week, the there’s more to [the data industry] than just the building of stuff.” – Kyle Winterbottom (12:56)

• “The whole soft skills piece is becoming absolutely integral because it’s become—you know, it’s night and day now, between the people that are really investing in themselves in that area and how quickly they’re progressing in their career because of that. But yeah, most organizations don’t even think about that.” – Kyle Winterbottom (18:49)

• “I think nine times out of ten, most businesses overestimate the importance of the technical stuff practically in every role. … Even data analysts, data scientists, all they’re bothered about is the tech stack that they’ve used, [but] there’s a lot more to it than just the tech that they use.” – Kyle Winterbottom (22:56)

• “There’s probably a big opportunity for really good product people to move into the data space because it’s going to be well paid with lots of opportunity. [It’s] quite an interesting space.” – Kyle Winterbottom (24:05)

• “As soon as you get to a point where if you can help to drive adoption and then you can quantify the commercial benefit of that adoption to the organization, that probably puts you up near the top in terms of percentile of being important to a data organization.” – Kyle Winterbottom (32:21)

• “We’re forever talking in our industry about the importance of storytelling. Yeah, I’ve never seen a business once tell a good story about how good it is to work for them, specifically in regards to their data analytics team and telling a story about that.” – Kyle Winterbottom (39:37)

Links

• Kyle’s LinkedIn: https://www.linkedin.com/in/kylewinterbottom/
• Orbition Group: https://www.orbitiongroup.com

Transcript

Brian: Welcome back to Experiencing Data. This is Brian T. O’Neill. Today I have Kyle Winterbottom, the CEO of Orbition Group. I’ve been calling this Orbition Group for so long; I was very embarrassed to miss the I, so I guess my—I’m having a refraction appointment coming up soon to get my eyes checked. So, I’m glad I know this is the Orbition group. I’m not going to say the wrong one anymore because I don’t want to further that mistake. But Kyle, welcome to the show [laugh].

Kyle: Thank you very much for having me. Is a very interesting story about the name of the company, not in the formulation of it, but just in the fact that everybody thinks it’s all its own group. And part of when I set the business up, I wanted to capture a word that was—one word, ironically easy to pronounce that people would remember, and everyone gets it wrong. So, probably need to have a rethink on [laugh] on the name.

Brian: [laugh]. The first thing I just thought of right now and not to turn this into a design audit, but I was curious about your logo and whether or not there’s something in the typeface about the tracking between the letter the spacing between the letters that maybe people are missing the eye, but I’m looking at it right now, I’m like, no, it says Orbition [laugh] very clear. That’s funny, everybody’s saying that. Anyhow, that’s totally cool. The more important thing is, why are you here and why did I have you on?

So, you’re a pretty unique guest here in terms of, this is episode 118-ish or something like that, it will probably be, and you’re coming out of the recruiting area that’s finding technical talent for data and analytics teams, which is unusual. Most of the time, we have data product leaders come on this show. But that’s kind of what I wanted to talk about. There’s an interesting vector here, which is like, let’s talk about the finding of data product leadership talent, and is this a thing in the market? What’s going on with it, et cetera?

So, that’s why I wanted to have you on here. And so, that’s what we’re going to dig into [laugh].

Kyle: Yep. Looking forward to it.

Brian: Yeah. So, when we had our kind of pre-chat, you had mentioned, there actually is a fair amount of stuff going on here I feel like in the last, I don’t know, 6 to 12 months, there’s been a lot more talk about data products in general. Which, it’s not new on this show, but there seems to be more activity here. So, I’m kind of wondering, can you just give a lay of the landscape in terms of, what are companies looking for in the space of data product management? Do they know what it is? Is there a consistent meaning for what a person is that’s doing this type of work? Can you kind of just give us a landscape look here?

Kyle: Yeah. Well, I guess to answer your first question, do they know what they’re looking for, I don’t think many organizations across the whole data analytics spectrum actually know what they are looking for. But I guess in terms of the product piece, as you very rightly said, you know, it’s obviously, it’s been around for a very long time, but in the data analytics world, it’s the last 12 months or so is kind of really seen the prominence. So, we’re now seeing, you know, a lot of organizations starting to build out data product teams, and you know, that’s translating into seen a lot of jobs being formulated around, you know, the data product owner, the data product manager, you know, having product teams embedded into central data analytic teams or fragmented or hub-and-spoke models. So, there’s definite demand there.

And I think, you know, I kind of started to see organizations getting more interested and excited about the kind of concept of treating data as a product or, like, a product, or however, the terminology needs to be framed and at the back-end of 2021, I kind of put my hat on the peg and said, “I think this is going to be the next big thing as far as jobs go.” And I think, you know, what we’re starting to see now in the 12, 18 months later is that, you know, that is starting to come to fruition. So, I think it’s—you know, I think I might have mentioned this to you when we were offline, but I had someone recently say to me, “I’m in the data product space and I want to become a data scientist and what should I do?” And I was like, “Well, here’s the answer: don’t.” [laugh]

Brian: [laugh].

Kyle: Because, you know, data science will have its time again, but it’s kind of on pause, if you want to call it that, and data products is where the next big wave is going to be. So, you know, I think you’re in exactly the right space. So yeah, a lot of talk about it. Organizations starting to move to it, build out, you know, data product teams, which is really exciting, but equally fascinating.

