#192: One Dimension of Data Strategy: Approaching Data Work Strategically with Emilie Schario

Episode Transcript

[music]

0:00:05.7 Announcer: Welcome to the Analytics Power Hour, analytics topics covered conversationally and sometimes with explicit language. Here are your hosts, Moe, Michael and Tim.

[music]

0:00:22.2 Moe Kiss: So dear listeners, we have some really exciting news for you. In June, June 20 to 23rd, in fact, the whole Analytics Power Hour team will be at Marketing Analytics Summit, and we have one ticket for a lucky, lucky listener. So if you would like to grab that free ticket, please leave us a review on your favourite listening platform and then drop us an email at contact@analyticshour.io. Now, if you unfortunately miss out on this free amazing pass to Marketing Analytics Summit, then you’ll still be able to use the code APH15 to get a 15% discount. We hope to see you there.

0:01:02.0 Michael Helbling: Hi, everyone. Welcome to the Analytics Power Hour. This is episode 192. We don’t really have a problem buying up just like a sweet stack of modern data tools, and just like IKEA furniture, it’s sort of really tempting to just crack the box open and see how far you can get without reading the instructions or anything like that. But across a lot of data teams, the result of this “action-oriented” approach is that data stacks are assembled with a little to no documentation of what’s been built and why and maybe no purpose behind the tools and how they’ll benefit your organisation and maybe it’s just time for a little data strategy. Hey Moe, how do you feel about data strategy?

0:01:49.4 MK: I feel like I need all the help I can get. [chuckle]

0:01:52.1 MH: Yeah, it’s one of those ones that’s sort of like squishy, like you just grab it and it’s like “Oh, it’s gone.” Alright, Tim, what about you? You’re the quintessential analyst.

0:02:03.3 Tim Wilson: I think the quintessential analyst would know what data strategy is. I feel like it’s easy to define as data science.

0:02:10.4 MH: That’s me too, I guess. I’m Michael Helbling. So we needed a guest, obviously, one who could bring their experience in data strategy work to the show, and that person is Emilie Schario. She’s the Data Strategist in Residence at Amplified Partners, prior to that, she was the Director of Data at Netlify, she’s held data and operations leadership roles at GitLab and other companies and today, she is our guest. Welcome to the show, Emilie.

0:02:36.8 Emilie Schario: Thank you. I’m so happy to be here. Thanks for having me.

0:02:41.0 MH: Yeah, we’re delighted to have you here because obviously you can hear we need help understanding what is data strategy.

0:02:48.2 ES: And Tim needs to learn about all the negative parts of being a data scientist.

[laughter]

0:02:53.6 MH: That’s right.

[laughter]

0:02:56.1 TW: Yes.

0:03:00.7 MH: So I think we could probably start there or maybe talk a little bit about how you found yourself over the course of your data career, migrating into this area specifically, that might even be a good place to start.

0:03:09.9 ES: Yeah, I, like a lot of people, started as a data IC and I got frustrated pretty early on with the lack of impact I felt I was having to the business. And so I would run these analyses and ship these reports, and then no one would ever look at the dashboard again or I felt like I was drowning in these asks. And the more I talked to people, I realised it wasn’t just me. This feels like the cycle of sadness that most data professionals and most data teams are actually trapped in, is just not feeling like their work is really moving the needle for companies. There’s the stat from HBR that 85% of all big data investments are considered wasted. And I come back to the stat all the time because imagine 85% of all investments in anything being considered wasted. So as my career has grown and matured and moved forward, what I have really focused on is not just, how do I grow as a data professional, but how do we make sure we’re allowing our teams to be more impactful to the business? And for me, that has looked like growing into management and strategy style roles. For other people, I’ve seen it be being very technical senior ICs. Whatever the case, I think there’s a lot of paths to push and focus on impact to the business instead of really cool engineering work that we did.

0:04:48.3 MK: So I feel like exactly what you’re talking about is probably the thing that keeps me up at night. I’ve been doing a whole lot of thinking about how we as a data team talk about the value that we’re adding to the business, and I do feel like there’s often, like you said, this feeling of just answering constant tickets in a queue and you just get ticket after ticket and you answer them, and then the analysis goes out into the ether. What I’m, I guess not understanding quite yet, I think there are multiple different ways obviously you can show value and I’m really starting to try all different things and see what works, but I feel like you’re way ahead of me on that. So what do you think is the best first step? And do you think businesses get it, like the need for your type of role?

0:05:44.3 ES: I think that companies don’t need to have a data strategist per se in order to have an effective data strategy and actually say that it’s the responsibility of your head of data, whether that’s a manager or a director or a VP, it depends on your organisation and the specific context of the team, but like the, “How do we make sure our strategy is driving the business forward?” question, is a little bit different from, “We are drowning in reactive work, how do we flip that on its head?” And that is tackle-able at every level of the org.

0:06:19.7 ES: You don’t have to even have a data strategy in place, you don’t have to have a big picture or a manager. If you’re a single IC, a data team of one and you feel like you’re drowning in inbound requests, take an analysis, whatever analysis you’re gonna work on on Wednesday, and I want you to block off two hours on your calendar on Thursday, just two hours, it’s not forever. You can find two hours, I promise. You spend more time scrolling Twitter and taking pee breaks during the work day. Like two hours, block it off, make sure it’s time for deep work.

0:06:55.1 ES: If you are not a deep work person in the morning, block it off in the afternoon. Find these two hours. I want you to take an analysis you’re working on, wrapping up, and just shift it 15 degrees. So if the question that you were working on answering is, which landing page leads to the best pricing page conversion? Or what is the path that most people take to the pricing page? Or whatever it might be, I just want you to shift that a little and say, “And which lead sources are getting most people to that landing page?”

0:07:31.1 ES: Or, “Which attribution sources, which platforms are sending the highest quality leads?” Whatever it might be, take the analysis you’re already doing and just turn it on its side a little bit. And then instead of pushing out just the answer to the question you were asked, I want you to push out the thing you found. Sometimes the thing you found is, all of the data sources are the same and lead quality across Facebook and Instagram doesn’t matter. Sometimes no result is the insight, right? But you need to find something and push it out to the org, and then do it again the following week, and again the following week, and what you’ll find is the number of inbound requests coming to the data team will go down. And the reason for that is you are now suddenly, for the first time in the business’ history, pushing information to the company, instead of the company pulling information out of the data team. And that little switch does wonders for creating space for the data team to really drive and decide and prioritise the work that they’re doing. So it starts though with any one person just shifting an analysis 15%.

