Building A More Data-Centric Company With HubSpot's Bridget Zingale

Media Thumbnail
  • 0.5
  • 1
  • 1.25
  • 1.5
  • 1.75
  • 2
This is a podcast episode titled, Building A More Data-Centric Company With HubSpot's Bridget Zingale. The summary for this episode is: Have you ever shown up to a meeting where you have one version of numbers to report only to find someone else in the meeting with a completely different version of those numbers? And then you spend half the meeting figuring out whose numbers are right? In this episode, Bridget Zingale, the Global Director of Analytics at HubSpot, is going to help us all avoid this all-too-familiar situation. Bridget and Sean talk about how the HubSpot analytics organization has evolved, what it means to promote data literacy at your company, and why the biggest risk in building centralized data teams is a departure from context. Like this episode? Be sure to leave a ⭐️⭐️⭐️⭐️⭐️⭐️ review and share the pod with your friends! You can connect with Sean on Twitter @Seany_Biz @HYPERGROWTH_pod

Bridget Zingale.: We are moving to a place where if you want to be successful as a company, everyone, literally in my opinion, everyone has to have some level of comfort with using data. They don't have to be an analyst, but they really have to understand how to use it to inform their work. I think that that will be really the deciding factor of what company has come out on top in different industries.

Sean Lane: Hello. Hello. And welcome to Operations, the show where we look under the hood of companies in hypergrowth. My name is Sean Lane. Have you ever shown up to a meeting where you have your own numbers to present only to find that somebody else in that exact same meeting has shown up with a completely different version of what should be the same numbers and then you spend half the meeting figuring out whose numbers are right? Anybody who tells you that this has never happened to them is either lying or they just don't work at a company with any other humans. So many businesses today say that they are data- driven, that data is at the center of what they do, but what does that actually mean? And how do you pull that off in a meaningful and lasting way? Luckily, our guest today has done exactly that and came into our conversation, armed with all the answers to help the rest of us do the same. That guest is Bridget Zingale, the global director of analytics at HubSpot. Bridget has spent the last five and a half years at HubSpot and has seen the sales and marketing software company explode from scale up to public company and beyond. With over 3000 employees and 73, 000 customers around the world, Bridget and the team at HubSpot are a pillar company here in Boston and a role model in so many different areas. In our discussion today, Bridget and I are going to talk about how the HubSpot Analytics Organization has evolved over time. What it means to promote data literacy inside of your company and why the biggest risk of building centralized data teams is the idea of being removed from context. But first, being the global director of analytics, which in addition to being a really cool title can mean a whole bunch of different things. So I asked Bridget to take me through what the analytics team does inside of HubSpot.

Bridget Zingale.: We joke all the time actually that the word analyst is incredibly broad inside the walls of HubSpot and our organic approach to this area, I think also led to some interesting team structures that I'll talk about in a second. But an analytical team typically actually runs the gamut from having very technical skill set. So data architecture, people really familiar with data warehouses to the more strategic business analyst who might be running business models, working on forecasting and demand planning and everything in between. And so the range of skillsets is really fascinating. And I feel like this area also, you end up meeting people with such an interesting range of skill sets as well, because I think it's a place that either people found their way to through really interesting paths or I think universities are now developing more specific educational tracks around this area, which is great. But for anyone a bit more senior, we're a bit of an autodidact, did a lot of working and learning on the job, and as a result, accumulated all sorts of interesting skill sets and talents on the way. And so what we've done in our current structure is we have a centralized team and then we have distributed teams. And again, I think anyone going from the startup to the scale- up phase, which is the point at which I joined HubSpot, reaches this realization around data. Because in the startup phase you basically just have wonderful, beautiful growth from everywhere and it's amazing and you're just sailing along. And then you hit this point where you've reached a certain size and margins become critical. You have to be incredibly strategic and very efficient. And that's where data becomes critical. And so that's around the inflection point where I joined. And so all of the distributed teams across HubSpot spun up just from organic needs within the various business units. What we're doing now is we are pushing on the centralized org to say, okay, you guys actually should be closely tied with product. And if you think about the flow of data within an organization, you have the product teams who are responsible for making sure that raw data is accessible, APIs, inaudible there's many different sources. Then you have data engineers who should be centralized, who work on ingesting that into your warehouse instances. And then you have an interesting split, which is what we're finalizing right now which is, certain datasets are so universal that it actually is helpful to have a team of what we call analyst engineers formatting them into tables that can be taken all sorts of directions, right? So revenue is a really good example of that, right? Everybody needs that data sets and anything around centralized systems is another thing that we're pushing on them to say, hey, you guys are actually the closest to this. So when we talk about rotation, enrichment logic, what's actually happening, what our success rates are inside of that process, that should be on that centralized team too. And then from there on top of that, you have all of these contextual data sets. And that's where being linked up to the business context is so critical. So usage data, email data, workflow data, all of your contact data, the way that's used really depends on the request and what we have found, because I think there's often this conversation of what's better centralized or distributed. And I don't think it matters. I think it's really just, you have to be super aware of the risks. So in a decentralized model, communication and transparency is absolutely key and you just have to make sure that work isn't duplicated, people are sharing, definitions across the org are aligned. And then in a centralized model, the risk is just around a departure from context, right? So having that knowledge, the splits between, for instance, how the executives need things to tie out to company goals versus how different teams need to optimize their programs, those differences that come up are so important to how people use data across the business. And a good example of that is like the definition of a customer can actually be different depending on how people approach it, right? So we have of course a company- wide definition of what a customer is, but we have teams that treat different people as customers, regardless of whether or not they actually fit that description. So in these instances, you need to make sure that the data serves all needs and the teams that are dealing with customers can use their definitions to optimize effectively. And that you also still then roll things up to the more proper definition that the company uses to report out.

