0:00:00.0 Michael Helbling: Before we start the show, we have an announcement and a request. Julie, you want to explain what we’re cooking up?

0:00:05.0 Julie: Sure. We are planning a listener call-in show. It’s an opportunity for you to submit your questions for one or more of us to respond to on a future episode.

0:00:15.7 MH: Oh, this is exciting. I mean, basically, this is a time for you, our listeners, if you’ve ever wondered why we never seem to discuss some specific question that you have.

0:00:24.5 Julie: That’s right. And logistically, we’ve made the process for gathering your questions as easy as it could possibly be.

0:00:30.6 MH: Ooh, by providing an internationally capable toll-free telephone number with a voice mailbox?

0:00:35.9 Julie: Okay. Well, almost as easy as it could possibly be. You just need to record your question with whatever recording mechanism you have on hand, and then email your audio file to advice@analyticshour.io anytime before April 12th of 2024.

0:00:51.3 MH: Okay. That’s still pretty simple. You know, and whether it’s a voice memo app on your phone, a one-person recorded Zoom, or any of a number of other options, we just need a reasonably decent quality audio recording that includes your name, where you’re calling from, and your question.

0:01:07.0 Julie: That’s right. Well, technically your question’s the only thing we truly require, but we’d love to have your name and location too. Oh, and as a reminder, we will be playing these audio files on our episode, at least the questions that we select for the show. So, our legal team has said we have to note that your submissions are in agreement for us to actually publish your audio as part of the show.

0:01:29.8 MH: Ah, lawyers. So, that’s it. Record your question, send it to advice@analyticshour.io before April 12, 2024, and hopefully we’ll be able to answer it. Anything I’m missing?

0:01:44.0 Julie: Just that listeners can visit analyticshour.io/advice to get a few more specific details on how to submit your questions.

0:01:51.3 MH: Okay. That’s useful. Analyticshour.io/advice before April 12th. Got it. And now, on to the show.

0:02:02.0 Announcer: Welcome to the Analytics Power Hour. Analytics topics covered conversationally and sometimes with explicit language.

0:02:10.1 MH: Hi everybody, welcome. It’s the Analytics Power Hour. This is episode 242. Ah, the passage of time, like little connected dots on a graph or something. And, you know, I’ve frequently said that one of the greatest things about analytics is the community of people around it. I mean, from my earliest days, back on the Yahoo Web Analytics forums to the many Slack groups of today, it’s just a lot of great people coming together to help each other on their way. And as with all things, some of them lose their meaning and go away or become less relevant to our journey. I mean, even this community, perhaps, that has formed around the podcast over the years. Well, we’re going to talk about it a little bit. So, let me introduce my two co-hosts. Hi, Moe Kiss.

0:02:58.6 Moe Kiss: Hey, how’s it going?

0:03:00.3 MH: Are you ready to get nostalgic about data communities?

0:03:02.5 MK: Oh, I’m actually feeling a bit emotional about this one.

0:03:05.4 MH: I’m too. I have a lot of feelings about this. I’m excited to talk about it. Val Kroll, how are you doing?

0:03:13.6 Val Kroll: I’m good. And if I make it through this whole episode without crying, I’ll be surprised. I’m just kidding. But also feeling very emotional for real.

0:03:21.8 MH: You’re taking a load off of me, so I don’t have to do that. So, good. That’s good. I’m Michael Helbling. And we wanted to invite another data traveler on this journey. So, we have Pedram Navid. He’s the head of data engineering at Dagster Labs. He was previously the head of data at Hightouch and has worked as a data engineer and data scientist at startups and large financial institutions. And today he is our guest. Welcome to the show, Pedram.

0:03:46.3 Pedram Navid: Thank you for having me. Can’t wait to make all of you cry.

0:03:48.4 MH: Oh, that’s perfect. Yeah, I mean, it is interesting. You know, this topic sort of rose to the surface in a blog post that you had written, I think, in the back half of last year where you’re sort of like, you had a blog post about how this all used to be fun. Analyst community. Do you want to talk a little bit about that? Like what was happening and what were some of the things you were thinking about at that time?

0:04:17.6 PN: Yeah, it’s funny. Just every now and then you get into one of these moods, right? It’s late at night. That’s when my blog posts are the best, I think. And I just have this thing that I got to talk about. And this blog post came really out of, I don’t know, just a reflection on where the state of the data community has gone for me personally. It’s not a reflection of how it is for everyone. I don’t think like this is a universal experience, but it did seem to touch, you know, some nerve of some people. And it’s just about how I think when I first started learning about data, the community seemed a lot smaller, a lot more welcoming. And through that, I think we were all sort of figuring this thing out together. There was no right way of doing data. Everything was kind of a mess. And it was really fun. So, we learned a lot from each other. And in doing so, like the people we would talk to and learn from they became my friends. And over time, I feel like the community has grown, as it does when things are successful. And sort of the vibe of that community has changed to a place where I just don’t recognize it anymore. It doesn’t feel the same as it was when I was like in the middle of it and enjoying it, I think, as much as I did earlier in my career.

