[00:00:08] Lauren Burke: Welcome to Women in Analytics After Hours, the podcast where we hang out and learn with the WIA community. Each episode we sit down with women in the data and analytics space to talk about what they do, how they got there, where they found analytics along the way, and more. I'm your host, Lauren Burke, and I'd like to thank you for joining us.
All right. Welcome back to Women in Analytics After Hours. Today we have Jerrie Kumalah joining us. Jerrie is an Analytics Engineer with a passion for democratizing data and empowering new data users. She believes in the power of data and finding ways to make it approachable and useful for all.
I am so excited to have her here with us today. So welcome, Jerrie, it's so great to have you on the show.
[00:01:00] Jerrie Kumalah: Thanks for having me. I am actually super excited to be here.
[00:01:04] Lauren Burke: I am too. I'm so glad you were able to join us. So just to start out, could you tell us a little bit more about your background? I know you took a pretty unique path into data and just some of the stops you had along the way.
[00:01:16] Jerrie Kumalah: Yeah. So I started in public health. I got my Master's in Public Health at UNC Chapel Hill in Health Behavior. So not necessarily super data heavy, but definitely with a foundation in biostats. And my first job out of grad school was actually as a researcher collecting primary data, so doing interviews and even collecting biological samples. And one thing that struck me when I started that path was trying to understand how they were gonna use that data to inform programming, which is a big part of public health, right? You generate insights and then you think about scaling that as a population level.
And so I was able to get an opportunity working as a Social Epidemiologist, which was relatively new for the state of North Carolina at the time. And so basically thinking about using surveillance data sets to start thinking about the social factors that influenced disease at the population level and using that to help inform policy and programs and initiatives and lead trainings and things like that. And what I loved is just like, that was my first real taste of starting to see the power of data when it comes to like, especially issues in communities and helping, you know, create voice around the things that are hard to quantify, but also equip communities on how to think about it, how to start thinking about collecting the right sets and how to compliment their narrative with data that existed.
And so I continued in that path in one way or another in local government and the nonprofit arena. Basically using data as a way to inform programs or evaluate programs and initiatives or identify areas of opportunity. So always shaping kind of that story and really kind of thinking about it from a population health perspective.
And as I kept going, data science was really booming and how we were talking about data was really shifting outside of the health space. And I felt like in public health it was pretty, um, rigid in how we thought about data and we were really slow in terms of like how we thought about processing information.
So for example, you'd collect surveillance data and it could take up to a year or more to clean, restructure, and then translate that into digestible chunks for policy making or programs or for local health departments to start using it to inform their initiatives. And so I definitely realized that there's a lot I could learn outside of that space.
And also excited about the opportunity that existed outside of public health. And that's kind of what led me to eventually analytics engineering, but started off, um, more in this quest to discover what data science was all about.
[00:03:56] Lauren Burke: That's awesome and I love that you called out the possibilities there are with social and community data because there is so much of it out there. And a lot of times, like you mentioned, there's not a lot of it being looked at.
We just had someone on the podcast a couple weeks ago, Leah Bowers, who is working with a lot of social data sets and trying to help communities and nonprofits understand how they can better help the people that they're serving. So I love that you mentioned that as one of the reasons you're starting to see the power of data.
Yeah. But you mentioned that you transitioned into tech from public health. So what inspired you to make that career change?
[00:04:33] Jerrie Kumalah: So what really inspired me, in addition to like really wanting to just solve problems differently than how we were solving it in the space that I was used to. Right. And you mentioned kind of like social data. So I did a lot of social determinants of health and health equity work and people wrestle with like, trying to think about that. I felt like I could learn a lot outside of public health, in terms of being more strategic in how we think about data.
And then the big driver really was flexibility. So I actually manage a chronic condition. And so for me, one of the important piece was also being able to continue growing in my career, but being in a space where also I had a little more flexibility in how I did that work.
And at the time, tech was one of the only spaces where you could work remotely. Like, that they provided that option. There were a lot of companies that were more willing cuz that was pre-COVID, right? This idea of working from home was in so foreign and this idea of being able to still problem solve and be impactful and support like these things that I'm passionate about felt like, okay, I can still do this in all these different spaces and hopefully one day go back to my health space. Um, and so that's what led the shift from public health and local state government to the tech industry.