Sean Lane: When people ask the best way to establish a data team centralized, decentralized, insert other option here, the answer isn't that simple. But based on Bridget's description, I think that the core building blocks can be that simple. First, she says, you have product teams that make the data accessible. Second, you've got data engineers to ingest the data, and then you've got a bit of a choose your own adventure in terms of what you do with that data. The most important thing though is understanding which definitions are universal while recognizing that sometimes a centralized approach might actually cost you something vital, context, and something that seems as simple as counting the number of customers you have, all of a sudden gets very murky. By the way, I'm going to save you the dictionary search that I had to do autodidact means self- taught. Anyways, I want to better understand how Bridget and her team fight that departure from context that she was referencing in a company that has over 3000 employees and only about 25 of them are analysts. How do those analysts not just stay close to the data, but to the business context as well?

Bridget Zingale.: This is part of the reason that the distributed teams just needs those technical resources, right? And we insist across the board that if you're an analyst, it is part of your responsibility. And I don't think this is totally, I don't think this... I don't know how common this is, but we basically insist that you have partners and people that you work with to absorb context on a regular basis, right? So part of being an analyst inside of HubSpot no matter where you are is working with product teams, working with marketers services sales to understand exactly how they're using the product, what their process is like, what they think their definitions are. When this all kicked off, we had a huge process around that, understanding what various people across the business thought were the definitions of key metrics, how they operated on a daily basis. And it's something that is just a constant effort. And that's how especially for companies that are, I think in tech, especially things just change really fast, new data sets crop up all the time. That's how you maintain the ability to jump into some raw data and say, okay, yeah, I know exactly how I'm going to need to format this and in which different ways, and then how to surface it in different areas, because I've absorbed all this context, right? So we rely on analysts to make those translations in a way that I'm not sure is entirely common.

Sean Lane: We've talked about this on this show before, and Bridget's point bears repeating. The context of her analysts comes from time spent within the business units and they view this as something that's rare and advantage to them. Isn't it crazy that spending time with those internal customers is rare? It makes it all the more impressive that a company of HubSpot size is making the point to do just that. But as operators, if we follow Bridget's lead, if enough of us do that, we can make this type of approach the rule, not the exception. And Bridget is the person by the way to listen to on this, because the model wasn't always in place like this at HubSpot before. In a business, many would imagine to be more mature, she and her team put in the painstaking work to move from a world where most data and reporting were happening in silos to the more advanced and hybrid model that they have today.