0:05:33.0 MK: I need to ask because it’s definitely a sentiment that I share as well. And I do think size is partially, I guess, to blame. But I think I had always just assumed it was me, that I was different, not that the community was different. And that like, obviously, before I was in the weeds more, I was kind of doing the day to day analyst work. And so I guess there was like a lot more commonality or common ground, whereas like now, I find that I gravitate towards different things and that I’ve, I wouldn’t say I’ve stepped away from the data community, but like my engagement in it is very, very different. Is, do you think that’s fair or like have you had that experience?

0:06:17.4 PN: Yeah, it does feel fair. I don’t know what the cause is. Like, am I stepping away because it’s different? Is it different because I’m stepping away? It’s kind of hard to know, right? It’s true that like when you’re earlier in your career, maybe there’s just less that you know. And having these people that you can all kind of learn from and work together is really like invigorating in some ways. And then, you know, as you grow in your career, it’s often maybe less of that pure icy work. Maybe you’re thinking about different types of higher level order problems, and those aren’t the types of things you go to the community for. And so maybe the community changes because you’ve changed. I don’t know. I feel like it’s still different. I talk to a lot of people and they feel the same.

0:07:02.7 PN: I don’t know what people who are just coming into the community feel about it. Maybe for them, they have that same feeling that we used to have. It’s possible. But I think just like it’s just a numbers game. When you have a community of 3000, 4000, 5000 people, most of you get to know each other really quickly. And when that grows to 20, 30, 40, it’s impossible to really build a community that’s the same as what it was. So, I don’t know.

0:07:27.7 MK: I think there’s also an ability to like weed out the jerks faster when it’s 3000 or 4000 like when everyone knows each other pretty well and you touch base with someone and someone’s like, oh you should talk to this person they’re working on that problem, like those degrees of separation are closer and you tend to, I’m actually going to talk about Adam Grant a little later, but his concept of like givers and takers, you end up with like a culture where there’s more givers in the group and that’s the like predominant culture of the community and that kind of prevails because they are in the mass. Whereas I think once you start to get to like the 40,000, 50,000, yeah, like you’re potentially touching base with people you don’t know, people who might have stronger opinions, might have very different intentions like in the giver, taker side of the equation of like trying to get information or access to people or things or whatever, but without the same desire to give back.

0:08:22.4 PN: Yeah, if you want to get super pessimistic about it, we can do that. It’s also like because there’s value in the community, people who aren’t in the community now see the community as a place to extract value rather than to provide value. And so when no one cared about data, the only people who were in data communities were people who cared about data. When there’s now a lot of buzz and funding and money and capital, then it becomes a place to pitch your company and your startup and your product and to fish for ideas and to complain about vendors that you might not like. And all that stuff becomes much more prevalent, right? And my hope is that no one cares about data anymore sometime soon so we can all go back to…

0:09:08.5 VK: Take it back.

0:09:09.7 PN: Just being us.

0:09:11.3 MH: Or the original users of data communities, recruiters, you know.

0:09:15.6 PN: Exactly.

0:09:17.6 MH: There’s always one. We just had a little meetup in Atlanta about a month ago. And one person was there and was like, oh, I don’t recognize that name. Recruiter. You know, it’s like.

0:09:29.5 S?: Oh. Oh.

0:09:29.6 MH: Yeah. It’s fine.

0:09:30.1 PN: Recruiters are people too.

0:09:31.4 MH: I mean, yeah. As long as they behave like a person. But yeah, it was fine.

0:09:34.3 MK: I don’t mind recruiters coming to staff because I’m like, they often, no, I actually, I feel very differently about salespeople. That to me is not okay. Recruiters, I’m like, they actually are trying to help companies fill roles. Like, yes, I guess there’s an incentive or whatever. But like, I do sometimes find that like the right people are not finding out about the right roles. And so I’m like, if a recruiter can help solve that problem, I’m fine. And also, they should be paying sponsorship dollars to be part of that community in whatever format that looks like, whether it’s like at a meetup or a conference or whatever.

0:10:06.9 MK: Recruiters I have a very okay relationship with. People pitching products, that’s and I think, Pedram, you really hit the nail on the head there where you’re like, suddenly it became sexy, suddenly there was money, and I think that changed the dynamic a lot as well. And I also potentially there are people attracted to the industry now because, you know, it became the sexiest tight job title to have and, you know, a way to get a good pay packet and all that sort of stuff. Whereas before, like you said, it was the people that were really motivated and interested in the field.

0:10:36.3 VK: Yeah. Trying to figure it out together. That vibe you were saying, Pedram. Definitely felt that.