0:08:49.2 TW: Well, so… And that’s… I mean, there’s… You gave a very kind of prescriptive, which was a little bit, I think, just I’ll say think strategically, which maybe… I mean, strategy is like a tough word, but though one way to take that is, you as an individual contributor can actually stop and slow down and just fight the resistance, the urge as an analyst, we’re kinda… Tend to be wired to be people pleasers, somebody wants X on this time frame, and we want to give them X on that time frame, and so kind of I think there’re… Seems like there’re lots of ways to say, “No, step back, take a deep breath, make sure you’re really clear on what they’re really asking, challenge yourself to say, ‘Don’t think that they gave me the answer.’” ‘Cause we run into stakeholders coming and saying, “I need you to pull the top entry pages and the most comment passed through the site.” And it’s like, “Okay, you can do that, but it’s probably not gonna be that helpful, so they’re gonna hit you with five more requests and you’re gonna be spidering out forever and you’re gonna be drowning in tickets.” And so what I’m hearing, that turning 15 degrees, boxing yourself off, taking a deep breath, is like, “No, think a little more strategically. Challenge yourself to think and go look at something else.”

0:10:13.9 MK: The funny thing is, Tim, I feel like you have been talking about essentially the same concept for a very long time, which is like, the question they’re asking might not actually be what they wanna know. So you kind of are always like, “Take a step back, yes, answer the literal question they’ve asked, but then also answer the thing that you think they actually wanna know.” But I think the difference about how Emilie is describing it, which is why I really like this framing is that, the way, I guess, Tim, you and I have been discussing this is, you’re basically being like, “Do two things.” Right? Like, “Take the piece of analysis, which was the additional ask, then add this other thing, which you think is the actual ask.” The way Emilie is describing it, I don’t know if it’s by adding the 15%, but it makes it seem more achievable, because you’re like, “I just want you to do this one, extra, little add-on, versus essentially… ” And there’s something about it mentally that I think it could resonate much better.

0:11:15.2 ES: So I think there’s actually two different threads there, and it’s not an either or. So I wrote a blog post with my colleague from GitLab, Taylor Murphy, who’s now Head of Product at Meltano, called Run Your Data Team like a Product Team. And the TLDR there is, when we think about product managers and how they scope what gets built, people will say, “I want this widget,” but the product manager doesn’t… They work with UX research and they work with engineering and whatnot, but they don’t necessarily go out and build that widget. They step back and they say, “What is the problem you’re trying to solve? How do we solve that problem?” And I think that is absolutely something we need to be doing on every inbound ask. But the second piece is going beyond that and saying, “How do we reduce the amount of time that the team is spending in a reactive position and turn it so that we can get more time being proactive and working on more strategic priorities for the business?” And the way we do that is by shifting this pull of information from the company, from the data team to the company, instead where the team is pushing information to the company.

0:12:38.0 TW: I can hear people having had this conversation with people, ’cause there’s a part that is communication and a little bit of being like Jiu-jitsu and finding how to shift it. I feel like I hear time and time again like, “Yeah, that all sounds well and good, but I’ve been told I need to produce this, this, this, this, this, this and this,” which I understand, but it’s like, well, if you at least recognise that you’re in a cycle that you’re not gonna get out of like saying, “I can’t change it, so I’m gonna keep doing it,” that’s a shitty strategy. So just definitionally, you gotta try something different or it’s not gonna change. And it seems like that’s the… And it can’t be… It can’t be, “We’ll just work harder,” because that frenetic pace will just fill whatever time you give it. So I think that is this piece of saying, aggressively defend, maybe you can only defend 5% of your time to start, but make that 5% really valuable. Start pushing it to try to affect that shift to where you’re having people come at you differently. Although I still… I mean I’ve got specific analysts in my mind who I…

0:13:58.1 TW: This is such a… Their reaction to somebody… They’re just wired to more like, “Oh, they’re not loving what I’m giving them, so let me just find ways to do more.” And I wanna take those analysts and just grab them around the neck and say “This is not helping anyone. You’re stressed out, you’re not delivering value and you’re not thinking strategically.” So sorry, I just went to a… I think I just got triggered by my own thoughts in this discussion.

0:14:29.4 MK: Dark past.

0:14:30.9 TW: Those people drive me nuts, yeah. Oh, I wish it was past. [laughter]

0:14:36.5 ES: It’s certainly easier to be able to carve that time out if you have a supportive manager who’s onboard. As a… I remember when I first sat my team down in our Monday team meetings at Netlify and I said, “We’re gonna start carving out time to do this every week,” and it first started off small, like two hours, and then half a day, and then a whole day, and people were like, “But how?” Their faces were just this like, “What do you mean we’re gonna not do work? There’s so much work to do.” And being able to say, “I’m gonna run interference, I am your blocker, don’t worry about your stakeholders. Trust me here,” was, one, that is my job as a manager, if I wanna push this team forward to be impactful to the business, I need to be willing to get in the way of whatever is impeding my team, and in that moment, it was those stakeholders being like, “But deadlines.” There will always be more deadlines, right? There’s always gonna be the next super important project, and so if you don’t carve out time to do the proactive work, it’s never going to get done. And then the thing that we then started pushing this information out, and I was like, “We need a way to better organise this information.”

0:16:00.0 ES: So we at Netlify, we had an internal team handbook, it was a statically generated site, and I added a blog to it, internal blog. So you have to have an @netlify.com email address. And one thing I had noticed and kind of been a little frustrated by is every time someone would join the company, they would throw a time on my calendar and they’d say, “So Emilie, you’re director of data, what are all the things I need to know?” and I’d have the conversation with them, and then the following week, someone else would join and they’d say the same thing and I said like, “This isn’t working.” And so we started logging all these things that the data team was finding into this internal blog and we would write the blog headline as the takeaway.

0:16:41.9 ES: So you could scroll through the blog and just see these headlines that were like, takeaway A, takeaway B, takeaway three, and blue is better than purple, and yellow is gonna eat red, whatever the takeaways were. And what I found is that executives started refreshing this page multiple times a day. They wanted to know as soon as we were pushing information out, and that’s where I was totally convinced I was going to continue to run the interference because people were coming at me a lot less. I was able to create that space for my team to do that proactive work because the rest of the business was feeling the impact of that information going out. And so it does happen. It’s easier if you have that leadership for you, it’s certainly hard if you’re a solo analyst trying to fight the good fight. Team efforts are usually easier.

0:17:37.7 TW: I’m a big fan of kind of tracking what you’re doing and what the results are and having some sort of a catalogue, whether you call it a learning library or a hypothesis library or whatever, because to me, it seems like it’s very… It is a… That collecting data on your own self, like “How long am I spending? And what is it? And was there value in that really? And that’s… And again, I’ll watch analysts who say, “Well no, they need this weekly report. And like when was the last time that there was anything that came out of it?” “I don’t know, but they said that they need the weekly report.” I’m like, “Well, you’re spending four hours, so either shrink the time down or maybe you should figure out if there’s more… ” But just keeping track, I think as an individual or as a manager of, “Look, where are we spending? This is not… ” People have a reaction to tracking their time because they’re like, “Oh, you’re gonna tell me I’m taking too long for this?” It’s like, let’s just assume everybody’s doing a good job and is reasonably efficient. If we see that half of the team’s time is being spent on stuff that we can objectively say is a waste, then that’s useful. Let’s go figure out how we can just take 20% of that time and repurpose it, I think.