[00:05:49] Lauren Burke: That's awesome, and you actually ended up at one point at a healthcare company or a health tech company, so you still got to bring in some of that feeling of working with the impactful data and making a difference, but doing it in the data space as opposed to the public health space.
[00:06:03] Jerrie Kumalah: Exactly. It was like going home for me.
[00:06:07] Lauren Burke: That's great. And so when you were making that transition, you had already had experience working with data, maybe in a slightly different way, but you'd been doing it for many years. So do you feel like your background and your experience in public health influences the way you approach your data roles now?
[00:06:25] Jerrie Kumalah: It significantly influences how I approach data problems. And I am actually really glad I was trained the way that I was because I always think about it in terms of what we're really trying to answer. So the business need, and really being able to zoom out because I've always had to think about what is the most impactful and be very, very strategic in what I'm going to invest my time and resources. Especially because in public health it's slower and you are working on a massive scale.
And so I do the same thing now. The hardest transition for me was translating it to the language that they use in this industry. It felt like, for a while I had a moment where I felt like is there on my, does my background really translate really well? And it just felt that I hadn't been equipped yet with the right language to say what is similar and what's different.
And so it felt like learning a whole different, um, language for sure. It was like, oh, this is what they mean. Oh, we've done this for a while. Or this is how you think about data architecture. This is how we've talked about it in MySpace before. Right? And so it was more spending time translating and better understanding like how my foundations were still really relevant and how my skill sets working with stakeholders and people and communities and getting people on the same page was still something very, very useful and needed in this space.
[00:07:50] Lauren Burke: That's so interesting, and you mentioned earlier that just the scope and the way you're working with data looks a little bit different. And often in public health, you're following a more rigid structured protocol. You have many steps that you have to follow exactly to a T and an industry, it really doesn't look like that. You're seeing things happening as you need them to. Right. A lot of more innovation, a lot more MVPs coming out just because you're really on a time crunch.
Are there any other differences like that, that you were surprised to see when you made your transition?
[00:08:20] Jerrie Kumalah: Yeah, I think one of the biggest ones is that some people are aware of the limitations of the data and some others are not, but are still willing to make really big decisions off of very small and incomplete pools of information in industry that you cannot do as nonchalantly in the health and in the public health and healthcare space.
You have to be able to clearly define what's missing. And that's also why we're slower in some ways, is all the limitations, right? Like, what are you inferring? Why are you making that assumption? How did you collect it? Why did you collect it this way? You know. And is it, uh, random or not? Like you, you start really leaning into some foundations and statistics, but also really thinking about how do you wanna be able to generalize this information?
Whereas in industry, um, yes, there's AB testing and thinking about how you break down certain things, but that's not as much of a limitation because the, the approach is more jump in and act. Right. It's don't be afraid, don't slow yourself down because of the limitations. Be aware of them. But like don't use that as a limiting factor, which is possible because of it's smaller scale tends. You tend to be product focused or systems focused in a much different way.
So for me it's been a huge shift as well of realizing it's okay, you can take more risks, you can be bolder, you can play with a data set in a very different way than I had been exposed to before.
[00:09:48] Lauren Burke: Yeah, I think that's such a interesting and helpful way of putting it. One of the things that jumped out, as you were saying that was I feel like a lot of time in academia in research, when you are looking into a potential question, you're thinking about a problem, you are not really sure what the answer is going to be, and you're okay with that.
And an industry, a lot of times, especially as a data person, someone's coming to you saying, "I think this is what's happening. Can you prove it for me?" And they're going to be much more aligned to whatever you come out with, if it also aligns to what they're hoping for.
[00:10:25] Jerrie Kumalah: Mm. Yeah, exactly.
[00:10:27] Lauren Burke: So now you are working as an analytics engineer, and you've been doing that for a few years now. And that is still kind of a relatively new job title within the industry. So what does your typical day look like as an analytics engineer?
[00:10:41] Jerrie Kumalah: I know it's a great question cuz I feel like analytics engineering can mean very different things I did in different places. I like to define it as a cross between someone who knows data analytics. So all the way, you know, to how you present the story to data engineering. So being able to flex across that entire pipeline, but with the business function top of mind.