0:10:42.1 PN: Yeah. It’s changing, though. I think it’s getting better than what it was from the fever pitch we had in 2022. I think the people who have short attention spans and just follow the money have realized there’s not that much money in data after all. So, they moved on. They’re in AI now, and I’m sure they’ll find their next thing tomorrow. So, it has calmed down a little bit when it comes to modern data stacks and all that stuff. It’s not as wild as it used to be. But yeah, I think you’re right. It has been, it does feel different. And maybe that’s what happens to every community at the end of the day. I feel, I’ve heard this from people in like web development have expressed this exact same sort of feeling. And they’ve gone through this like we used to be a small group of people who believed in some type of thing. And then it became this in-recognizable force that sort of just took over. So.

0:11:34.8 VK: Or even like UX design, I felt like they kind of dealt with that too like everyone who did graphics or UI all of a sudden they were UX designers, right? And like that was like a huge bubble and influx into that community as well.

0:11:47.1 PN: Yeah. As soon as you put a name on something.

0:11:48.6 MH: Almost every subreddit that gets popular basically goes down this, on this journey.

0:11:57.4 S3: It’s time to step away from the show for a quick word about Piwik PRO. Tim, tell us about it.

0:12:03.6 Tim: Well, Piwik PRO has really exploded in popularity and keeps adding new functionality.

0:12:08.7 S3: They sure have. They’ve got an easy-to-use interface, a full set of features with capabilities like custom reports, enhanced e-commerce tracking, and a customer data platform.

0:12:19.5 Tim: We love running Piwik PRO’s free plan on the podcast website, but they also have a paid plan that adds scale and some additional features.

0:12:26.8 S3: Yeah, head over to piwik.pro and check them out for yourself. You can get started with their free plan. That’s piwik.pro. And now let’s get back to the show.

0:12:39.3 MK: I wonder if it’s a good time for me to tell a story that might make things a little bit less pessimistic.

0:12:45.9 VK: Yes, please.

0:12:46.5 MK: If you guys are open to it.

0:12:47.4 VK: Very much so.

0:12:47.5 MH: Yeah, yeah.

0:12:48.4 MK: And it’s funny because reading Pedram’s blog, I just was like, oh, I feel this so deeply. So, when I started in the data community, I was very lucky, of course, as most people are aware, both my sister and my brother-in-law are in the data community. And so introduced me to Measure Slack pretty early on. And it was a very small Slack group at that time. I don’t remember exactly how many, but it was very manageable to keep up with. And I was having issues one day with some of my R code. And I remember multiple people jumping on and someone who lived in New York city got on a Google meet with me and helped me troubleshoot through my code. And I just remember like, crying with relief afterwards ’cause I was working at a company where no one else knew R.

0:13:35.2 MK: I didn’t know anyone in the data community yet that I could like turn to for help. And that was like really the start of my involvement in the Sydney data community because I felt like if someone did that for me, I wanted to be able to do that for other people. And to this day, it’s like still why I work in data is like I think that we do have some of the most incredibly generous and smart and wonderful people that have this, yeah, desire to want to make it better and to like bring people along. And it’s it has been the driving force for why I love my career and why I’ve think it’s really important to give back. But I’m guessing we all have a little bit of a story like that that started us off.

0:14:14.6 VK: Yeah, absolutely.

0:14:16.8 PN: I don’t know if I actually have a story, but I have felt this like exact same thing. It’s why I joined like these Slack groups. Locally Optimistic was a really big one and the dbt Slack as well. Every data job I’ve had until probably this one, I was the only data practitioner at a company.

0:14:33.8 MH: Yeah, it’s a big one.

0:14:34.3 PN: Plus you have these startups. You’re often one day to hire and a team of 20, 30 engineers. And I’d always get jealous because they would have peers to review their code. I would like submit pull requests to myself and review my own code. Great job Pedram.

0:14:49.3 S?: Great job.

0:14:52.4 S?: That’s so diligent.

0:14:52.5 MH: I remember that.

0:14:52.9 PN: Just to feel like I was a part of the team, you know? And it’s hard ’cause like you have doubts about the work you’re doing. You want to ask questions. You don’t know if you have the right approach. And often you want to talk to someone and I’d pull in a software engineer and they’d have no context on like anything I’m doing. So, it was like not very useful. And then you’d go and find out locally optimized mistake or dbt Slack and you’d start asking questions. And there were people who suffered in the same way you’ve suffered and they’ve felt the same problems you have. And that was, I think, the beginning of a good data community. You know, thing for me. It was clear that there were people out there like me. And I think that’s the first thing we all kind of look for.

0:15:34.4 PN: And that’s what made data like so special, I think in the early days, ’cause I used to work at a big bank before and I did data there, but it was just like, whatever. We were just trying to like answer some questions that people asked. But when it became something bigger, something where we could like share best practices and learn from each other. And like you said, everyone I talked to, which is like offer up a wealth of information. It was very, very different. And now, I mean, that still happens. I still have people I can go to. But it’s just harder to, I think, develop that sense of community when you’re just too big, I think, in some ways, right?

0:16:13.8 VK: Yeah. There’s different things that you need from the community as we go through time and different things that bring the community together. I can speak specifically that my foray into data communities was specifically within digital analytics. And I don’t really know the digital analytics world without things like the Digital Analytics Association, because that was like so nascent to my early experiences, but we all have different needs. And I think that people start to identify with different parts of the community or roles get more specialized or they get changed and morphed. And so I think that that’s one of the other kind of causes that I see.