Brian: Yes. So, explain this to me, I guess, from your take. The desire to create these data product teams must be a reaction to something that’s going on in the business. “We’re having problem X, and so therefore, Kyle, we need to go find data product teams to fix it.” What is the X that they’re trying to fix? What do they think they’re going to get with that?

Kyle: That’s a great question. That’s a great question, and to be honest with you, I’m not too sure many organizations know the answer to that themselves. I mean, I think, you know, generally speaking, what I’ve seen happen more times than I care to imagine over the years is that most organizations get sucked into the hype around certain topics, right? So, data science five years ago, I must have sat in countless corporate offices saying, “You don’t need to hire data scientists. I promise you, you’re going to regret this.” Lo and behold.

But I think organizations get sucked into that trap. Now, I think data products, obviously, there’s starting to get a bit of hype around it, which I’ve got no doubt will start to lead organizations to look down that route, just because they see and hear about other organizations doing it, and things like that. I think what I’ve seen has been the thing that is helping, what it’s helping organizations to do is to drive adoption. And I think that’s at the moment in time, is kind of what it’s really being used to combat now. You know, I’ve been quite vocal about what do product professionals think about the data product landscape currently because it’s nowhere near as robust as, you know, actual, genuine products.

I think what I find and what I’m seeing is that it’s almost like the role itself at the minute is being used as almost like a Data BA, right, you know, where they’re kind of sitting in that spoke as the data person who knows the technicalities, but really, their job is to gather requirements, make sure this stuff is getting used, et cetera, et cetera, the product piece around, you know, the service offering around that comes with a product, how often it’s updated and maintained and new features. I don’t think we’re anywhere near that yet. In fact, I think in many instances, you know, it’s the same work that’s been done before, we’re just kind of now calling it a product. Which is interesting, but it seems to be working, which obviously then leads more businesses to do it. So.

Brian: Even if it’s anecdotal, you said, it’s working. Are you saying you touch base with people after they place a candidate and you’re like, “Did it work?” Like, “Are you getting the value?” And they’re saying yes? Or like, what’s your evidence that that’s having an impact on adop—and when you say that, do you mean adoption, user adoption of the data products is increasing as a result of hiring these people?

Kyle: Yeah. Yeah. So, I think that, yeah, the result is that—and this is, you know, normally the perception of the data leader, right, so there’s probably something in there, they’re probably also not going to say [laugh], “No, this is terrible and not working,” but equally, yeah, I think, you know, most data leaders feel that by approaching data as a product, it’s allowing the business users to kind of get their head around it more that there’s something here that’s formulated, that’s there for you to use, that’s going to make you more efficient at your job, as opposed to hey, we found this data and here’s some insight and, you know, make do with that what you will. So the, kind of, putting the guardrails around it as something tangible to use seems to be helping business users to just implement and execute, which you know, as I said, I don’t think there’s much more to it than that. I’m sure I’m doing a lot of organizations a grave disservice at this point in time because there probably are many organizations that are doing data products stuff really well.

But majority speaking, you know, when we speak to data leaders and obviously people that we interview on our podcast, and through the day job and in our events, et cetera, you know, the product piece really seems to be that kind of initial engagement piece that gets business users excited to use it, basically.

Brian: Got it. I’m actually glad that the people doing this hiring, the executives and leadership teams are connecting the dots between adoption and the work of product. And, like, this is an area I’ve been focused on for a long time and the interesting thing with data products, to me, at least in my world view of what these are, is it’s like, the word product in there is actually more of a verb than a noun. It’s a method of designing solutions that are human-centered with this idea that you’ll get the adoption if you can increase the satisfaction part. So, product is, like, a way of working for people that are native to product management and design. It’s really about the method; it’s not just about the output of labor, you know [laugh]?

And so, I think we get hung up a lot in the making of stuff, the making of the outputs, but really product management and design as a field is more about the verb and the process to me the way of working. So, the analy—you know, the requirements gathering as part of it, I think of that more as the problem finding instead of requirements gathering because requirements often sounds like, go get what everyone’s asking for, and then serve it to them. And that’s actually not what we need product people to be doing. Product people are actually out to find the hidden problems that are under the surface that are not articulated. Because you don’t want to give them a machine learning model just because they ask for it, you need to make sure that’s the right antidote for the problem they have, and a lot of times they don’t know how to express the problem except using buzzwords or they’re trying to help you by telling you what tech to use because they think it’ll help you know what to go make. And it’s a trap [laugh].

There’s a great book on this called The Mom Test, which I recommend for product people. It’s really, the book was written more for entrepreneurs and people in startup spaces, but the main idea of the Mom Test is, one of the big things that I like about it is that—and this ultimately becomes a product person’s job once a startup gets going, and the founders aren’t quite as attached to the product itself—the product person, you one hundred percent own the solution, but the customer, the user one hundred percent owns the problem. You don’t get to tell them what their problems are, but they don’t get to tell you what to make. And you don’t cross those boundaries. And so anyhow, I’m going off on my tangent here a little bit, but I’m actually happy to hear that the increasing of adoption here is what they’re associating this with. That’s something I’ve been talking about for a long time, other people have been talking about as well, so I think that’s cool.

Kyle: Yeah.