0:18:56.0 TW: Coming from somebody who eschews managing teams.

0:19:00.5 MK: My issue with the time tracking, and it’s not that I’m anti-time tracking, but I do…

0:19:05.9 TW: I’ll say time tracking and task tracking are related, but they’re not necessarily the same.

0:19:09.3 MH: It’s more time boxing, I think. This is not necessarily a time tracking.

0:19:14.6 MK: I think time boxing is maybe more important than… Yeah, because otherwise it starts to become like, “We spent X amount of time on this.” And it becomes about the amount of time you spent on it and not the value that was… Because if you were spending a lot of time on something, but it’s adding lots of value, that’s a good thing, right?

0:19:33.3 TW: But let’s make that separate from time and task just… But I don’t want you putting words in my mouth and saying that I’m like…

0:19:39.3 MK: I’m not intentionally putting words in your mouth.

0:19:40.8 TW: I know, that’s why I’m trying to not let you go on to say, actually having a list of what did we do? That is one piece, and if we put time against it as well, but there are times where it’s like, I don’t know where the day went. I just did… I had a guy who worked for me who was like, “The ankle biters, the ankle biters are killing me.” I’m like, “Well, write down what the ankle biters are.” And I don’t care… We’ll just assume every ankle biter took 15 minutes, but if you got 40 of ’em, then, yeah, we gotta figure something out. Something out on that, but okay, I’ve headed down a path. I just, I’m not gonna go die on a hill of time tracking.

0:20:16.0 MK: Please. Good. Now, I’ve gotta move on because, Emily, I’ve gotta get your thoughts… Okay, this is a really fucked up scenario from work, which thank God no one at my work actually listens to the podcast ’cause they’ll never hear this, which is great. We have had someone new start in the team who is coming in to basically build out a new forecasting model, and it’s really high value, we’ve mapped out a strategic plan for the project, and she said to me yesterday, so like yes, we have this section of the team that are drowning in tickets, and a big part of that is that we also do a lot of the data warehouse builds. Say something breaks in the data warehouse… Like you get in the cycle. And she was kind of like, “Look, the team lead kind of asked me to jump into some tickets and I push back,” and I was like, “Please push back, please push back,” because the second you start jumping… The second you get into those tickets, like really tactical tickets, you’re never gonna get out of it, and I really need you to do this really important project. But I walked away also feeling completely shit because I’m like, but someone else is going to have to do those tickets and maybe this isn’t fair from a team perspective or balancing the interesting strategic work or like I don’t know. I just… I walked away being like, “This isn’t the right solution, but I also really need to protect her time.”

0:21:42.8 ES: I think there is a people side of this as well as an impact to the business side of this. So it sounds from the way you said in the beginning that this forecasting project is the most important thing to the business that this person could be working on, so you probably did the right thing in coaching them to protect their time and fight to stay on what they need to work on. From a people side of things, you don’t always need to put the Michael Jordan on the team on the most important task. Sometimes Michael Jordan can pass the ball. And so it’s important that we make sure everyone gets the chance to take those projects, but we don’t need to split any given project. We can see this through completion and when the next project comes on up, you make sure that someone else is doing bug duty or slogging through the tickets that have come in.

0:22:40.7 ES: And so I’d like to handle these things in rotational ways where either everyone rotates through on call, or you do one day a week or whatever the motion that’s right for the team. Someone’s gotta do all those ankle biters Tim was mentioning, and I definitely don’t wanna get in the org in the habit of like, “Well, just ask the person you know best,” that is a terrible world. And so figuring out the best way is specific to your org in the context and there’s 40 million things to consider, but the one I would never lose sight of is, “What is the most important project that we need to drive forward,” and if the answer is that forecasting model, that’s the thing that needs to be worked on.

[music]

0:23:28.5 MH: Alright, it’s that time we’re gonna step away from the show for a quick word about our sponsor, ObservePoint. What’s it called when you embed an ad for something right in the middle of content that’s actually pretty relevant to the content in the ad itself.

0:23:46.0 MK: Contextual advertising?

0:23:48.8 MH: Oh, that’s it. I mean, it seems hard to imagine that an organisation could have an effective data strategy that doesn’t include plans for ensuring the integrity of their data.

0:24:00.8 TW: Totally, they’d be spending a lot of their time doing those ankle-biter tasks trying to get their data right, if the data is terrible. But ObservePoint is a great way to do robust, ongoing monitoring of the data collection for a website, like what tags are firing with what values along key user paths on the site, doing that 24 hours a day, seven days a week, and firing off an alert if and when a problem is identified.

0:24:26.1 MH: And beyond alerts, ObservePoint provides easy access to reporting that shows what those audits have turned up over time, which is a great way to measure the robustness of your QA processes.

0:24:38.0 TW: Plus, with the steadily increasing focus on privacy, ObservePoint’s automated audits can help ensure compliance to digital standards and government regulations for customer data, and that’s pretty handy.

0:24:48.9 MH: It is. So if you wanna learn more about ObservePoint’s many capabilities, go request a demo at observepoint.com/analyticspowerhour. Now, let’s get back to the show.

0:25:03.3 TW: So can I ask… We were talking a little bit before we started recording about, you’re at Amplify Partners, they’ve got stakes in many, many, many different startups, and you’re kind of the… The fractional data person who can come in and help get them going in the right direction. Many times I’m envious of younger companies because they tend to have less tech debt. If they grow too fast, they wind up in an absolute mess. But in the various companies that you’re seeing or working with, are you finding that you need to be helping them take a breath and do some educating about the data? Or is it because they kind of are naturally sort of product-focused, are they in a better shape with how they’re looking to use data? Am I… Is it wrong for me to be envious of younger companies with younger tech stack? Or what I also have in my mind is a bunch of tech pros in Silicon Valley just saying, “We’re just gonna capture everything, man, and throw it into the lake and then you can clear your way out of it.” And actually that doesn’t have enough thought to it, and it’s a mess. So what’s the reality?

0:26:30.3 ES: Both. All of the above.

0:26:33.7 TW: Okay. [chuckle] Okay.