0:16:48.1 VK: But before I tell my story, do any of you guys watch John Oliver? Please, someone.

0:16:55.5 MH: Not regularly. I know who he is.

0:16:57.0 MK: Sorry.

0:17:00.9 VK: Okay. Well trust me there’s like this one segment where he says like and now this and it’s usually this like sizzle reel like making fun of like local TV newscasters like repeating the same catchphrases like, well when I played football back in college, and it’s like. So, it’s a really funny segment. So, I feel like when I tell like little tidbits of this story it’s like that you can make a sizzle reel out of this because the number of times that I say I almost quit digital analytics like it if it weren’t for the community because I had a very similar shared experience with Pedram that I was kind of like a lone wolf within my company, my department. And I thought I was terrible at my job. Like I was like, I am failing left and right, but it wasn’t until I got around other practitioners and having like some shared experiences, some shared trauma like we were just discussing to say, we’re going through this fire together like let’s help each other out and like let’s raise all ships. But that’s what kind of pulled me back in and made me feel not so alone and got really excited about it ’cause people were all solving problems in different ways.

0:17:58.2 VK: And it was similar challenges, but you’re trying to solve it in the context of e-commerce and you’re in B2B. Like, wow, I wonder how you grapple with these kinds of challenges or barriers or… Certain companies have more privacy risk or different things like that. So, it just became really interesting to dive into the way that other people were solving problems. And it was really fulfilling to know that like that giver taker concept you were talking about, Moe, that at some times in your career, you’re going to be more of a taker, and that’s going to ebb and flow, and the ability to give back is also super, super special. So, I’m very thankful for communities like DAA and Measure Slack and things like that to be able to engage with our people.

0:18:40.6 MH: Yeah, I definitely had a little flashback, Pedram, when you were talking about that, because I remember a project I was on where I was the implementer and the QA person.

0:18:51.3 VK: Account receivable and…

0:18:53.1 MH: I was trying to explain to my boss how this is not appropriate for me to be QA-ing my own work. It’s like, what are you doing? But it is interesting, because when you first are starting out, and I think you made a really great point also, like back in the day, and this is probably still true in a lot of communities and startups and things like that, but certainly even in companies and even large companies, there’s only one or two people that are doing analytics of any kind. And so just to be able to run into anyone else that shares your passion or interest or struggles, that was such a big deal historically. And with larger data teams forming, you can sort of form a community within the four walls of where you work sometimes now. And so that is a difference, I think. I also think as you progress, as we’re talking, one of the things I think a lot about community is people my age complain a lot about how all the data communities always just talk about how to use tools and this tool and that tool and this tactic.

0:20:01.1 S?: Oh, really?

0:20:02.1 MH: And yeah, because we’re old and grouchy…

0:20:06.9 S?: Back in my day.

0:20:06.9 MH: But also back in my day, but because the needs that sort of I would have from a data community are not necessarily, well, they’re actually pretty valuable to have all those tool-based conversations, because a lot of times you need to figure something out about something. And so you go do a search through a Slack group and you find some great little threads about it. But the conversations have shifted. Like, Moe, you lead a big team of people. You have different challenges and needs. And so the data conversations you’re having are, how do we create a great data culture around what we’re doing? Not necessarily how do we implement this particular piece of technology?

0:20:46.8 MH: And so I think that also changes it. So, place and career tends, I think, to also shift our viewpoint a little bit. So, when you’re coming up and learning, and I think Pedram, you kind of mentioned this too, like the kind of as in an IC role or individual contributor, you’re kind of pulling in information to be able to do your job. And when it gets a little more esoteric or a little more global in scope, you kind of are like, well, I don’t know if these things, but like I still get a ton of value out of some of the discussion topics in some of the channels. Like the Locally Optimistic channel has a data leaders channel, and there’s some amazing conversations that happen in there.

0:21:30.6 MK: Okay. I’m going to say something asshole-ish.

0:21:33.7 MH: Yeah, go ahead. Do it. Say it.

0:21:35.4 MK: I can’t keep up. Part of it, and I know part of that is size. I look at some people like in Measure Slack, and I’m sure we can all think of who they are, who were exceptionally thoughtful, who are constantly running back and helping people. I’m like, how? Where do you find the time to do this? Because I am drowning. I can’t keep up in my work Slack, let alone, I think I’m in like six other community Slack groups, and I struggle with like just the volume. And I don’t know, how do you get value? And like, soon as you said that, Michael, I was like, I’m in Locally Optimistic, and that is a channel I should be in, because holy shit, I have been thinking for so long about the fact that I need to be around another group of people that are like in the same place of their career as me and facing the same challenges. I need to be in that. And then I’m like, that is one more Slack channel that I need to find time to read every day. Am I just being grouchy? Like is that?

0:22:35.3 S?: Yes.