Brian: That said, is this a hype? Are we—I mean, feel free to react to what I just said to, but like, I’m always worried that we’re creating a hive—like, this whole big data bullshit that was out there for so long. It’s undefinable; it doesn’t mean anything; it’s not helpful. I don’t want to be contributing to that, and sometimes I worry talking about this for so many years that it’s creating fake baloney that’s not helping anybody. Does this feel like hype to you? Like, it’s just going to—and you’re not going to offend me if you think so. Is it just hype-y or is it meaningful?

Kyle: Well, I think the data analytics industry in general has got an awful habit of falling into the traps of whatever the next big thing is—

Brian: Yeah. Mm-hm.

Kyle: —right, as the silver bullet. So, I think there’s always a risk that it is a hype. I think that the slight difference, in my view, at this point in time is that there seems to be some tangible wins off the back of this, right? You know, like, if you think about the big data landscape, you know, that’s fizzled out, no one’s you know, no one’s talking about big data anymore, really right, but ten years ago, it was—because it was a massive, massive tech play and you need to build this big data tech stack because big data is the next thing. And I think across the data analytics industry, what we are starting to slowly but surely realize is that, you know, all of the technology is just there to serve a purpose.

And you know, I talk about this quite a lot in the, you know, the content that we produce, but I think we’re at a point now where it’s becoming more and more clear, day by day, week by week, the there’s more to this than just the building of stuff. And the whole methodologies around that plays an important part, but you know, we’re getting to the point now where I think businesses understand there’s a lot of the soft stuff that’s really important. And I think that’s probably where the slight differences here is that, you know, the product piece is starting to bridge that gap a little bit. But I think to kind of go back to your earlier point, I think, the data analytics industry because of its, you know, immaturity in comparison to other functions within a business, what I see, literally on a daily basis, is that different terms, different buzzwords, different job titles, they all mean different things to different organizations, right?

So, we held a roundtable event, which we do every single quarter, and this was probably the middle of last year, so we’re probably approaching 12 months ago now, and you know that one of the themes of the topic was data products versus data as a product and is there any difference? And, you know, you’ve got 20 fully-grown adults, data leaders, you know, director, VP, CDOs sat in the room, almost literally arguing about the definition of [laugh] data product or data as a product, and what’s the difference and what constitutes what? And, you know, they, literally someone saying, “Well no, I don’t think that’s a data product; that’s data as a product.” And I’m kind of sat there thinking, you know, at this point in time, does it really matter if it’s serving the purpose of what we’re trying to get it to do? So, [unintelligible 00:14:32], I think there’s more to this than it’d be in another hype cycle of something else.

It feels like… it feels like there’s been success quickly with this and people are starting to see some tangible results. But again, you know, I think the important thing to probably reiterate is that data products to one organization probably means something different to what it means in another organization, especially when it comes to the role itself. Because as I’ve said, sometimes it’s really a Data BA type of role that’s badged up as a product owner. Often, then it can be, you know, the person that sits in the middle for more of a communication standpoint, right, so the, “translator,” you know, in quotation marks. Of the times it’s actually, we are building physical products that are, you know, beyond just the dashboard that we would have created six months ago when we weren’t calling it a product. So, you know, so I think there’s a lot of organizational context to this, but it feels like there’s more to it. And there’s been some wins quickly, which has allowed organizations to kind of get behind this might be a route forward.

Brian: Got it. Got it. Can you tell me a little bit about how, when someone comes to you looking to hire talent, are they expressing the skill gap that they need or they coming in the door saying, “I need a data product something?” Or are they saying, “I got to get some people in here to help me increase adoption of our stuff?” And then you’re like, “Well, there’s this data product role is kind of owning that.” Like, which way does it come in?

Kyle: Both ways, to be honest with you. It really depends on the organization and obviously there are levels of maturity around that. So, you know, one part of our business is the kind of talent advisory piece, which is kind of what you’re describing there. So, that helps us to understand, you know, actually, what you’re trying to achieve as an end goal, how you piece all of that together with what talent you’re going to need at what points in the journey. What’s the actual purpose of that role?

Okay, well, that role is this. This is what it’s called in the market, these different titles, this is what it costs, et cetera. And then obviously, the second part is, go and executing on that and putting that in. So yeah, we know, we’ll get organizations come to us saying, “We need a data product owner.” And then other times, it’s, you know, we’re kind of looking for someone that can sit in the central data team, but is kind of going out to the different domains, but there needs to be strong enough technically that they know how to build some of this stuff themselves.

But really, what we want them to do is engaging with the user and understanding how the user is going to use this and start to help the data team to build this. And then you can go, “Okay well, this is kind of what’s coming up and this is the type of stuff that you need to be thinking about.” So yeah, often comes down to the maturity of the organization and, you know, how well-versed they are in understanding what they are looking for. So yeah.

Brian: And I’m curious, after all that’s said and done, do you do them get a list of 25 technical skill bullets that they have to have mostly, and that’s it?

Kyle: Right [crosstalk 00:17:24]

Brian: Because I find those people are—

Kyle: [crosstalk 00:17:26].