0:26:34.8 ES: I would be envious of young startups because I think startups are a lot of fun to work at, independent of what role you’re in and their data stacks. I find start-up life to be particularly exciting. But what I see a lot of times with technical founders, which is what amplifies investment pieces around technical founders. And so that doesn’t always translate to, they know all the things about data that they need to know. Even some of the data tools founders, well, they’re phenomenal engineers, they saw a problem, they dealt with the problem regularly, and it frustrated them so much, they started their own company around it. Great, good for them. But that doesn’t mean they know the best metrics to measure success for your sales funnel, or the great way to set North Star Metric for their product.

0:27:28.0 ES: If you’re in B2B, then you don’t care about weekly active users per se, you care about active accounts, right? If you’re selling to a business, you care about the number of businesses who are active. And for you and me who are in data day in and day out, that might seem incredibly obvious. But if you’ve always spent your career working in Consumer Tech, and you’ve been frustrated and now you’re building your own tooling or technology, and you’ve started your own company, and now for the first time you’re in the B2B world, it can be really confusing. What do you mean, we’re not indexing on weekly active users? And so there’s a knowledge piece that almost has to come before the technology, and I spend a lot of my time there because that is much more impactful than, “Let’s get permissions on your snowflake instance right.” That’s important too.

0:28:26.2 TW: But knowledge… Knowledge, you’re saying… Knowledge of, think about the business, think about what makes sense.

0:28:31.7 MH: Business acumen.

0:28:32.9 ES: Exactly, yeah.

0:28:34.8 TW: Yeah. Okay.

0:28:35.2 ES: How do we measure and define success for our company? And data is important, it’s how we do that. But a clear strategy not to beat a dead horse here today, we should have started off with a drinking game maybe.

0:28:51.6 MK: I was about to say, “I feel like I need to open a bottle of wine with you and just be like… And talk for the next two hours.”

[chuckle]

0:29:00.9 TW: As Michael takes a big swig of his moonshine.

0:29:02.4 MH: That’s right.

0:29:03.0 TW: So… He is in Georgia and that was a Mason jar. So…

0:29:06.2 MK: But… Sorry. Emily, about this data strategy, right? I have written data strategies, I’m not gonna lie, they’re not my favorite things to write, I have read a gazillion data strategies. The thing… So we had Cassie Kozyrkov on the show a while back. And one of the things I was talking to her about afterwards was people from the business are kind of like, “You need to have a machine learning strategy.” And Cassie’s approach was kind of like, well actually, what you should be thinking is like, “Here’s a problem. Now, what’s the best way to solve that problem?” And you might find machine learning is one of the ways to solve it. Something that is much easier, like some, I don’t know, rule base something or rather might be a much easier way to just go about it. And my feedback was kind of like, “That’s cool and I get that. But I need to think about it from a hiring perspective, from a resourcing perspective.” I can’t be like, “Oh cool. We’re gonna solve this problem with machine learning now, and then we look around and we don’t have any machine learning engineers.” So I do need to have that strategy first, but then I can say there’s two things are in conflict, and I don’t know how to resolve that.

0:30:17.7 ES: I think the problem there is that people think that they write their strategy and they pinned it to the wall and everyone read it and great, we’re done. That’s the strategy, right? And instead, your strategy is constantly evolving, every day you’re gonna get… Every minute you’re gonna get a new input and you have to reconfigure all of it. And so an example that I like to point to… I don’t know, Moe, you’ll have to tell us if they also did this down under. But in the beginning of the pandemic, Netflix rolled out top 10 right before the pandemic, right? They rolled out top 10. What’s top 10? Count distinct order by descending, right? Not fancy algorithm, right? Not like what your neighbour is watching, it’s top 10.

[chuckle]

0:31:09.9 ES: And I think you have resource constraints that affect what you can and can’t do. Somebody forgot their laptop in the office and couldn’t go back to go pick it up, I guess. And so they’re like, “Top 10 it is.” I don’t know if that’s actually the story but it makes sense in my head. I think that the lesson there is that we have to figure out a way to do the most impactful thing we can with the resources we have. And the strategy isn’t about the specifics of how we’re going to address it. In fact, I’d argue that’s much more operational or tactical. The strategy is about understanding what are the problems we’re trying to solve at an org level, right? And that cascades into the operational plan, which is, “What are the head count we’re gonna have? What do we need to hire? What is the tooling and technology and the budgets and all that?” And then the tactical piece is every day, what are… What are the projects that Moe, Tim and Mike are gonna be working on over the next sprint or two or three, right? But your strategy isn’t about the machine learning problems that you’re gonna solve in the next quarter, it’s what are the business things we need to move, and then you figure out the pieces with the resourcing you have to get there.

0:32:27.4 TW: I do feel like that’s one of those things that can go in circles, and I think we’ve talked about it, I’m sure on the show before, when there are times where the business, the key stakeholders, the senior stakeholders really get their heads wrapped around what the… The “How” rather than the “What,” they’re like, “We need to do… We need to do customer journey analytics where we link together every touch point with the customer,” and it’s like… You try to say, “Well, wait a minute, let’s step back and let’s think about what we’re really trying to do, let’s not be in the hand-wavy world of theory, let’s talk about what problems we’re trying to solve.” So I think that winds up becoming a communication challenge of saying, “Look, I know that you want to do AI or I know that you wanna track every customer touch point,” and then you’re in these discussions of saying, “I know what you’ll want to do,” but there’s also the reality of what we can do like that… And so I feel like there’s that case as well, of where sometimes the business is just kind of living in the clouds and you’ve gotta say the strategy is actually trying to get a little more specificity around what is it we’re really trying to solve in the next one month, three months, and then go down from there.

0:33:53.0 MK: Okay. Alright, so… I feel like this is like everyone, coach Moe on how to do her job better at session again. So we talked about the fact that a strategy is not something you write, and you put in the shelf yadda yadda yadda because we’ve all experienced that, but I suppose…

0:34:07.6 TW: Well guess what consultants do. That’s good money.

0:34:11.3 MH: Alright, it’s time for the quizzical query, the Conductrics quiz, the conundrum that puts away your humdrum, I guess, because it’s always interesting. Alright. Moe, and Tim, are you ready?

0:34:27.4 TW: Ready.

0:34:28.8 MK: Definitely.

0:34:31.0 MH: Let’s get started, and before we do, let me just tell you a little bit about Conductrics. The Analytics Power Hour and the Conductrics quiz are sponsored by Conductrics. They build industry-leading experimentation software for AB testing, adaptive optimisation and predictive targeting. Go find out more about how they could help you in your testing and experimentation programs at conductrics.com, conductrics.com. Alright, Moe, guess who you are representing?

0:35:01.2 MK: Who?

0:35:04.0 MH: Till Butner, I hope I get his name right. He’s a good friend of ours.

0:35:07.3 MK: Excellent.