0:22:35.9 S?: Yeah.

0:22:36.3 S?: Yeah, you are.

0:22:39.7 MH: No, it’s a real problem. I don’t know. What were you gonna say, Pedram?

0:22:44.9 PN: It gets harder. I think part of the thing is, I think when you’re earlier in your career, you’re less sure of yourself. And so because you don’t have that confidence, you believe that you need to learn more. And you need to, like you have this imposter syndrome, and I like have to upskill myself and I have to learn all these things. I have to absorb all this information. Because everyone else knows everything, and I don’t know anything. And so that motivates you to read all the Slacks and all the blogs and try all the tools. And then you learn them all. And then you realize none of that shit actually matters.

0:23:18.9 MH: Or you start reading a blog post, you’re like, okay, this person doesn’t know what they’re talking about. Move on.

0:23:25.4 S?: Block.

0:23:25.5 PN: That’s another great skill to learn is to figure out when people are writing about things and they’re wrong. Took a long time for me to figure that out.

0:23:31.7 MH: Yeah, me too.

0:23:33.5 PN: Learning why people write is actually a really interesting thing. And looking at people’s incentives and motivations. Because when you’re young, I think you don’t see any of that. I think you just, and I’ve started to see that myself. Like I will get posts from people who are earlier in their career, and they share it with me. And they’re like, “Oh, doesn’t this look really interesting?” And I’m like, “Yeah, the vendor thinks the vendor’s tool is great.” What a shock. Have you tried something simpler? Maybe we don’t need to go down that rabbit hole. But they don’t see it that way. They just see this, what appears to be an informative piece. And it tells people what to do and what the right way to do things is. And if you don’t know better, you’ll just believe that because you don’t have the experience to do it. And I think as you age, as you get older, as you get wiser, if you will, you become better, or maybe more equipped to handle that stuff. And that drive, I think, does go away a little bit. You don’t feel the need to keep up with every single conversation, right? And maybe that’s part of why it feels different because now you are probably too busy managing a team of data analysts instead of being a data analyst yourself.

0:24:37.0 VK: I will say just really quick, back to what you were saying, Moe, on the keeping up. One thing that I did is when I went out on Mat leave, I actually turned off all notifications to everything just so that I could just focus on what was new in my life. But when I came back, I got really tough and critical about what was worthy of being pushed to me and not getting overwhelmed by notifications counting up in Slack. And so I set a lot of custom parameters for notifications or terms in certain channels that I want to be notified for. So, when I was really trying to develop internally with my team, what is our POV on SRM checks, sample ratio mismatch? I wanted to be aware of what were people saying around this topic to just help inform what we wanted to say or how I felt about it.

0:25:22.4 VK: And so as an example I allowed the notifications to come through when that term was used. And so I didn’t get overwhelmed by all of it, but I’m not getting organically the things that I wouldn’t think to look for. But I don’t always have time to jump in to look for that, but that was just one thing that worked for me. So, I just wanted to throw that tip out.

0:25:39.5 PN: So, the trick is to have kids.

0:25:40.9 VK: The trick is to have kids.

0:25:43.5 MK: To be fair, it does force you, I think, to be more thoughtful with the spare time that you do have…

0:25:49.8 VK: Totally.

0:25:50.2 MK: Because you will have less of it. But Val, okay, just for everyone out there, what are the other terms you have left? Because I did this for MMM and incremental. Incremental was a stupid word to tag because I thought of measuring incrementality and I’m like, I want to be in all those discussions. We obviously do use dbt and we use the word incremental build a lot. And so I started to get tagged in every single data warehouse message about anything. And that was a really stupid word. So, I’m curious what the words are that you use to try and manage which conversations to stay on top of.

0:26:30.8 VK: I think I try to just update it over time based on things that I’m interested in. So, after that was a topic that I was no longer interested in. Obviously, Google Optimize was a topic that I was interested in people understanding how they were going to grapple with that sunset and what their plans were next. And so I would say that it’s just mostly whatever I’m interested in, or maybe it’s related to some client work that I’m diving into. So, it evolves over time. Sometimes it’s tool specific, sometimes it’s not. But yeah, that’s a good one. An example of that’s a broad match. You’re going to get some false positives there. But I guess…

0:27:07.0 MK: Pedram, how do you keep up?

0:27:10.3 PN: I don’t. I don’t. I was just thinking about it and I realized I don’t. Let me open up dbt and just see. Yeah, it’s all bolded. Nothing.

0:27:18.0 VK: That’s overwhelming.

0:27:19.6 PN: Except for tools Dagster. Anything related to my company, I will look for, right? That’s an easy one. So, I work at Dagster. I look for things related to Dagster. People talk about it. That’s a very easy one because I’m a data tool and a data community. But the rest, I can’t keep up. I don’t use Slack, I think, for learning anymore. I use Slack for finding answers to questions I may have. So, it’s a great, sometimes a great search repository for those communities that have enough context. But outside of that, I think I just wait for it to hit me. There’s so much out there that I just believe that if something is useful, it will hit me in the face.