So what that looks like for me right now is that I'm embedded in a marketing team. And I'm focused on thinking about the opportunities for automating data whenever possible, streamlining and standardizing data quality. So really thinking about what kind of questions do they typically try to answer? How do we make sure that it's consistent? There's good data quality around it. So monitoring that information, but also being able to zoom out far enough that you're not building data models that are hyper-specific to one particular question or ask, right? You're sort of like really thinking about the data infrastructure in a way that is optimal for the particular business unit and the type of questions that I typically ask.
But also keeping in mind the larger like core analytics engineering function of being able to kind of think about the system across the board. We use dbt, which is one of the tools that has definitely further defined the analytics engineering function. So it's become one of the tools, like that's often directly associated with analytics engineering, right?
But some places don't necessarily use dbt but still have analytics engineers. So my day-to-day looks like a mix of spending a lot of time coding or meeting with folks or kind of addressing bugs, trying to define kind of how to think about data, and just kind of generally always problem solving.
[00:12:26] Lauren Burke: That's awesome. And I know you've talked about this before, but a lot of your work as an analytics engineer also kind of aligns with the strategy side of data. So when you were coming into this new role as an analytics engineer, did you know that strategy was going to be such a big part of your role, or is that something you found along the way?
[00:12:43] Jerrie Kumalah: So I was lucky enough that how I fell into analytics engineering was also because of my love for data strategy. And so it aligned really well. I didn't realize it could be such a core part of the role, to be honest, but I realized it's such a strength to be able to think of it that way.
And so what I mean by data strategy is you have to really understand where the data is coming from, how you need to transform it, analyze it to make it as useful as possible. So all the different components that we often think about, but we don't spend time thinking about it consistently. So everybody approaches it differently. Right. But really the power is also in knowing what your approach is and what is most important in that process. Like, how we even think and define data quality is part of strategy, right?
How you think about monitoring it consistently or when you're thinking of transforming that information, is there a consistent pattern of like the level of granularity for how you are defining like your problem solving levels. So it's been one of those where it goes really nicely, hand in hand with this analytics engineering function for me, because it leverages my strengths and it allows me to be able to think big picture and then work through small pieces of the puzzle in different ways depending on what I'm working on. Which not every role gave me that opportunity, and that's why I felt like and I still feel like analytics engineering is that perfect cross for me where I can use all my skill sets and I get to jump into many, many different types of problems depending on where it sits in that whole like data pipeline.
[00:14:20] Lauren Burke: Yeah, I think that is such a absolutely necessary thing to have on an analytics or data team is just someone having that strategic vision in making sure that that's being applied and followed every step of the way. And so a lot of times you have people that are thinking about their roles in very different contexts, right?
You have analysts and data scientists who are thinking about how they can get the best insights to make decisions off of these data sets. Right? And then you're thinking about people on the architecture side, making sure that the data is flowing well, it's locked down, and sometimes there's a disconnect between that.
I feel like analytics engineering, as you've described it, is a really great flow between those roles and introducing a lot more collaboration opportunities than previously existed.
[00:15:09] Jerrie Kumalah: Yeah, exactly.
[00:15:11] Lauren Burke: Awesome. So how can others that are in similar roles, maybe not analytics engineering specifically, define their own data strategy to help influence how data is being used in their roles and on their teams?
[00:15:24] Jerrie Kumalah: Yeah, I think it's, um, first zooming out a little bit and seeing what your team tends to invest the most amount of time in. So we often think that we spend an equal amount of time across the entire data infrastructure and pipeline. Meaning, yeah, like, oh, I spend this amount of time generating insights and like working with stakeholders and building data models.
But the reality is, uh, that's not what it looks like, right? So you might notice that and depends on the level of maturity of your organization. So if you notice that you are in build mode and you everything, you spend a lot of time building, Like the foundational pieces so that people can be more self-serve or all these different things.
You need to be able to understand where your company is at and where you fit in. I do think it's when you have a better understanding of like where time is spent, you can also be more realistic in how you spend your time and what you wanted to further investigate or build out.