0:35:08.0 MH: And Tim, you are representing David Bissainthe. Bissainthe. I’m not sure how to pronounce his name correctly, but we may have met him before at Super Week, I don’t know. Alright, here is the quiz. In the TV game show, Who Wants To Be A Millionaire, contestants are given the option to use various lifelines to help them answer the questions that they are not sure of. One of these lifelines, the ask the audience lifeline, has each audience member vote for the answer they think is best, each audience member gets one vote and all votes have equal weight. The contestant can then see the audience vote share for each option and if they choose to can pick the one with the most votes. This approach to learning is similar to what was known as ensemble methods in machine learning, where several weak learners are all combined in some way to create strong learners, so they are kind of like Voltron of machine learning.

0:36:10.5 MK: Kind of.

0:36:12.1 MH: What is the name of the general ensemble approach that is most similar to ask the audience, is it A-stacking, B-boosting, C-bagging, D-facilitating or E-spanning.

0:36:27.9 TW: Oh, I think I might actually know this one.

0:36:32.3 MK: Yeah, I’m like, there is some familiarity for the first time ever.

0:36:37.5 MH: Okay. One might have some answers.

0:36:41.5 TW: Should we do at least an elimination a piece before we…

0:36:46.0 MK: But then I’m like, “Now that I’ve been overly confident, I’m probably gonna rule out the right answer on guess one,” but it is what it is, I’m going to eliminate facilitating.

0:36:56.5 MH: Facilitating, D-facilitating and yes, that is correct. That can be eliminated. That is not one of them.

0:37:03.5 TW: I will eliminate spanning.

0:37:06.7 MH: Spanning, E.

0:37:07.7 TW: Eliminating spanning.

0:37:08.7 MH: You eliminated correctly. In fact, there’s a little note that says E is made up.

[laughter]

0:37:16.6 MH: That gives us stacking, boosting or C-bagging.

0:37:20.0 TW: Oh, so you either go for it, there’s like… If you go for it, ’cause if you eliminate, then the… The power is with me.

0:37:28.4 MK: Yeah, yeah, you’re right.

0:37:30.3 MH: That’s true. If you think you might know it, I think you go for it. Yeah, good strategy.

0:37:34.6 MK: Well, it’s one of two, which is why I’m not sure.

0:37:37.6 MH: Oh okay.

0:37:40.7 MK: I am going to guess B, boosting.

0:37:45.5 MH: That it is boosting?

0:37:46.5 MK: That it is. Is.

0:37:48.5 MH: That it is boosting.

0:37:51.0 TW: I think it is bagging, so I’ll go ahead and…

0:37:52.8 MH: You think it’s C. Okay, so B or C, and so if one of those two is correct then we’ve got a winner, although… And what if it was A… And then neither of you…

0:38:00.6 TW: Then we’d be screwed.

0:38:02.1 MH: Well, we just…

0:38:04.0 TW: I think Moe and I both were thinking it was boosting or bagging and I’m pretty sure it’s bagging.

0:38:07.8 MH: The answer is C-bagging, which stands for bootstrap aggregation. Of course, in our game show lifeline, we don’t take bootstrap samples of the audience, but for the scope of the show, what the audience members each know is fairly independent of each other, sure there might be families or groups of friends, but the audience isn’t selected based on their knowledge of the topics, and we collect each of their predictions and give each of them an equal vote in the final outcome. One might want to argue for B-boosting. So one might Moe, and it’s similar to bagging except in both weight samples differently based on how hard they are to classify and gives the base learners that perform better a boost or extra weight in the final answer. In our millionaire example, each audience member gets an equal vote, which is maddening since there are always members of the audience that have no idea of the answer, but can’t keep themselves from voting anyway, and idiotically just randomly picking something.

[laughter]

0:39:06.9 MH: So that’s the answer which means that David, you’re a winner, Tim, great job, and that is the Conductrics quiz for this week. Big shout out to our sponsors, Conductrics, for putting together a great AB testing experimentation software, which you can check out at conductrics.com, and putting together the Conductrics quiz. Let’s get back to the show.

0:39:32.3 MK: One of the things though that’s in my mind, because obviously Cambay is such insane growth and things change daily, let alone hourly. And I understand if we take a step back and we look at the what are the problems we’re trying to solve, but I think the thing that makes it complex is like, number one, the problems change very quickly, but number two, every time we’re trying to get buy in, we have to educate the business, so it’s not just about changing the strategy because things have changed a little bit, it’s like, “Oh, now I need to go back and re-educate, why was the thing that we’re doing not the right thing anymore, and we should do this other thing.” And how do I… The education bit is actually the longer bit. That is a cultural thing that I just think is hard, and is the result of the fact that data literacy is still low.

0:40:36.9 ES: Yeah, low.

0:40:36.9 MH: Another very difficult to define, term.

0:40:41.7 ES: And measure and feel and… Yeah.

0:40:45.3 MH: Yeah.

0:40:46.3 ES: The one thing that I had at the top of my agenda every week in our team meetings at Netlify was the data team mission, and that did not change, that was not moving goal posts. And so I could say the data team exists to empower the entire organisation to make the best decisions possible by providing accurate, timely and useful insights, and that was our anchor point, and we could use that as the thing for, does this project or this problem to be solved, does it fit within our mission? No? We need to figure out a way to not do it, right? We need to not be distracted by all the shiny opportunities, we need to do the thing that’s going to push the business forward, and this is how we contribute to it through our mission. Sometimes the answer is yes, and it’s, we need to figure out how to drop everything else to make this thing happen, the most important. And sometimes it’s a team effort where you need all 11, 15, 25 people on it, sometimes it’s three of your Michael Jordans, and sometimes it’s not, but there’s no one right answer, but this, this, then growing and every day is different problem, I will say is my favourite part of start-up life, because it is so much fun, every day is hard.

0:42:14.0 MK: We can talk after the show.

[laughter]

0:42:18.1 ES: Every day is different, and I think about my two and a half years at GitLab, in that time when I joined the company, I was person number 282. I left two and a half years later, mid-pandemic, and the company was over 1300 people, and that whole time I was being stretched, I worked on projects around… I was in calls with investors, I was in pricing and packaging projects, I was in expansion into new countries and geographies, including China. All these really hard outside of my realm projects and being able to push myself and be stretched by the business, it wasn’t just about me, it was I had to fill the job ’cause the job had to get done. I think that is so fun and that is truly learning by fire in the best way.

0:43:15.0 MH: And I think a lot of times frameworks come into this as well, ’cause there is… Everything is changing all the time, but the types of problems generally will fit into a series of frameworks so that we can kind of say like, “Okay, that’s the category for this or what we expect, or what the business is looking for.” And sometimes that can also help bucket things and give it a sense of priority as a result, because it’s sort of like, “Okay, if we’re on a drive for new customer acquisition and growth than something that’s impacting, like loyalty and retention, while very important, is not necessarily the thing that’ll be the top of the list for the team to work on,” and that sort of like having frameworks like that. But I think that’s the two things when I think of data strategy or business acumen, understanding what the business needs to do, and empathy for the business itself to actually care…

0:44:04.3 TW: Easy there, easy.