0:28:03.5 VK: It will find a way.

0:28:04.7 PN: That’s the job of the marketers, right? That’s what they’re supposed to be doing this whole time, right? So, I assume the marketers will be good at their job and they will find a way to make someone tell me about the thing if I should care about it. So, I mean, I have friends. I talk to my friends and that’s, I think, the number one source of information. What are they interested about? Yeah.

0:28:23.9 MH: It’s sort of like coming up with a routine that handles the signal-to-noise ratio in a way that’s manageable, really.

0:28:31.9 VK: Exactly. Well said.

0:28:32.9 MH: Because that’s really what’s happening. I remember, and in your article, you kind of made mention of this when Twitter, now X, kind of changed hands. It’s a crazy place, but I have it so heavily curated that I get exactly what I want out of it and it’s good. It’s fine. I have it right where I want it, but I’ve worked really hard over however long I’ve been on Twitter to…

0:29:02.6 VK: Train it.

0:29:03.5 MH: Like, get it that way. Yeah, exactly. Now, I guess the challenge is to do that with LinkedIn, which is…

0:29:07.8 VK: Oh, boy.

0:29:08.0 PN: Oh, boy.

0:29:08.7 MH: Yeah, exactly. I just can’t. I just can’t that’s why I just don’t even try. But the same thing with those communities is I have interests and things I’m… So, I skim a lot of stuff, but I think personality also has something to do with this. I’m huge on information intake. I’m always reading something, constantly absorbing information. I can’t stop. It’s actually a real problem. But I don’t make the assumption that other people come with that because I also have noticed over my life, back when there was a thing called Google Reader, I had a RSS feed of pretty much every analytics blog that existed in the world. And I would go to other people and be like, well, have you read? And I’d be like, well, here, I’ll share this with you. And they’re like, “How do you find all this?” I was like, “Well, I’m just interested.” And it’s like, “Oh, you’re not interested. Oh, okay. So, I’m the weirdo.” And that’s okay. I just started to learn not everybody’s as committed or idiotic as me to want to read about every possible thing going on.

0:30:14.7 MH: So, it’s a matter of where the personality stacks up. Okay. I want to switch gears. Let’s throw this in a different direction, which is…

0:30:23.4 MK: I do too, but I’m not sure we’re aligned on which direction we’re going in.

0:30:26.7 VK: I was just about to say, same.

0:30:28.1 MH: I’m going to go with the one I want to go down, and then we can see where…

0:30:31.5 MK: Okay. Alright. Alright.

0:30:33.4 MH: So, I think another thing I think about a lot today, especially around data communities and our industry, is the difference in who is in them back in the day and now who is in it now. And specifically, I think about it like when I got into the industry, the only people that were in analytics were all the misfits and weirdos who didn’t fit in in any other industry. And so we were all kind of a little bit odd. And now analytics is a real thing. You can actually study it in college and get a degree in it. And it’s a real career as opposed to a thing, no, you couldn’t explain to anyone else 15 years ago. And so people come out of college into analytics careers on purpose now, not by accident. And that wasn’t true 15 years ago. Nobody in the analytics 15 years ago or 10 years ago, even maybe 12, I don’t know, but started out with like, when I grew up, I’m going to get into analytics. Whereas now people legit do come into that. And I think it’s changing how we interact with each other’s communities as well. I don’t know. Let me get your thoughts on that.

0:31:47.5 PN: Yeah. I think it has because things are easier, right? There isn’t that like discoverability problem I think that we had earlier on. I remember podcasts were actually a great source of like this for me when I was earlier in my career back when I wanted to be a data scientist. And I thought that would be a really cool job to have. There was all these data science podcasts that they’re all defunct now. I think there’s only one left that I still listen to. And that was a great way to learn about how people are doing a state-of-the-art data science or what tools they’re using. But now, like you said, you learn that at school and everyone can learn what the best way of doing those things are. It’s kind of like an established pattern. It’s become a practice and it’s become something we understand the fundamentals of a lot better. It’s not perfect, but we know we should version control and we have these tools that we can use. And ggplot is better than Matplotlib.

0:32:41.6 PN: And these are all just truths that we all live by and we can just benefit from them. Whereas prior to that, we had to dig. It was almost like you were a DJ trying to learn what the latest songs that you wanted to play at a club were. And now Spotify came and just told everybody what the best songs are. And now you’re like, okay, well, my job’s over. I don’t really have to bother with that. I can just click play and then music comes. So, it feels the same way. There’s not that element of I got to go find things anymore. You just believe that good stuff will make its way up to the top because it’s good.

0:33:14.3 PN: And to be clear, we’ve gotten a lot of good things. There’s a lot of great tools that have come out in the last 5 to 10 years. You look at DuckDB, for example, or even dbt. You look at Polars. They’ve just came out of nowhere. There’s just wonderful open source tools. I don’t know why they’ve put their time and effort into making them great, but they have. And now we have them. And we can go and do things that previously would have required you to set up a Spark cluster and learn about the JVM and all this garbage that we didn’t really want to deal with. So, I don’t know. Things have gotten so good that it’s almost like you don’t have to go digging for how to do simple things that used to be really hard.