Brian: —yeah. It’s just like, well, you want all that which requires, by definition, non-technical skill is the type of skill needed to do this type of work with. This includes research and problem finding and design and all these kinds of human-factors type skills, as well as knowing the business and all this other stuff. So Python, Jupyter Notebooks, data engineering, pipeline building, Tableau, Picker, whatever, that’s not going to help you with those things. But like, so much of the time, when I—I mean, I don’t look at job descriptions very frequently, but when I do, I’m always like, “Well, good luck finding that person.” And you’re just going to end up hiring another technical resource that’s not going to help you with that. Like, I don’t know, is that—

Kyle: Honestly, that’s nine times out of ten, across every single job in the data analytics landscape, even at the data leadership level, right? You know, we see all the time, we’re looking for a chief data officer to own data analytics and we want them to get commercial value and design and deliver and execute the strategy, right? And then you look at the requirements of job and it’s a Python, GCP, Kubernetes. And I’m kind of like… there’s something not quite adding up here, you know? But, I mean, that’s just where we’re at in the industry.

I mean, I bang on about this all the time. And I think, you know, even I posted this morning about this very topic in that, you know, the whole soft skills piece is becoming absolutely integral because it’s become—you know, it’s night and day now, between the people that are really investing in themselves in that area and how quickly they’re progressing in their career because of that. But yeah, you know, most organizations, they don’t even think about that, you know? If I had a dollar, for every time I’d heard this person is great technically, but he really struggles with XXX, you know, XXX meaning insert any non-technical skill into that component. And I sit there and say, “Okay, well, let’s look at the job description. Let’s look at the job advert. Okay well, it doesn’t mention anything about any of those skills that you’ve just said they’re missing. And did you assess them in the interviews for having those skills?” Like, you know, “No.” It’s like, “How good are you at Python? How good are you at this?”

So, you know, there’s, there is a big disconnect there that, you know, I think without kind of delving into the kind of, you know, troubles of most organizations’ recruitment processes and practices that often stems from the disconnect between the people that are the hiring manager for that job and the knowledge transfer to the people who are actually responsible for engaging in attracting talent. So yeah, you know, I think it’s the same across every job, to be honest with you.

Brian: I’m curious then, how are any of these companies then screening for the required skills in this space if it is a new skill? I mean, assuming we could get past the technical part of the interview. I mean, God forbid, they’re actually focusing on that in a data product management role. I mean, it’s yes, you do need to know how the sausage is made, but your job is not to be in the weeds, writing code. It’s just to understand the capabilities and all of that and to be able to have an educated conversation when necessary. But if you’re spending all your time doing that, like, you’re not doing the role. So, I’m kind of curious, how are they screening for this? Like, do they go out and just kind of hire, like, a seasoned PM, like, out of the software space and then—or like, I don’t know, like [laugh], do you have a sense of how they’re doing that?

Kyle: So, I think, obviously, you know, I can only speak on behalf of the organizations that I speak to and spoken to and we work on behalf of. I mean, a lot of the time, it’s advice, right? It’s these the things that are actually going to be important to you for this role. And I think often, as happens very frequently, but you know, most organizations start out with, “This is what we think the job is. Here’s what we think is important.”

They go to market for that skill set, they bring that skill set in and do you know, five, six interviews and then quickly realize this person’s probably not got enough of this, or they’ve got too much of that and not enough of this. So, the process itself often starts to rewrite the narrative a little bit for organizations, but you know, most of them do start out first on that technical track. I think what has been interesting is that the data product space, we’ve tended to see people who are already working in the data team somewhere or in some kind of data capacity, so they might have technical grounding, but they’re probably not, you know, not the Python whizzes of this world. And, you know, we’re starting to see people that maybe want to move into data analytics as a career, you know, so moving out of a different domain where they’ve got 10, 15 years experience in something else where they have that kind of commercial mindset and skill set, but they’ve been I don’t know, you know, a frequent user of data or whatever the case may be, they might have some technical grounding as an example, right?

And I think it started out as almost a re-skilling of people into those types of more forward-facing roles where they understood that this person necessarily didn’t need to be so technical, that’s then started to build up the data p—you know, organizations calling that data product so that then when they go to market, they’ve almost got a benchmark of okay, really, the important things are this. But again, that’s based on maturity of businesses having gone on that journey themselves and kind of come to that conclusion, often. So, you know, I think nine times out of ten, most businesses overestimate the importance of the technical stuff practically in every role, right, you know, maybe outside of the engineering space, for example, where it’s practically all technical stuff, right? But yeah, even data analysts, data scientists, like, all they’re bothered about is the tech stack that they’ve used. And it’s like, well even those roles, there’s a lot more to it than just the tech that they use.

Brian: Right.

Kyle: So yeah.

Brian: And would you say a lot of the people getting placed in this then are not coming out of a product management role, say in the software space, but rather they’re coming out of some other role and making a career leap into this new job title? Is that more frequently happening?

Kyle: I think that’s what I’ve, that’s what I’ve seen, but I do think that’s starting to change a little bit. I think as the conversation starts to unfold more, that actually, you know, what we’re calling data products here probably wouldn’t pass as a product in many [laugh] in many other instances. I think the actual product people are now starting to get sucked into that conversation and obviously, data analytics is, you know, on a big upward trajectory, has been for a few years and probably will be for the next 10, 15 years, right? So, there’s probably a big opportunity for really good product people to move into the data space because it’s going to be well paid, lots of opportunity, quite interesting space. But yeah, I think the starting point for a lot of businesses has been, you know, maybe moving some of the people that are better getting out into the business speaking to people, influencing, you know, all of that type of stuff, getting them—and a lot of them might have come from a central data team, right, but they might be not the best technically, so they think, well, these this person’s got, you know, the forward-facing skills that we need to be the, kind of, liaison and go-between between our central data team and our business users.