0:44:06.6 MH: It requires empathy Tim. To be a great… If you wanna be a great product management team, which Emilie, I think you’re one of the advocates for treat your data team like a product team, product teams require a lot of empathy and not just empathy for like who’s complaining the loudest, which Emilie, I think you made this point earlier, it’s not just sort of doing whoever is coming in and banging the bossy pants the loudest, but actually having a fully engaged empathy for the entire user base to think big picture about what’s supposed to be done. So I thought that was a really great point as well.

0:44:38.8 ES: I think you see that too, as companies grow. A lot of times they will, for lack of better terms, bring in the adults, like they hire the people who have done the thing before, and…

0:44:49.9 MH: I don’t like it when that happens.

[laughter]

0:44:54.1 MH: Just because I’m never one of the adults.

[laughter]

0:45:00.9 ES: And the reason they do that, I think, is that they’re doing exactly what you said, they’re pulling from the frameworks of having done it before and what they can reference, and what are the patterns they’ve already seen that they can apply in this situation.

0:45:15.2 MH: Yeah, but like you said, I think that’s where fluency with data becomes a huge differentiator because usually the people with the frameworks have very low data fluency and people with high data fluency have not yet spent enough time engaging with the right level of the business to sort of have a nice transfer of understanding, and I think that’s where paradigms and data strategy become a really nice bridge, hopefully.

0:45:42.3 TW: What I was hearing is you were saying in the mission example, and if you kind of abstract having a clarity of purpose, which is like question number one. Are you really clear on what you as an individual or your team, and you can fall in the trap of slap dash, I may have some current examples of where like, “Oh, I was told I had to write a vision statement, so somebody wrote this and now we’re reminding people what it was.” I’m like, “But you didn’t put thought into it.” But put that aside. If there’s that clarity of purpose, it does seem like I’ve never done this, but now I’m tempted to run through… It could be reflection at the end of the day and say, “Let me just make a list of the things I worked on and how well did they align with that purpose.” ‘Cause the reality is yes, you will be required to do things that don’t align with it, but if that’s 50% of your time, then you probably are like, either you actually aren’t clear on what your purpose is, or you’re not spending your time on the right stuff.

0:46:48.0 TW: I’ve never done that, but now I’m tempted to say, Oh, what is it I’m really here to do? And can I break down and then ask myself, How did I get sucked into doing things that don’t align with that mission?

0:47:01.5 MK: That’s interesting Tim. That’s very, very interesting. Tell me more about how you’re gonna set some boundaries Tim? I’d love to hear more about this.

0:47:11.3 TW: I’m running through that right now and saying, Wow, that’s a small number. It’s like the number zero has some very interesting properties. Okay, the therapy continues.

0:47:25.3 MK: We did start every team meeting with it, in case you’re wondering, and I thought it was a great way to do that at a team level.

0:47:32.5 TW: But it’s interesting because I’m literally… Well, let’s just say I’m hypothetically… Let’s just say hypothetically, somebody I know, I have a friend who’s at a point where every one of his team meetings starts with that, and it is not particularly clear or focused or well thought out, it can have the… It can fall into the checking the box, ’cause it came from on high, you gotta put this in, you gotta start every team meeting with it as opposed to, “Wow, I’m watching my team make poor decisions on where they’re spending their time, maybe what we should do is have some real clarity and we should remind them and we should have a discussion. We should have people volunteer, everybody share the thing you did in the last week that was the least aligned with this, and let’s talk about how that happened or something.” I’m sounding like a consultant. This is shit I’ve never, actually…

0:48:26.1 MK: No, but it’s… Well, I don’t know, the only thing is like that could also reward you or become… I’m not gonna lie, my team would become a badge of honour competition of who can do the like worst to least a line thing for shits and giggles, but…

0:48:40.5 TW: Well, but that was… It’s actually interesting, ’cause we had May Allison a few episodes back and she talked about the data curbing where, I don’t know, it was once a week or every couple of weeks, they’d grab pizza at the end of the week and they would kind of have a… I don’t know that it was from the same reason, but that same… I wonder if the ability to say what did we do that really shouldn’t have done because it wasn’t aligned with our strategy, our purpose, our… And not to say in a perfect world, you wouldn’t have spent time on that, but let’s figure out how that happened and just surface it, which I guess all that gets back to that thinking strategically, just recognising when am I doing stuff that’s what I should be doing versus not.

0:49:24.2 MK: Another way to think about it is, if you think about it slightly differently Tim is how do we zoom out every once in a while to make sure we’re working on the right things. Right? And figuring out the right cadence, whether you do a quarterly goal setting or reflection or whatever it might be, or weekly or monthly, it’s really up to you, but I’d argue that businesses need to be doing this team, data teams need to be doing this within companies, but also we as individual team members should be doing this for ourselves when we think about our careers, when we think about where we’re trying to go and how we’re getting there and the progress we’re making, and our own ambitions, we should also be doing this exercise for ourselves for outside data, just in terms of, am I moving in the directions that I want to for my job? For my career, for my family, for my life, for whatever the things that are important to me?

0:50:18.7 MK: And so I think these are just kind of general principles and skills that also should be applied to data instead of just data-specific things. The irony and totally messed up thing here is that in my head, I’m like, how do I carve out the time to do this? I’m that person with the skeptical phase, even though I know that I tell other people to do this and I’m probably going to pass all of your sage advice, and I’m getting into this habit where I start to be like, “Is it the weekend yet, so that I can actually have some time to do the work that matters?” Which is honestly the most fucked up thing ever, and I am… I’m doing all the bad things.

0:51:05.6 ES: Tim, you’ll send Moe a calendar invite for two hours to block off two hours on her calendar, right?

[laughter]

0:51:12.9 MK: Emilie, this is the messed up thing our company has meet-free Wednesday, right? Meet free Wednesday. And so no one is allowed to put any meetings in your calendar until 2:00 PM on Wednesday, which is lovely, right. And that is intentionally there to do that, but what do I end up doing on meet-free Wednesday? I go through the 300 unread emails, I have to go through 50,000 Slack messages, I have to do performance reviews, I end up doing the ankle biters.

0:51:40.3 ES: That is the problem, you gotta eat the frog girl. You have to do the hard things first. What is the thing that you need to be doing every week that’s actually gonna move the needle for you and your team, and that needs to be where you start your day on Wednesdays. Definitely not email.

0:51:56.7 MK: I know, I know.