Bridget Zingale.: Yeah. Well, again, it's interesting because there was a certain level of maturity, but around data, again, I think this speaks to that whole startup to scale up inflection point. No one was really using data the way we are now, because no one really needs to when everything is just up into the right forever. People don't really feel... Basically every experiment you've run ends up working out, you're just not as motivated to really dig in. And then you hit this scale where you are so big, you have to be so conscious. You're building massive teams and you have to make sure that everything is informed with data because that's how you keep those margins healthy. And that's how you keep that up until the right growth once you've hit that scale. And so what I think happened with HubSpot, I would say that started becoming more and more of a trend. And they also started hiring people like me with analytical backgrounds who really wanted to dig in into this area. And you also had this moment of, as people were trying to start to use data, you had these hilarious looking back and at the time stressful meetings where basically different VPs from different parts of the org would report on the same metrics, totally different numbers, right? Because they were getting what they could from different people. We weren't really had definitions. And so really the move towards the structure now of really defining ownership lines and saying, okay, this is where the centralized handoff is, really defining that data process of like this is what product handles, this is what the DEs handle, analysts engineers, and then distributed analysts is to make sure that we have that alignment, which also keeps the business so much more efficient, right? And then suddenly VPs getting room and everyone has already seen the same numbers and has used to the same numbers and has been using them on a daily basis, right?

Sean Lane: Imagine that.

Bridget Zingale.: Right. That's the dream. And so we've gotten to a good place there, but it's interesting because even with all that effort, that example of the customer data still persists, right? Even with that mission of metrics that you would think would be so straightforward really aren't because there's always this split of how a team operates around a certain audience and how the company defines levels of their audience. And so-

Sean Lane: Well, that was my next question which is, when you're describing this world and you gave some examples of some of those centralized core numbers, right? Like revenue and usage data and contact data, when you say that, I imagine this big HubSpot like central dictionary, right? And I'm curious as you guys continue to grow and as you mentioned new use cases come up, what does it look like to say, okay, this is now a new thing that we're going to add to this dictionary? And we have all agreed that this is the reason why this needs to be in here. And we've all also agreed that this is the definition we're all going to abide by in this centralized place. Because I imagine that's a constant question that comes to your team but also not an easy thing to maintain.

Bridget Zingale.: Yeah. Oh, gosh. It's sure isn't. It is definitely collective ongoing effort and it is something that I think we have a good handle on internally for all of those metrics that the execs really see and use and roll up to those larger meetings and where it's more of a struggle is around what we call program optimization metrics which is just like, well, how does this one team, when this one team says this word, what do they actually mean? And tracking all of those as they develop. So yeah, we have a lot of resources and we have all sorts of fun processes around documentation. And I think it's just an ongoing effort because when things get busy, documentation is basically the first thing to fall away, I've noticed.

Sean Lane: Totally. Can you give me an example of a program optimization metric? What does that mean?

Bridget Zingale.: Yeah, for instance, one, I think really great example is when you talk about close rates, which probably everyone has some form of. When you think of the way a sales team needs to optimize close rates, what they want to see should be representative of what they see and feel in their CRM. So in that world, we would either say, hey, here are all the domains that hit you in this time period and here were your work grades against all of them and here's how many of them closed. Okay. Great. And that feels straightforward, and in some cases you actually even represented as a more of a ratio because they're seeing actual contact records come in. And so they'll want to see that number reflected and then it's the domain that closes. So it's a bit of a many to one. But you're trying to format things as, hey, here's what you saw inside of your world. And here's what happened downstream over this amount of time inside this influence window that mattered. But even those influence windows are often much more open- ended, right? It might be just like, hey, here's what closed forever, like at any time, whereas the execs and your overall company actually need those close rates to be used in different places that would then have to tie up to a company goals. So close rates is a really good example because the close rate that we often report out to the business has a lot of suppressions baked into it, has various rankings across different lead types baked into it. And so what that does is it prevents inflation or over counting if you want to use it for some forecast. So for instance, one of the things that we strip out for the exact level close rate is our e- commerce channel, right? And so the rep would still see, hey, if this came to you and you worked and you closed it, it's baked into your close rate, but we actually remove it in this other version, because that close rate we use in a lot of forecasting models and e- commerce is forecasted separately. So to avoid that double counting, we just strip it out, right? So you end up with a lot of different cases like that. And it's really just about how are you using this? Is this a human that just needs to see their activity and optimize against it, and roll that up to a manager, that can help all of them? Or is this something that you're going to end up using in different models or stacked up against all of these other pieces that are knocked out separately, in which case you have to take those things into account.