And that’s what’s the—you know, that’s how it kind of started to get bandaged up as data product because it was then, I don’t want to say it was a marketing ploy, but it was, you know—and probably wasn’t even intentional, but it kind of feels that we got to that point, almost by accident, right? You know, like, there’s this… we’ll call you data product people because you’re taking our products to our users who are our customers, and so on and so forth. And the job really was about getting them to use it, I can understand what they want, help us to design around that to, you know, do whatever we can to get them to use it. Because the whole adoption conversation, I mean, it’s been rumbling on now for, what, many years, right? And we’re still relatively in the same place. But as I said, I think the product piece has started to move that along a bit, which is probably why, you know, data product today or [as a 00:25:31] product, however we’re coining it, is seeming to be getting a little bit more traction and airtime, the longer time goes on.

Brian: You mentioned this is going to be a well-paid space as well. So, let’s talk about money for a second, if you’re okay with that. What’s the salary bands, like, for this? And can you compare that to say, a data science role or something that our listeners might be more familiar with? I’m trying to help them kind of see, like, the investment is equal to, you know, one-and-a-half of these that you might have or something. Can you talk a little bit of, you know, mid to large-size enterprise, which I’m guessing is where you’re typically going to see this type of role. Like, what kind of money can people make in this space?

Kyle: Yes, I think, you know, broadly speaking, salaries across organizations, within reason, are relatively similar, right? You have a few outliers, obviously, big tech players pay, you know, substantially more, but think everyone else is within a certain remit. So, I mean, from what I’m seeing, and from the roles that we’ve, you know, been working on and where we’ve been helping businesses to build out their product teams, it’s been relatively similar to, you know, you have data science-type salaries, for the most part, obviously, I think, you know, what we’ve seen generally across the industry is five years ago, it was all about data science, they were the people getting paid the big bucks. That kind of slid away a little bit and then, you know, the rise of the data engineer has come to the forefront now. So, I think, you know, data engineering is probably the most well-paid space and the kind of, you know, “Practitioner,” in quotation marks, kind of landscape across data.

But yeah, the product stuff is up there, for sure. You know, so I think most of it is very, very similar. I’d say, you know, it’s kind of on par with data science. But again, I think what we are starting to see across the industry is more well-rounded teams being formed, which is really good to see, actually, you know. If you think even a couple of years back, practically every person in the data analytics team was a hardcore techie, right?

Brian: Right.

Kyle: Even your analysts were, you know, they were hired based on how good they were using Tableau or whatever the case may be, right, as opposed to some of the softer skills. And we’re now starting to see many organizations hiring people out of, I don’t know, change and transformation, or out of communications and PR, right? So, we’re getting to this point where data teams aren’t all about the core data job titles; there’s more to it than that. So, I think there’s probably an appreciation, you know, if we were having this conversation two years ago—well, we probably wouldn’t be having this conversation two years ago—

Brian: [laugh].

Kyle: —[crosstalk 00:28:06] there, right? So, you know, the, this type of train of thought didn’t really exist. But now, you know, I think businesses are waking up to the fact that we can invest all this money, we can build the greatest tech platform in the world, the greatest data lake, warehouse, whatever, you know, whatever we’re [crosstalk 00:28:24] here, but until people start using it, it’s all you know, it’s all relatively useless. So, I think there’s been a big organization, or there’s been a big shift to actually okay, we probably need to start thinking about some of the softer skills, but how we connect the dots to the central data team who are building these products, and, you know, so on and so forth. So, yeah, I mean, money-wise, I’d say, probably the same in the realm of data science at this point in time.

Brian: Can you share any ranges, like in Europe or even the US? I don’t know how—what your company, the gamut you cover geographically with your work, but can you share some ranges for people that don’t know?

Kyle: Yeah, we work across the UK, Europe, and the US, I think. Obviously, there’s some—I mean, it depends on the size of the company often, right, but I think, you know, in the realms of data science and data products, you know—and again, it depends on experience, right, is it middle level? Is it senior? I don’t know. You’re probably looking $150 to $180 as a base.

But again, it really varies. If you’re speaking about a Fortune 100, they might weigh heavily more towards bonus in stock, as opposed to base. So, it is slightly different in the US because we can get some organizations where they’ll pay an obscene base, but you don’t get much else. So then, you know, a competitor might pay a relatively low base, but you know, load you up with stock and bonus and things like that. The UK and Europe market is relatively similar, I guess, in terms of rates across the board.

And again, depends on experience, so you know, data scientists, you’re probably looking at $50 to $80k and that would get you a, you know, probably someone with a few years experience up to someone slightly more senior. You then start to creep into the realms of management, leadership, et cetera, et cetera, et cetera. So, it can depend. It really, you know, depends on the size of the business, the maturity, where there are at, whole host of factors. But yeah, hopefully, that gives you a bit of a flavor.