0:51:57.6 ES: There will always be more emails.

0:52:00.0 MH: You gotta give yourself permission to focus on the most important thing.

0:52:02.2 MK: I know. I’m just in a…

0:52:04.6 MH: It’s tough.

0:52:07.1 MK: Yeah, this is like I’m in a swamp at the moment, it is very swampy.

0:52:11.3 MH: The way I described that recently to a friend, Moe, and I don’t know if it’ll resonate with you, but maybe with others, is you have to get comfortable living next door to the understanding that you are definitely not going to get to everything on your list ever again. And now it’s just a matter of, okay, then if I can’t get to everything, what’s the most important thing for me to get to? And you have to sort of live with that, and it’s a tough transition to make, because especially people who are highly conscientious and really do a great job and value accomplishing kind of all the things they see in front of them. Eventually you reach a point where it’s like, “Nope, that’s never gonna scale that way,” and you’ve gotta learn a different way of managing it. And that’s not really like a data strategy thing, but it’s definitely like a head space thing, where you get to that meet-free Wednesday, and sort of like, “Okay, well, I’ve gotta clear this other stuff up because it’s on my to do list and it’s like I need to reorder my to-do list based on the biggest, most awesomest thing I can do.”

0:53:11.5 MK: I do wanna be clear, it’s not like I’m just doing emails and Slack, but the issue is also, I am doing that 15% stuff at the moment, but I’m probably doing the 40% of… Actually, I’m doing this work to push stuff to the business, because in my head, I know the strategy and I know that that work is gonna help us get there, but I can’t expect everyone else in the team to know and understand that unless I actually take the time and put it down in paper and explain to everyone like, this is where we’re going, and get their buy-in and their thoughts. So that’s the problem is like I am doing the pushing bit, which is great, but… Yeah, maybe you guys just caught me in a really bad week, which is also possible.

0:53:58.7 TW: I feel like we’re dicking each other out on that front.

0:54:03.7 MH: Yeah, and listen I’m no better. I go waste all my time answering emails and get reign of Slack messages and be like, “Oh, this person just emailed me on LinkedIn, I’ll go set up a call with them next week.”

0:54:12.2 MK: Oh I’m definitely not doing that…

0:54:14.5 MH: No, no. I’m the worst.

0:54:15.1 ES: I only clear out my email inbox on Fridays, it’s not uncommon.

0:54:22.7 MK: Tim? Me? I am not the worst in the world with emails.

0:54:25.5 TW: She clears it out on Fridays.

0:54:28.1 ES: Yes, I read emails when they come in, but right now I’ve got 30 or so emails that need some sort of more than sentence response. But Friday afternoon, the last two hours of my day when I don’t have energy to do any deep work, I’m gonna clear out my email inbox, and that’s the best productive way of Friday afternoon for me.

0:54:55.0 MK: I feel vindicated.

0:54:56.7 MH: Yeah, we should do that. You know what we do is we start to wrap up this podcast because we’ve got to…

0:55:04.2 MK: Can she come back? Can she come back?

0:55:07.0 MH: Yeah, of course, why not? [chuckle] Or I imagine probably, Emilie, you’re shortly gonna have your own podcast probably or something, so…

0:55:15.0 TW: We’re good with her coming back, it’s gonna be…

0:55:17.2 MH: Okay, hey, I don’t know. No, it’s been such a pleasure having you and talking about this ’cause there is a real world things that are happening and every data team is going through and struggling with these kinds of challenges, so very timely topic always. One thing we love to do is go around the horn and do a last call. Something that our listeners might find interesting that we read recently or looking at. So Emilie, you’re our guest. Do you have a last call you’d like to share?

0:55:42.0 ES: I do. So a little peek behind the curtain for the people listening, but before we started this call, Moe mentioned someone she knew who had been meaning to reach out, and my last call advice that I will leave is send the email today, like today before this episode is over, of like the person you’re looking to talk to or pick their brain or something. There is a phenomenal blog post that I will… Y’all will hopefully share in the notes on how to send a good cold email to someone you’re looking to get something from, written by Claire Carol, who has been a guest on the pod.

0:56:24.0 MH: Yeah, that’s right.

0:56:27.6 ES: And send the email today.

0:56:30.0 MH: I like it. Anybody who’s really looking to be brought down, you can reach me at… [chuckle]

0:56:36.9 MK: Okay, I’ve had this on my last… I’ve had this on my last call list for a while, and the really nice thing about this is that the author is here, so she can explain it in her own words. But basically I read this article by Emilie and I ended up sending it to the entire 160 present Data Team at Cambay, ’cause I was like, “This makes sense.” For the first time ever, I have had so many people try and explain to me what re-basing from Master is and just me somehow not getting it. And then suddenly Emilie’s article, there is no such thing as a non-technical data analyst. I just was like, chef kiss. That’s me doing a chef kiss. It was perfection.

0:57:22.7 ES: Okay. I have to cut you off. That is also Claire Carol’s example, not mine. I wrote an early example and Claire was like, “Actually, someone taught me what re-base is,” and I think it’s a better example. And so, she let me share that story. I think the sentiment is still there.

0:57:42.8 MK: How crazy! [laughter]

0:57:42.9 ES: But credit where credit is due, that is a Claire Caroll thing.

0:57:46.6 MK: Of course, she also is very, very brilliant at explaining things.

0:57:50.0 TW: There’s video of the two of you sitting next to each other on a panel, if I’m…

0:57:54.4 MH: That’s true.

0:57:56.5 MK: There is.

0:57:56.6 TW: Emilie wants to get there, Emilie and Claire, I think.

0:58:00.1 MH: That’s right.

0:58:00.6 TW: I will also say my son, I might have said this before, my son gave me a Pluralsight subscription so I could take a Git course that he thought was really good, that he had taken during an internship, so I learned re-basing from a son… My son’s Christmas gift, so we’re not a weird family at all.

0:58:17.1 MH: Alright, Tim. What’s your last call?

0:58:18.2 TW: So this is a… At this point, a pretty old Twitter thread, even as we’re recording it, it’s a while, it’s a bit old, but it was a tweet that started as one of those like, these are the books I’ve read, it was by Bojan Tunguz, @Tunguz, T-U-N-G-U-Z, we’ll link to it, where he said, “Hey, these are the books that I’ve read to do kind of everything I need to know about data science and machine learning. If you read these 10 books, you’ll be 98, 99% of the way there.” That’s fine. A pretty long thread, 99.9% male because of Twitter and that sort of post.

0:58:57.1 TW: But it popped up because actually JD Long had commented, because somebody midway through the thread said, “There’s like really nothing about statistics in this,” which then started a back and forth that got a little snarky, not complete Twitter assholery around, do you need to know statistics or not to do data science. And JD kinda came in on like, “Huh? Well, that’s… I can imagine doing data science.” So, we were talking earlier about the definition of data science, to me, it was still a very interesting thread to watch the different perspectives and the breadth of the space.