Sean Lane: Bridget's close rate example is complex. There are different suppressions, different use cases, comparisons, forecasts, it's a long list, but ultimately it comes down to two things who is the audience of that particular metric and what are they going to use it for? And keeping those two things straight while constantly adding new terms to the HubSpot dictionary is the challenge. Then once you have your definitions in place, the next question is what you do with all the data that you've got. I'm constantly asking myself at Drift about the medium I should use for different pieces of data or how to best present them. When I present the data, should I use Salesforce, or should that information be in Looker or in a slide deck? I wanted to get Bridget's advice on how she thinks about where that data lives and how to best present it to different audiences.

Bridget Zingale.: Yeah. It's such a good question. For a lot of our reporting, we have what we call internally source of truth dashboards. And so we'll bubble up metrics for different internal audiences, circulate the dashboard, get feedback, and then everyone refers to that going forward. We schedule them to hit inboxes. So everyone's always looking at the same thing. And this is actually part of the reason we have these splits, right? Because to your point, for the sales' rep dashboard, which is in many cases in Looker, because of the flexibility, it allows around granular cuts and influence metrics across disparate datasets but that close rate should line up exactly to what they're seeing in the CRM, right? Maybe the only reason that we have it in Looker is to say, oh, and by the way, if you cared about other lines of influence around these metrics, that's what this looks like. We develop those dashboards and then we just try to line them up to that operational world. So for the reps, it should reflect what they see in the CRM and for marketing, it should reflect what they're seeing in our marketing tools. And then most of the exec stuff is all in Looker. And so source of truth, I always joke that is a misnomer, but it's catchy and people like it. I can't remember who came up to it. I think it might've been our VP of marketing John Dick. Anyways, having those as the jumping off point for people has been really helpful.

Sean Lane: And the other part of what you were describing with putting those dashboards together or putting those suppressions in, that makes sense inside of each of those program optimization metrics is the people, right? The people on your team who have to put those puzzle pieces together and make those decisions. And so you mentioned a handful of different types of roles that might fall under this broad analyst term but I'm curious, how do you coach the people within your team to be able to connect those dots and make sure that they're taking in all of those different audiences' perspectives when they're creating these dashboards and these metrics?

Bridget Zingale.: Yeah. The first step is really familiarity with the operational area that they're trying to serve. So if you're trying to create something for the reps, you meet with those reps. We'll do everything from a few hours of shadowing so you can see how that moves through the CRM. You can pick up what their definitions are because what ends up happening I think operationally is you get these really successful teams who are not even used to thinking of, for instance," the definitions that they're using in certain areas" like what even is a lead, for instance, to different people. And so having the analysts attend team meetings, shadow people for parts of their day, tends to be some of the building blocks that we'll use when projects kickoff is really this getting an analyst embedded on that team temporarily so that they really absorb what this team does and what tools they use, what definitions are a default across them? And then once they've observed that looking at the data and to translate it becomes a much easier process. So it's interesting because I don't think it's typical really to a pure analyst role, honestly and I think maybe as we grow, what we will have is a split of people that basically are in a sense data translators, that could be honestly a full- time role is just observing that, understanding how to translate it to an analyst that does the more technical work

Sean Lane: I want to spend more time on this concept of data translator, the different audiences you are working with likely have differing of comfort with data and all of the audiences likely don't have the same level of comfort that an analyst or a data engineer does. So how do you make sure that the information you're presenting is being translated the way you intended or the way that they originally needed?

Bridget Zingale.: On the marketing side we talk a lot about content analysis and padding analysis. So we can offer them like, hey, here's all the amazing things you guys pushed out, but here are the ones that are tied most closely to these downstream business metrics. So can we lean into this area? It's really about offering them of a visual or a data set that allows them to immediately jump into, okay, I know what to do with this. I now know how to change this campaign. Or I now know what to say to my reps so that we can increase these really critical metrics.

Sean Lane: Are you guys actually spending time enabling those marketing folks on how to use that information as opposed to okay, here's the information, good luck? HubSpot, I feel like is famous for both training in the marketplace, but also the internal training that you guys do with your employees. And so I'm curious, how do you improve the overall data literacy of the employee base inside the organization so that all the hard work that you and your team are doing is getting as much value as it possibly can.