0:33:52.6 MK: Do you know what’s great? This is literally what I wanted to talk about.

0:33:56.3 MH: See, I knew it. I knew it, Moe.

0:34:00.4 MK: But I also feel bad because Pedram has painted it in a like, look at these amazing tools. And I do think there has been this incredible evolution. But one thing I did also pick up on is the monetization of the data community sometimes means that very good things also disappear or go down a path where people can’t use it anymore because they’re trying to find a way to monetize their product or whatever. And I’m not going to lie, I do find that quite difficult. I actually, and this is the saddest thing, I don’t remember it. There was a SQL tool that I was obsessed with and I adored it. I loved it. I worshipped it. And then literally they got acquired and they’re like, and we’re deprecating this bit of the product. And I just was like, and they just wanted to, they wanted to take the engineers. Like that was literally why they acquired the company. I think it was like PopSQL. Yeah, I think it was PopSQL maybe. But anyway, it was a SQL tool that actually was beautiful, which is why I really enjoyed it and very usable. But I feel like this does start to happen, especially in the like more open source side or like free products. And Pedram, I feel like you have a lot more kind of exposure to that world than I do.

0:35:09.8 PN: Yeah, it does happen. I mean, I think dbt is probably the best example for me of this tool that really like felt like it was built for me as an analyst and it solved all my problems. It spoke to me, they were developing for me and it was so great and I was so happy. And then they raised a bunch of money and they had to make money and they had to like, they had to run a business, which is very annoying, right? Like I wish they didn’t have to do that. And I understand that nobody wants to work for free. So, they did. And now they’re mostly on the enterprise side, a lot of their features are gated to dbt cloud on the enterprise side as well. And it sucks as like a practitioner, you’re like, I don’t need that product. I don’t need like enterprise is not the right fit for me at my one person data company startup, right? Like I just can’t use these things anymore.

0:35:56.3 PN: And also it feels like they’re not really investing in the core open source package as much, right? Which is fine, it’s their prerogative, but it still sucks. It doesn’t feel great. And so I don’t know, what do you do? Like people have to make money.

0:36:09.9 MH: Well, and it raises a question, like to what extent do the vendors in the space, is that their responsibility to kind of do that contribution? Or is it just sort of we all gathered around them and kind of laid it at their feet?

0:36:25.7 PN: It is like a plan now, which is unfortunate. I think when this first happened was it was, I think, this motion was maybe more accidental, right? You’d start open source, and you would hope you would find some type of market fit that you could like monetize. And then the dream is you continue to invest in both and then it works out.

0:36:46.9 MH: It occurred to me while you’re saying that I was like, Oh, yeah, we’ve had the dbt community manager on the show.

0:36:55.9 MK: But I think that’s the problem, right? Is, and I say this with all the arrogance of someone that works at a company that I think has done this quite well, in terms of we have an exceptionally good free product. And then we do have a paid product. But there is a very strong company culture that our free product needs to be absolutely exceptional. And that we don’t want to degradate our free product at the detriment of paid users. And I do wonder if maybe for some companies, they’re getting that balance completely off of like, they’re putting all their eggs into the and like you say, gating features like…

0:37:32.0 PN: Yeah. So, Dagster Labs also is in this model. We do open source and we have a cloud product. We invest very, very heavily into open source. Most of our engineers are working on the open source platform. Really a lot of the stuff that comes into enterprise, it has to come into open source first for it to even work. And so we believe very strongly in that. That said, we are a company trying to like sell a product. So, we do also have some things that only work on the enterprise side. I don’t think it’s wrong to do, to have a model that does that. I think it’s actually very healthy. ‘Cause if you look at other tools that are cutting in the space, people start open source projects out of love and passion and that becomes very tiring and very just ungrateful. People are ungrateful and they want so much out of you, they’re not gonna give anything back. It dies down and it becomes into sort of a shell of what it used to be. And so I think you do need a business model that can support these tools if you believe in open source. I think like you said as a company culture, I think it’s like do your founders actually believe in open source because they love open source and they think this is the right way to build software or is it a market strategy for them? And that’s really going to define how the product goes. Yeah.

0:38:44.9 MH: The cynicism will reveal itself at some point.

0:38:47.9 MK: I’m really wondering, I feel like the four of us have similar experiences maybe with the data community and/or at the, let’s politely say more senior side of our careers. What advice would you give a young person who’s starting to really get involved in the data communities around them? Particularly if there’s someone that’s passionate about continuing to drive the culture that we’ve all talked about that we love of like giving and sharing learnings and all that sort of stuff?