So, um, when I was in the early startup phase, there was a lot of more like fixing errors. Because everything was ad hoc, right? And tried to think about what does standardization look like so that when a question comes in, you're not like, oh, I'm gonna answer it right now, and that's it. But also starting to think about what does that look like in a world where she can answer this question on her own, because I think that is more powerful so that we can start thinking about insights in a different way.
And so data strategy for me is like understanding the level of maturity your organization is in. So are you always like just running through, um, fixing things and like doing ad hoc things or you know, space where you have these systems and processes in place. And turns out a lot of time right now is spent in coaching up, right?
Like your stakeholders in defining processes of what that means. Documentation, um, just for as examples. But when you understand where that is, then you can. Figure out how you think about your personal projects and how you have those conversations, um, with your manager, but also with stakeholders so that they understand that yes, you know, you could problem solve, right, address their issue like that.
But the reality is a lot of the focus is going to be on one angle of that problem, which might be standardization, or it might be building the mini, you know, the smallest piece of the larger picture because that's what makes the most sense right now. Um, it just allows you to have more refined conversation with the different people in the company and the team.
[00:18:01] Lauren Burke: Right, and one of the things you mentioned before with that is having that data strategy, even just for yourself and how you're going about working with data in your role, allows you to improve the way you're managing stakeholders and empowering them to understand and help you use data to make better decisions.
[00:18:19] Jerrie Kumalah: Mm-hmm. Exactly. Yeah.
[00:18:21] Lauren Burke: Are there any kind of skills that you've learned when you're managing stakeholders, when you're trying to figure out what people are really looking for when they're asking a question that you think would be helpful to share with others?
[00:18:33] Jerrie Kumalah: Yeah, I think the first one that seems very simple but has been very helpful for me is create a shared language. So don't assume that you are, um, both understanding the problem from the same angle, so defining what exactly it is that you're talking about from. What is engagement? How do you define it?
How do you look at it? What does it mean, right? Like when you say it doesn't look right, what, what does that gut instinct tell you? It doesn't look right. Like, what are the steps you follow to better understand why it doesn't look right to you? And also trust in that the subject matter expert. So if someone is not in the data world, right, but they're subject matter experts, let's say in engagement, I'm just using this random one. Trusting that the way they see information is important. Even if our data systems do not reflect that, because then it helps you better understand what's missing and then you drill.
And for me it's like constantly asking why. Why is the simplest thing you can do, but it's just asking the why. Like why you say it doesn't feel right? Why is that a problem? Right? And so there's the part of really understanding why, what the friction is and why it matters to that individual and what it's supposed to inform. Sometimes it's because there's pressure from the board, right? It might not be something that's personally important to them, but it's important to how we present that whole team and that matters. And so you need to know like what that means, right? And how understanding level importance.
And so in my past lives, what that looked like, um, which I haven't done as much in recent years, is I would shadow people. So I've worked where I had to really help build how they collected information about even like, Parks and recs and I would go to the rec center and see like from when someone comes in and they register to this particular community center, what does that look like?
Right from data collection to where it's stored, who's reporting it, what kind of errors are more likely, um, or forestry, how they would count trees in Baltimore. And so like being able to like, ride on the trucks and see kind of like what that process was like. And I use that as examples because there's an opportunity also for just understanding the flow of information and not necessarily like data generation.
So if they're not creating the data, it's more like how are they using it? Have you seen them use it? It helps you understand the gaps and it helps you better understand. What you're building for, but it gets you really invested in the whole piece because you also want them to know that you are a champion and you're trying to make their life easier.
The big thing that I've noticed is that people are afraid to slow down. They like to just go, go, go. Um, I do like to slow down. To a fault, maybe sometimes it's not always. Sometimes I need to be quicker. But I'm okay with the, this practice of getting people to pause and focus on the right things because then everyone is bought in, is on the same page and has a shared understanding of why it's important and how you champion that person.
And a lot of people wanna be more data savvy, but there's this bridge in how we talk about concepts that can be very alienating. And so the other piece is helping people understand what you're doing. It's just a simple thing, you know, like what does a data pipeline mean? What does monitoring really mean when you're saying you're doing that?
And why are your steps so filled with so many different pieces. So being able to kind of like break down your flow and why it's important and why it needs to be consistent and why they should, your stakeholders should be engaged is just as important.