Bridget Zingale.: Yeah. Absolutely. I love that term too, data literacy, it's so critical. So that process that we talked where the analyst becomes embedded on the team, that happens more intensely at the beginning and the end, right? So at the beginning, they're trying to observe all the contexts that they can work with the data to create the findings that we think these areas will need. And then the other piece at the end is that enablement. And so there's a lot of it that happens project by project, team by team, where the analyst continues to work with them strategically, join those meetings, offer their insights and guide strategy for a while that way until it becomes more habitual and nature to those teams. And then there's also programs we run just more generally, but actually we're resetting how we do some of them. So for two years, we had what we called reporting champs, which was an enablement program that ran bi- weekly. And it was people across the org that had been nominated by their business area as not an analyst, but someone who was going to be a champion for making sure that people were using data and digesting these reports and understanding how to use them to inform their strategies. And so it was everything from offering them office hours to holding very specific sessions, or on new datasets, how to understand them or new reports, metrics, etc. And then we had a community built around it where the champs were also supposed to help each other to help us broaden our enablement with our still reasonably limited analyst, outside of our analyst headcount. So those efforts have been... And then we just push out a lot of resources. I think those efforts are incredibly important and I completely agree that we are moving to a place where if you want to be successful as a company, everyone, literally in my opinion, everyone has to have some level of comfort with using data. They don't have to be an analyst, but they really have to understand how to use it to inform their work. I think that that will be really the deciding factor of what company is come out on top in different industries. And so those efforts are critical, it's just a matter of balancing investment, finding ways to... We try to templatize a lot of things to your point, like glossaries are really critical. Those source of truth dashboards are really critical. We do a lot of Loom videos where an analyst will just turn on Loom as they're doing something and then post it on the Wiki to just say like, hey, are you doing something similar to this? Just go ahead and watch how one of these analysts would do this and how they would decipher this data and what learnings they take away from it. So there's some easy things that pretty much any company can start doing. And then you just have these longterm dream scenarios like the Data University that Airbnb spun up, which I had just heard such great things about. So it's a balance, but yeah, I know we touched on this before, but data is becoming like... The literacy term is just so apt because it's reading, writing these basic skills, people are going to need basic skills around just looking at data and understanding how to use it in their business area.

Sean Lane: I love Bridget's enthusiasm around this concept of data literacy. Here's how she put it to me using writing as an analogy. You don't have to be a novelist or a professional writer, but you need to know how to read and write, simple enough, right? Data's the same way. So whether you have the reporting champions program like HubSpot or Data University like Airbnb, my challenge to all of you is find what you can be doing to improve data literacy inside of your organization. And look, I get it, not everyone has the same level of sophistication or resources of a HubSpot or Airbnb. We certainly aren't there yet at Drift and many companies going through hyper- growth likely don't have a team of 25 analysts on hand. But the way that Bridget and her team have organized can be an aspirational North Star for all of us. It's a long road to get there. So I asked Bridget what her advice would be for those of us taking the next step on that long road to being a more data centric organization?

Bridget Zingale.: Yeah. Oh, God. And I would say, honestly, listen, I'm very pleased with our progress, but we also have a long way to go.

Sean Lane: I get it. I totally get it.

Bridget Zingale.: But it's such a good question. And it's a little bit dependent on industry and type of company and what your overall those are. But I would say actually the first step is probably that step around definitions, right? Because whether or not people are using data in a cohesive way, often people are using data without really knowing it which sounds weird to say, but what I mean is everyone has something that they're trying to watch, right? Something that everyone's working so hard, they all have something in mind of like, hey, I impacted this thing. And so that step around understanding those metrics as they bubble up and trying to align them across the business is so critical because it will set the foundation for the work that comes next in terms of saying, okay, you know what? Let's make sure actually that everyone is looking at this along the same definition. And then let's find out what other areas we think might be influencing this critical metric and get our hands on that data, right? You can take this step- by- step approach of, hey, here are all these people doing awesome things. What do they think they're impacting? Let's start there. Let's align those definitions. Then let's start digging in and trying to access our data to see what else might be influencing these areas and give everyone that much broader view into what's actually driving these areas that we care so much about and that we've hired against.

Sean Lane: Yeah. To your point from earlier, right? If you can turn that corner to the point where you have people showing up at meetings and everyone is on the same page about what the key things that they're measuring in that meeting and everyone has the same numbers like that indeed of itself is a massive step forward. Before we go at the end of each show, we're going to ask each guest the same lightning round of questions. Ready. Here we go. The best book you've read in the last six months?

Bridget Zingale.: Oh, man. Fiction or nonfiction?

Sean Lane: Doesn't matter.