Brian: No, that’s helpful. And I, you know, I would suggest, too, if you’re on the candidate side, management side, well, you can take this information for listeners who are in management or in the IC side looking for this role. But if you do this role right, it’s really going to be about the adoption piece, which then translates to the business value piece. And if you get good at explaining the business value, you can bring and you can quantify that through your solutions—which is a skill you need—you can probably leverage some salary increases well because you can actually quantify the impact of your work, which some other people might not be able to do that are mostly focused on doing the implementation of tools and building platforms and all this kind of stuff.

Kyle: Without a doubt.

Brian: So, something to keep in mind that can be used as a weapon or defensively [laugh] in negotiation, but I don’t know, that was the first thing that came to mind. It’s like, it sounds to me like they’re pricing this role like everybody else because we don’t really know what it is, so we’re kind of going to, like, seat it in the middle with everything else. That kind of is what I’m hearing? I don’t know. [laugh].

Kyle: Yeah, no, yeah, you’re absolutely right. And I think, broadly speaking, you know, that’s where most organizations pin most of their data analytics roles, right? You know, I mean, I’m very vocal on the fact that, you know, the chief data officer or the data leader in any organization of today, their job now is about identifying and quantifying and articulating the commercial value that them and their team are adding. That’s what we’ve gotten to now based on the last however many years of building these big tech platforms, you know, lakes, whatever and businesses come in to us with their hand out saying, “Well, where’s the ROI?” Right?

So, I think—but I mean, it’s a really fascinating space because it divides opinion. Like, you know, I speak to people all the time who are kind of like, “Yeah, that’s not possible.” And I’m like, “Hmm, well, I think it is possible.” So, I think, you know, as soon as you get to a point where if you can help to drive adoption and then you can quantify the commercial benefit of that adoption to the organization, you know, that probably puts you up near the top in terms of percentile of being important to a data organization, you know? So, whether you’re the leader or you’re the contributor, you know, if you’re the leader, great, if you’re not the leader, the leader will definitely want to have you and keep you and pay you well, right, because you’re helping them to deliver on their job effectively.

So yeah, I think—but we’re still in a place where it’s not common practice. You know, it isn’t common practice at all for data teams to be able to say, “We did this. That’s resulted in”—you know, whatever—“10 million on the bottom line that wasn’t there before.”

Brian: Right.

Kyle: We’re still not in that place, despite the debate in the conversation. You know, most people think we should be there. We probably shouldn’t be there, but you know, that I think that’s the thing that a lot of data analytics professionals are going to have to start upskilling themselves in is that, you know, the value creation piece and being able to put your hat on that, that was you’re doing. And often it’s not a case of they haven’t done it; it’s they struggle to articulate that in business lingo, right? You know, it’s almost the articulation of that, and probably the marking of that, you know? Most data teams are naturally not the best at marketing themselves because they don’t see the importance of doing that. But you know, we’re moving away from that, for sure.

Brian: Yeah, it’s funny, you mentioned this. So, I’m launching this—I don’t know if I told you this—data product leadership community. And I’m opening up that research and design process, so we have a shared Google Doc, where I’ve been inviting people that listen to the show and my mailing list to go and comment on what would the community look like in version one? What are the benefits and all this kind of stuff? And this value, being able to quantify the value topic actually came up, and there was several comments in a thread about this from two people I know that were—and the reason they struggle with this is they’re on the enablement team, so they build a platform on which other people then go, say, deploy a model and then the data—they’re like, well, then the data sciences, they get the credit for creating the thing that is user-facing that actually feels like it generates the value, but it couldn’t have been done without a platform on which to build it.

Which I understand is kind of like, “Oh, look at these nice faucets in the house.” It’s like yeah, but you need an entire sewer system and plumbing infrastructure before you could ever have those, so who gets the credit and how much is it worth? And there’s an art, there’s both art and science to quantifying this stuff and I think that’s where the challenge is, is we get lost in the precision part of it, where value is very subjective. And so, once you realize it’s subjective and you have to figure out, well, who’s the person that is actually deciding what your value is, their perception of the value has to be factored into this. And once you have an idea of what that is, then you can get into quantifying it with some math, like, you can run numbers.

But there’s a whole belief system, it’s just like, how much of the dollar worth? It’s like someone told you that the dollar is worth a dollar, but there’s a whole belief system just wrapped up in money. And so, I think that’s the thing people need to unlock here is that there’s a little bit of an art to this as well as the science of it. It’s not all about just running statistics and numbers; there’s an art to it, too, and it’s subjective.

And that’s okay. We don’t need to get super precise. You won’t ever really be totally precise because someone else may have a, like, no, the platform is not worth this over time. It’s, you know, it’s only what you build on top of it. And someone else will be, like, well, without plumbing, there’s no water, so I believe the infrastructure—who’s right? You can’t—it doesn’t it almo—there’s no absolute right and that’s the thing we have to let go of is there’s no absolute right, actually, to that answer. In my opinion. I don’t know.

Kyle: I mean, the interesting thing with that is that you can kick that down the road for days and days, right? You know—

Brian: Exactly.

Kyle: —so the enablement team is saying, “Well, without us, this isn’t possible.” The data science team are getting the credit in their eyes, but then, I don’t know, they build a model that helps sales forecasting, well then, the sales team saying, “Well, hey look, we brought the sales in, so it’s on us.” Right? And then your sales officers saying, “Well, hang on. I run the team. It’s mine.” Right [laugh]? You know?