0:59:31.0 TW: It didn’t even cross my mind that there would be people who are very experienced, and very successful kinda making the case for saying, “Yeah, you don’t really need to know statistics to do this,” I don’t think I fall in that world at all. So it’s just a long Twitter thread with some pretty kind of funny digs in it as well, but I have not been able to let it go, it has now been several weeks, and I keep thinking about the thread, so it was thought-provoking. What about you, Mr. Helbling?

1:00:00.0 MH: Well, I’m glad you asked. So recently, I came across an online data leadership community called Locally Optimistic, which Emilie, I think you actually contribute to, which I… That’s not how I found it. I actually found it through a company called Brooklyn Data, which is a company somebody told me a few years ago, like, “Oh, you should look out for them, they’re kinda doing cool stuff in the space,” it’s as a small company, they’re not so small anymore, I think they’ve grown quite a bit. But anyways, I was randomly perusing some of their website, ran across their founder, and I think he’s involved from the beginning on this, so I ran across Locally Optimistic, which if you’re a leader or an aspiring data leader, that’s a Slack community, that might be a good one for you to join also in addition to the Measure Slack, but it is… It looks pretty good.

1:00:47.5 MH: I’ve been a member for a few days, and it just looks like a really awesome place, a lot of great perspectives and things like that. So that is what I would recommend. And I know as you’ve been listening, you probably have done things at your company to think about data strategy, to address the challenges that local data teams have, we’d love to hear from you. Probably the best way to do that is through the Measure Slack or our Twitter, and or you can also send us an email at contact@analyticshour.io. We’d love to hear from you. And I think, Emilie, you’re on Twitter as well, or you will be after Lent is over. I think you’re out for…

1:01:24.7 ES: Yes.

1:01:25.6 MH: You’re out for Lent?

1:01:25.9 ES: Yes, every year.

1:01:27.1 MH: Which is nice. Way to have boundaries, very nicely done. But you wanna share what your Twitter is?

1:01:32.3 ES: Absolutely. This episode will come out after Lent, so you can reach out to me on Twitter.

1:01:36.7 MH: Oh, that’s right. So there you go.

1:01:40.2 TW: Way to go, Michael. Such a professional.

1:01:40.3 MH: Yeah, I know. I literally start with the… I remember… Yeah, I always forget that stuff. Thank you.

1:01:45.0 ES: My Twitter handle is Emilie, E-M-I-L-I-E, Schario, S-C-H-A-R-I-O. So just first and last name, real original there. But actually a much better place to reach out to me is in Locally Optimistic Slack.

1:02:02.6 MH: See, there you go.

1:02:02.7 ES: I am very involved in LO, I love hanging out there. If you’re an IC, if you’re a team leader, if you’re thinking about, “How do I be more impactful?” It’s a great place to be. It’s all volunteer admins who work in data or data… Who used to work in data full-time, like Scott, the CEO of Brooklyn Data, he’s now a CEO of a data consultancy, but he was VP of data at Casper for a long time. And it’s a great place with wonderful conversations, and I do a much better job of responding to Slack DMs than I do answering email.

1:02:37.3 MH: There you go. Well, you gotta go join Locally Optimistic then. Awesome.

1:02:42.5 TW: Definitely. If they email you on a Monday, don’t expect you to get an answer before Friday morning.

1:02:48.2 MH: Okay, fine. But anyway, and of course, no show would be complete without a big thank you to Josh Crowhurst, our producer, doing what he does to help get the show out to you. Alright, Emilie, once again, thank you so much for joining us, being our guest today, we really appreciate it, and we love the insights.

1:03:07.7 ES: Thanks for having me.

1:03:08.8 TW: Awesome. And fighting through the technical difficulties, which we will… It’ll be a challenge, Josh will… Josh, you earned your pay on this one.

1:03:18.5 ES: I didn’t wanna be a boring guest. [chuckle]

1:03:21.1 MH: No, not even for a little bit. Well, no matter what troubles you’re running into, whether you’re reactive or super proactive, remember, I know I speak for both of my co-hosts, Moe and Tim, and when I say, keep analysing.

[music]

1:03:39.9 Announcer: Thanks for listening. Let’s keep the conversation going with your comments, suggestions and questions on Twitter at @analyticshour, on the web at analyticshour.io, our LinkedIn group, and the Measure Chat Slack group. Music for the podcast by Josh Crowhurst.

[music]

1:03:56.3 Charles Barkley: So, smart guys wanted to fit in, so they made up a term called analytics, analytics don’t work.

[music]

1:04:03.2 Tom Hammerschmidt: Analytics! Oh my God, what the fuck does that even.

[music]

1:04:12.3 MH: And action.

1:04:14.3 TW: Why is there background noise?

1:04:15.7 MK: I was gonna say, is the rain so fucking loud that I need to move?

1:04:20.5 MH: I guess the rain is down in Australia.

1:04:27.3 MK: I know I’m being a massive pain in the ass, but is the rain okay now? The only other option is I go downstairs, but then it’ll be very echoey. The only caveat is that is has stopped raining, but if I go downstairs, it’s wood floors, whereas at least this is carpet.

1:04:42.0 TW: We call that… We call that confounding, if you’ve made two changes. If you both stopped the rain and changed your location, we cannot…

1:04:51.7 MK: Thanks, Tim.

1:04:52.6 MH: I’m not sure if you’re familiar with a concept called causal inference, but…

1:04:56.3 TW: All right, let’s keep going.

1:04:58.3 MH: Okay.

1:04:58.9 TW: Thank you for the address. Okay, so, Emilie… [chuckle]

1:05:02.5 ES: I am trying to keep my shit together here.

1:05:08.6 MH: But I am gonna do something like over recording shit.

[laughter]

1:05:14.2 MH: All right, let’s get started…

1:05:15.6 TW: Tell me more.

1:05:17.0 MH: No, no…

1:05:19.4 TW: Rock flag and eat the frog.

[music]

2 Responses

  1. Javier Morales says:

    Any chance you guys will post the cold email article Emily mentioned?

Leave a Reply



This site uses Akismet to reduce spam. Learn how your comment data is processed.

Have an Idea for an Upcoming Episode?

Recent Episodes

#257: Analyst Use Cases for Generative AI

#257: Analyst Use Cases for Generative AI

https://media.blubrry.com/the_digital_analytics_power/traffic.libsyn.com/analyticshour/APH_-_Episode_257_-_Analytics_Use_Cases_for_Generative_AI.mp3Podcast: Download | EmbedSubscribe: RSSTweetShareShareEmail0 Shares