Bridget Zingale.: So it sounds dark but it was really entertaining read which is, The Doomsday Calculation, which sounds especially dark given unfortunate recent times. It's really just about this interesting universal equation that effectively predicts the future in all sorts of areas and how that's been used. And it's a quick read for anyone who's interested in stats, but maybe didn't study it. It has to just a mild interest. I think it's written in a very engaging way, loops and a lot of historical figures. So I enjoyed it and I promise it's not as dark as the title.

Sean Lane: You're really hitting the director of analytics stereotype part with that book. I'm not going to lie.

Bridget Zingale.: I just finished it a month ago too. So it's gone away in the most inaudible. Yeah.

Sean Lane: All right. Next one. Favorite part about working in Ops or analytics in your case?

Bridget Zingale.: Oh, man. I have to list two, one, I find the work so interesting. I think Ops is... Obviously, analytics, these are the engine and fuel respectively of companies and they're really where things all come together and you make these critical connections. And so I just find that work so fascinating. And then I've found this in every area I work with, but I love walking into room. I love just learning constantly. I love walking into rooms and meeting people who I know have skills and knowledge that I don't have yet. And that knowledge sharing which I think just keeps things really exciting.

Sean Lane: Least favorite part about working in Ops?

Bridget Zingale.: Oh, boy. It can be stressful. I think the connections that I just talked about and how exciting they are, as there so often are disconnects, I think they are felt in the Ops and analytical arenas the most. And so I think that can cause a lot of frustration as you work things out. That's probably the negative, but overall, the pros vastly outweigh the cons for me on that one.

Sean Lane: That's awesome. Someone who impacted you getting the job you have today?

Bridget Zingale.: Oh, man. Our VP of marketing, John Dick. He joined, he was very analytically minded, which was great and part of the way I remember this happening is that I just would not stop talking about all of the improvements that could be made and then a month or two into his joining, he more or less was like, why don't you just run with this? I agree. And the switch to being full- time analyst was largely because of his vision for the marketing org. So that was a big step, which was great, which led us here.

Sean Lane: Hey, that's a great way to get a job.

Bridget Zingale.: Yeah.

Sean Lane: All right. Last one, one piece of advice for people who want to have your job someday?

Bridget Zingale.: Something that I look for when we're hiring is you have to love problems. You have to love really frustrating, messy, tangled up complex problems. And if you were the person that doesn't get frustrated too easily and will lean in and just fight, and honestly find it not only interesting, but enjoyable to handle those types of things and is incredibly curious, then you are already well on your way. I find that tends to be the biggest dividing line between someone who will stay in an Ops and an analytical role versus not is just, if you are the kind of person who just really loves messy problems. Honestly, sometimes in the interview process we'll toss out just weird questions, almost like riddles sometimes just to see whether or not someone can actually even answer it correctly, but just to see how they react. Just to see, oh, that's so interesting. I'm going to show you my thought process and just see how far I get. That's the mentality that I think is really needed for types of roles.

Sean Lane: Thank you so much to Bridget for joining us on this week's episode of Operations and thank you to Katie's and Golly for helping to set up the interview and make that conversation happen. Thank you, Katie. Also, want to send a special shout out to the data team at Drift who's doing all of the same work behind the scenes that Bridget and her team do at HubSpot, but they're doing it for us here at Drift. So Aroon, Akash, Mickey, Kyle, thank you so much for all the work that you do to set our organization up to be successful. If you like what you heard this week, please, please, please leave us a six star review on Apple Podcasts and if you haven't yet, please subscribe wherever you get your podcasts so you can get an episode of Operations in your feed every other Friday. Thanks so much for listening. That's going to do it for me. We'll see you next time.


Have you ever shown up to a meeting where you have one version of numbers to report only to find someone else in the meeting with a completely different version of those numbers? And then you spend half the meeting figuring out whose numbers are right? In this episode, Bridget Zingale, the Global Director of Analytics at HubSpot, is going to help us all avoid this all-too-familiar situation. Bridget and Sean talk about how the HubSpot analytics organization has evolved, what it means to promote data literacy at your company, and why the biggest risk in building centralized data teams is a departure from context. Like this episode? Be sure to leave a ⭐️⭐️⭐️⭐️⭐️⭐️ review and share the pod with your friends! You can connect with Sean on Twitter @Seany_Biz @HYPERGROWTH_pod