So, you can kick it down the road, I think. When I often speak about the importance of soft skills, I think one area that is massively undervalued and under-talked-about is the ability to build relationships, you know? And I think that’s often where this comes in internally, you know? If you’re a data enablement team that’s helping a data science team, whoever’s running those different teams and divisions should be working closely together in partnership, and saying, “Right, you know, if we come up with something great here for the business, well, you know, let’s agree what that’s worth to each team.” You know, and the many people I’ve interviewed over the years, in the day job, on the podcast, spoken to events, often they talk about that kind of, you know, setting the expectations up front to say, you know, if we do this and it results in X, then we want to be allocated X amount of value credit for that, you know?

Then it’s almost setting expectations up front and having a mechanism to do that. But I think, you know, most data analytics teams have been at this for a long time. They arrive at the point of there’s some uplift and then everyone’s squabbling over [laugh], you know, how important their part in that role is, as being everyone’s kind of, you know, grappling for the value because now we’re in a constant cycle of talking about, it’s all about value and value creation, so we’re kind of reactively trying to fight for the stuff that we’ve already done as opposed to, you know, trying to find a way, from the outset, to understand the subjectiveness of it, I guess.

Brian: Right, right. We could probably go on for hours about that, but [laugh] that’s a whole ‘nother topic. But I wanted to get just a little bit of practical stuff here before we wrap up. First, in terms of hiring data product managers, data product leaders, any suggestions on how to do that in terms of, like, job description stuff, and—I’d like to get—you know, if you can get fairly specific in this domain, that’s helpful. I know, there’s a lot of just generally bad job descriptions; I see it in the design field.

And there’s still just—like, what do you—what’s going to happen in six months after I’m at this job? I have no idea because you just listed a bunch of skills, but I don’t really know why you’re even hiring the role. Like, what would be different in the future after I’ve been here for six months? I mean, that’s what I would be asking if I were if I was still in the W-2 market, which I haven’t been for 15 years. But any suggestions on how to write a description? Is the job description even the important thing these days with this role? Like, where does someone get started? And any tips for how to find these people?

Kyle: Yeah. So, I mean, as part of our, you know, advisory business, this is probably the area that we speak most about, right in terms of talent attraction, it’s almost understanding—well, identifying, understanding, and being able to articulate—similar with the whole value conversation—exactly what it is that makes your business team opportunity role compelling. And believe it or not, I’m sure you’re going to be shocked to hear, Brian, but that’s not the list of tech skills that, you know, people are that bothered about. I think—

Brian: Yeah.

Kyle: You know, it’s ironic because we’re forever talking in our industry about the importance of storytelling. Yeah, I’ve never seen a business once tell a good story about how good it is to work for them, specifically in regards to their data analytics team and telling a story about that. And I think that’s often where there’s a disconnect, is that often businesses are led by the perception of whatever is above the door, especially big organizations. They really suffer from this. You know, they think just because they’re a big brand name that everyone’s going to flock to work for them.

And outside of Facebook, Google, Amazon, Twitter, Netflix, et cetera, who most people will go and want to work for, regardless of how bad they’re paid or how bad the conditions are because it looks great on a resume, everyone else is now kind of competing in the same field. But I think we’ve gone on a journey where many people now feel that they’ve been employed by businesses before and feel a little bit burnt by the experience they had. You know, everyone’s talking about how we’re moving, you know, on this big data transformation journey and we’re, “Data-driven,” in quotation marks, and they get in there, they’re delivering work that doesn’t see the light of day, and their impact within that team, their organization, it’s just invisible, right? So, I always talk about the concept of, how do you articulate how someone’s work will be visible, valuable, and impactful? I think that’s the thing that you try to—the story that you try to tell in a job description or job advert.

You know, we could talk for hours about this. Adverts and job descriptions are two completely different things, although most businesses create a job description and copy and paste it into a Word document and then that’s the advert, right? That’s a thing in itself. Like, the advert is there to—it’s a commercial, right? You wouldn’t see a BMW commercial on a billboard that listed all of the nuts and bolts that it uses in its tires, right?

But that’s what job descriptions are, right? You kind of a commercial for a BMW talks about what you become when you drive a BMW, the status, it gives you the feeling you get when you sat in the car, you know, all of that type of stuff. That’s what businesses should be focusing on, how they articulate what it’s like to work there. And then you can start to get into the components of what that looks like.

You know, what is the business goals and objectives? How is the data strategy designed to support that? What does this roll slash team do in order to support that data strategy? How is that individual person’s work going to be impactful towards that overall mission? Who is the data leader that they would work for? Do they have any credibility standing in the data analytics community and are they well known? Do they speak at conferences?

There’s a whole host of stuff that you can look at, which is why it almost infuriates me because it’s still never been easier to stand out from the crowd. Like all you’ve got to do is write a half-decent job description slash advert in the form of a story. Ironically, a lot of big businesses are bound by systems. So, I was having a conversation this morning and, you know, opportunity—wise, and I mean, this is a big, global automotive brand, right? So, by the very nature of it, they would get people wanting to work for them anyway because it’d look pretty cool, and I’m sure the job itself would be pretty cool.