[00:22:12] Lauren Burke: That's such a good thing to be thinking about. Because a lot of times when someone is asking a analytics request, you to create a data science project for them. They're thinking about it as just this end thing, this one singular project, but really, like you're saying, it's a bunch of different steps.
It's a bunch of different ways that you need to be interacting with others, finding things along the way, learning things, putting them together. I love that you mentioned that because I think that could help a lot of people, especially when they're setting up their data practice.
Just saying, this is the way that our projects typically go. These are the steps we'll take. This is where you come in. This is where I come in and along the way, this is where I will need your help and these are the things that potentially could be roadblocks. These are the things that could potentially help us do this quicker. I also love making templates just because.
[00:23:02] Jerrie Kumalah: Yes.
[00:23:03] Lauren Burke: I think it helps you so much if you can go back and then you have a template of how you worked on this particular project, especially if you worked with a particular person before, you can send that back to them and they will hopefully remember that process and be able to understand what stage they're at and how they can best support you.
[00:23:20] Jerrie Kumalah: Exactly, because the other part that we don't often like capture, like even in our, you know, when you're sprint planning or thinking through stuff is like the iteration piece is not something that you can capture really well in a ticket or in a flow, but yet it's essential to the success.
And value that you generate from a project is, um, not always just the end result, but that iteration process so that they're part of the decision making. They understand why it was a yes or no, why it was prioritized or not. And those are like conversations that you're having or finding ways to document.
And so I don't know the stakeholder piece for me, because I came from a people-centered field. It was just like really, um, it's important to how I think about things and um, feel like it brings a lot of value.
[00:24:08] Lauren Burke: I absolutely agree. I also liked what you mentioned about the flow and seeing where you are and where it's going, because a lot of times, like you mentioned, someone's just asking a question that someone told them to find out about and that might be going three levels above them for the final decision.
And so you need to be thinking about, oh, that's who I actually am, not in the real sense but technical sense, that's who I'm presenting to. That's who needs to understand and be able to make a decision off of this information. And I think that's something you, when you're starting out, you often aren't thinking about. You're thinking, this person asked me this, I'm going to answer it exactly as, as they intend for me to answer it.
[00:24:47] Jerrie Kumalah: Yeah, I, I used to do, uh, I like stakeholder matrix. I forget what it's called. Basically, I was, Also trained as a project manager, and I would just like, who is really making the decision? Who needs to know, but like stay informed, who is just like, has and understanding the power structure. So really understanding the whole like hierarchy, but why certain people need to be kept in a loop a certain way, or why the question might have been framed a certain way.
And it really helped like just give me a good sense of how to be thinking about like, that one simple question, right? Of engagement. I love engagement cuz that's always been one of like those where I'm just like, oh my gosh. Uh, it means so many different things to different people. And so being able to keep that in mind, like, who's needs to make the final decision?
Is it one, is it two? It helps you like, prioritize also how you're gonna tackle the work, like in terms of speed, and how accurate you need to be and things like that.
[00:25:41] Lauren Burke: That's so awesome and I, I bet it also helps you kind of figure out what people's true priorities are, right? Because two people could be asking, you could be working on the same project, and they're both thinking that this end goal is what we want. And the other one is thinking this end goal and this is our measure of success.
And without talking to them or without having that matrix from your previous experience, you might be like, "Ah, well which one should we really be going with? Where's this going in the end?"
[00:26:07] Jerrie Kumalah: Exactly.
[00:26:07] Lauren Burke: That's awesome. I love that you're bringing so many different roles into the way you approach problems.
You're bringing in the project management side, you're bringing in the people affected by the data and the people who are making decisions off of that data and all of that wrapped up together kind of helps you figure out how to work with each person at each stage and help you, just yourself, understand what you're truly working with and working toward.
[00:26:32] Jerrie Kumalah: Exactly.
[00:26:33] Lauren Burke: That's awesome. And before we wrap up today, I always like to ask about your favorite resource. So what is a resource that has helped you in your career that you think might help others who are listening?
[00:26:45] Jerrie Kumalah: So I thought about this in many different ways, but I feel like what's nice is that a lot of your other guests have mentioned resources that are important to me. So I will share one that we undervalue, which is reflection. I'm a journaler by nature, and what I mean is like being able to sit down and really assess kind of, um, your strengths, your weaknesses, and where you wanna explore.
And who you wanna talk to that kind of like exude those things that are important to you? It has helped me I'm realizing, especially when I rarely think about it the most in my career because I've had many pivots and have had moments where I question like, am I in the right direction? It doesn't look like anybody that I know or like this idea of success that I had in my mind.
But by being able to kind of reflect on it and connect with folks, that are in that ideal position that I'm interested in or have like stories that I wanna hear, like, how did you get there? It would affirm it. It really help has helped me like affirm that I'm in the right path or that I am growing or also determine like the areas that I wanna grow in professionally.
So just a week ago I was talking to someone and I looked at my notes from the past five years cuz I have little blurbs, you know, these different reflections. And the map is beautiful because I've actually steadily been working towards this big vision that I have for myself. But sometimes it's easy to forget.
And so reflection in the sense helps you like, map out the opportunities you need to seize, the people you should connect with. So like that network you wanna build and the skills that you wanna build as well. So kind of what to hone in instead of trying to do everything in one go.
[00:28:32] Lauren Burke: That's awesome. I love that. I think that's so helpful. I mean, we do retrospectives on the job, but we don't really do retrospectives internally to figure out where we've been and where we're going. And with journaling, right? You have a clear record of when I made this goal, when I was thinking about where I wanted to be in five years, this is what went into it, this is why I had these goals and this is how I thought I would get there.
[00:28:55] Jerrie Kumalah: I, I mean because priorities always change, but then all of a sudden you notice your patterns and you're like, I am still working towards this thing that I defined for myself and I thought I had forgotten about. So, yeah.
[00:29:08] Lauren Burke: That's so cool, and I love that you can probably also find that you've achieved some goals in different ways than you thought, or you had taken them off your list and you look back and you find, hey, I thought I wasn't on the path to achieve this, but I have actually in a different way than expected, gotten to this point and been able to find something new that I am really proud of.
[00:29:23] Jerrie Kumalah: Mm-hmm. It's a beautiful thing in many ways. I have been rereading old journal entries, especially in moments where I'm pretty down or, just questioning, um, where I'm at or what I'm looking for for myself. And then you feel kind of, it's like talking to you or past self. That's cheering you on in some ways and reminding you, you did your homework and maybe this didn't work out.
Now you're trying something else. Or, you know, you thought you didn't learn anything from this awful experience doing X, Y, Z, and then all of a sudden you're like, oh, I gained so much, so many insights that I now use to be more discerning and, um, more intentional in how I support others or think about problems.
So, uh, reflection, spend time doing that more alongside everything else.
[00:30:20] Lauren Burke: I think that's a great resource to add to your collection of maybe not physical resources, but things you should have on your mental shelf of things you're doing for yourself to keep moving forward and improving.
[00:30:25] Jerrie Kumalah: Exactly.
[00:30:27] Lauren Burke: So finally, how can our listeners keep up with you? Where can they find you, on the internet or just out in the world?
[00:30:41] Jerrie Kumalah: I am on LinkedIn, um, and also in a number of Slack communities. Mostly like Locally Optimistic, Data Angels, and Operational Analytics Clubs are the one that you'll find me the most active or just like lurking. But I definitely enjoy building community and so I love connecting with folks. So LinkedIn and Slack, always a great place to find me.
[00:31:03] Lauren Burke: We will absolutely link both your LinkedIn and all of those Slack channels because I'm in a few of those and they're just such amazing places to meet people who are as excited about data as you might be, and it's great to have that kind of community and just know that people like you are out there and they wanna talk about data all the time, just like you do.
[00:31:22] Jerrie Kumalah: Exactly.
[00:31:24] Lauren Burke: Well, thank you so much, Jerrie, for joining us. I thought this was an amazing conversation. We got some really interesting insight into data strategy as an analytics engineer and in other roles. And your unique path and some of the skills that you're able to bring into your analytics practice and use in a way that makes you a better analyst, analytics engineer, and person to have on a data team.
[00:31:46] Jerrie Kumalah: Thanks so much for having me, Lauren. This was awesome.
.png)