Everything that they told me was fairly positive, but then we started to review some of their job descriptions and the chief data officer is saying, “I’m kind of hamstrung here because what I’ve got to do on the back end is, I’ve got to answer this question in a box, and then the job description is formulated off that.” So, everything looks and sounds the same because everything’s been answered the same; it’s based on a set series of questions that goes into a system and spits out the other side this awful-looking templated document, right? So, in that instance, I was saying to him, “Well okay, you know, that’s what you’re working with, that’s what you’re working with. You need another way, another medium to tell your story.” So, you know, and this person is starting to get more active on LinkedIn, so I was like, “Great. That’s your platform. Go and tell your story on LinkedIn and drive people to that as opposed to looking at that god-awful”—

Brian: To that boring job description [laugh]?

Kyle: Yeah [laugh].

Brian: Where they’ll immediately be let down [laugh].

Kyle: Yeah, exactly. I mean, the story is cool, right, but just the process and the system that they’ve got to follow just doesn’t allow for that to come out. So. So yeah, I think broadly speaking, that’s really what’s starting to matter across every job. But obviously, you know, there’s—and this is the thing that I talk about a lot: there’s nuances to every role, right, so what matters to a data analyst or data scientist is hugely different to what matters to an accountant, hugely different to what matters to a marketeer, that’s usually different to what matters to a salesperson. You know, the list goes on.

Even within the data analytics spectrum itself, like, what a governance person cares about is wildly different to what an engineer cares about that’s wildly different to what ML engineer cares about, you know? So, the whole spectrum is almost—you know, I think businesses are very guilty—and understandably so because it’s very time-consuming—but they’re very guilty of, there’s a template, we follow that template, so every job looks the same and it just doesn’t speak to the specific audience that you’re trying to attract. And I think that’s the thing is, I mean, goodness, we speak about personalization non-stop, right, in today’s world, yet look at a job description and tell me how personalized that is to the specific type of person that they are trying to attract. It’s non-existent.

So, I think that is the thing. It’s understanding, in this realm, of course, like, what’s important to product people, like, beyond the stuff of what will you become? What is the future look like? How does that all sew together? What will I be doing day-to-day? How will my work be valuable and impactful and visible across the business? But then specifically from product lens, how does that sit? How does that all piece together? How does that—within that team, how does your work impact the users? How does your work impact the central data team? What does that relationship look like? There’s a whole host of things to it. So yeah.

Brian: It sounds like, “I just want to write Python code. Leave me alone. Let me do that. I’ll be happy.” Maybe there’s less of those people coming in the door. People want their work to matter more or get used? I don’t know. That’s-it sounds like [laugh]—

Kyle: Yeah, I mean look—yeah, I mean, if you rewind… I mean, probably pre-Covid, right, there were a lot of people that would choose an employment opportunity based on you know—and this is data analytics-specific—but will choose an employment opportunity based on what projects do I get to work on and how, you know, quotation marks, “Sexy,” are they?

Brian: Yeah.

Kyle: What is the tech stack, and how much are you going to pay me? That’s really the core things that, you know, and obviously, the logistical stuff like, how often do I need to be there and where am I based all that stuff? After that, you know, the world changed through Covid, but I think, you know, it already started to shift a little bit because people were just, you know, they’d go into an organization that do all this great work, it wasn’t getting used, no one knew they even existed, and then they were just like, “Well, this is absolutely pointless. Like, I want to be somewhere where I have a purpose.”

Brian: Right? Go figure [laugh].

Kyle: So yeah, exactly, right? It’s about getting to that point of understanding, of course they still want to work on sexy projects, of course they still want to use the latest and greatest tech, of course they still want the best salary they can, but they also now want something more. But you know, no one’s really talking to that, and the ones that do I find that—you know, we talk about a talent shortage all the time. You know, for many roles, there is a shortage if you were to look at this on a pure mathematical basis; you know, engineering is a great example. You know, there are probably a lot, lot, lot more jobs for senior data engineering folk than there are people that can do it, so in that case, there is a talent shortage.

But what you’ll find is the market is lopsided to the ones that can attract better people by telling a better story, you know? So, you might have a whole host, a big quadrant of those data engineers work at one or two companies because they’ve just figured out what it is that it takes to get them through the door by telling a great story. And, you know, yes, they’ve probably got the best tech and they’ve probably got the best projects, and they probably pay really well, but they’ve also managed to get them to engage with them, which is really part of the journey.

Brian: Kyle, this has been awesome. Thank you so much for bringing something really different to the show here. Where can people learn about Orbition Group? I’m assuming you can probably help people, management, the leaders that are listening to the show, if they are trying to staff up on data product managers and that talent, I’m assuming that’s something you can help with. Can you just kind of give a spiel? Where should people find you? What’s a good place to connect?

Kyle: Yeah, well, you can definitely find me on LinkedIn. I think I’m the only Kyle Winterbottom on planet Earth, so it shouldn’t be too hard to find. www.orbitiongroup.com. There’s a whole host of detail on there that you can scan. But yeah, LinkedIn is probably the best place.

Brian: And just my branding tip for you is you should also buy notorbition.com.

Kyle: [laugh]. Yeah.

Brian: And that’s a shout-out to my friend Brenda from Lycos. There’s a designer I worked with named Brenna and her website was notbrenda.com.

Kyle: [laugh].

Brian: I thought that was the best name ever. So, [laugh] Kyle, it’s been so good to talk to you. Thank you so much for coming on Experiencing Data.

Kyle: No problem. Thanks for having me. It’s been a pleasure.

Brian: Yeah, yeah.