0:39:19.0 VK: Don’t be just a lurker. [laughter] And this is actually somewhat, my answer is somewhat tied to one of the things I would love to explore with you all too, which is finding local communities like in person. So, we’re talking a lot about like the online, and the virtual and the global, which is amazing and great and always at our fingertips. And I love your story Moe about [laughter] the person in New York, but finding that local group and even if you feel like you’re super junior, like raise your hand to be a part of helping shape that group ’cause there’s always gonna be a role. There’s always gonna need to be someone who orders pizza and there’s always gonna need to be someone who does the social posts. And so like raising your hand and getting involved like growing your network that way I think is how you’re always gonna make sure that you have someone to say like, oh you’re working on this problem. Let me connect you to someone else who I know just overcame that challenge or had a previous similar role that I think those shared experience would be valuable to you. Like you were talking about previously as well, Moe. So, investing in that local community and even before it feels comfortable like raise your hand to get involved ’cause that’s the best way to help the future of it is to be a part of shaping it.

0:40:28.5 PN: Find something you’re interested in too. I feel like if you’re passionate and interested about something then it’s so easy to get involved, right? Like find that part of the data stack you really care about, right? Whether it’s the tooling or the analysis, visualization, whatever it is. Find that and then you’ll find people who are also like that and just go out and do it. I think so many times I hear from people asking that question. It’s like, what do I do? And just the answer is just go do it. Like there’s no answer that I can give you that’s gonna help you on the steps that you need to take. I think what you need to do is just go do it. Like it should just be obvious to you that there’s something out there that you can do and go find a way to make it happen.

0:41:10.4 PN: Like if you’re not happy with something, if you think something’s interesting, you wanna go write about something, you’ve read something, just go do it. It could be posting every day on LinkedIn. It could be posting on Twitter, it could be writing a blog post, it could be going to a meetup, it could be contributing to open source. Doesn’t matter what it is, as long as you’re doing it, I think you’ll get there. The worst thing you can do is like just every now and then I get these messages on Twitter, on LinkedIn from like someone I don’t know and they send me like a three page essay about like advice they want on their life and career. And it’s like, I don’t know you, why would I spend the time reading all this? And like trying to answer the question for you. You need to do your own work.

0:41:55.4 PN: That’s the first step is like, go and figure all this stuff out and then come with questions that people can answer and make it easy for them. If someone else searches me on Twitter and is like, Hey Patrick, what do you think about X or Y Or like have you seen a great place to do this or stuff like that. I’m always happy to answer, but every now and then I get this person who just thinks that they can trauma dump their entire life story on me and ask me for advice on how to move their career. And it’s like, I’ve never met you in my life. Like I don’t know how to start to answer this question. So, I don’t know. Looking at yourself in terms of not what you can take from people, but more how you can contribute. And making it easy for other people to work with you and help you. That will go like so far and people will want to work with you and they’ll want to help you out.

0:42:41.5 MK: It’s funny actually, I sent an email to someone the other day about a particular topic and one of the things that I really recommend, if you’re reaching out to someone, maybe consider don’t asking them. Like don’t actually ask them for something. Think about something that you can offer them and edit your email. Like this email I had was probably like three full paragraphs and I went through it with a girlfriend because I was reaching out to someone who like, I really think is a fantastic person and I know their time, they get thousands of emails like this, their time is really precious. And I literally went through and cut 90% of the content because writing a short email that is really punchy is hard work. But ultimately the goal is to be respectful of that person’s time. Like you said, Pedram, if someone sends you a two page novel, like you are not only asking for their advice, you’re now asking for their time too to digest all of that. And that’s a lot. So, like I think if you’re reaching out to someone to ask them, like maybe start with like, what can I do to help that person first? And then over the course, like you don’t know, there might be an opportunity to ask for help back as well. And I’m not saying that you should offer it only to get something back in return, but it’s like I think you need to think about your intent there really carefully.

0:44:04.5 PN: And show you’ve done the work too, right? Like it’s one thing that you’ve hit a brick wall and you’ve tried and you don’t know where else to go and asking. Like I help people all the time with stuff like that career advice people ask me and I’ll give them my best opinion on them. But I can’t get emotionally invested in like four pages of your life story before I tell you whether or not you should go to get your master’s degree. I don’t know [laughter] I don’t know the answer to that question.

0:44:34.9 MH: Yeah. I think in terms of bringing it back to advice around data communities for people who are more junior in their careers, I still think they’re extremely valuable. And if you’re coming up in your career, the nice thing is there’s so many that they can match your areas of interest because and Pedram you kind of like pick an area you’re passionate about, but there it’s almost like you could find a community just for that. I think the Test and Learn Community, the TLC is such a great example in the optimisation space of a very focused, committed, passionate community just around experimentation and optimisation and the level of expertise in that community is amazing and also know what it takes to curate and build that kind of community. It’s a lot of work. And so there has to be some very passionate people involved to kind of prolong that and keep it going.

0:45:34.2 MH: And so get in there and participate. And I’ve always believed very, very strongly like you get out of it what you put into it. And I remember there was a forum for Webtrends, which was like the first software I ever used in analytics, which is an old log server analysis software. And they had a user forum and I made it my mission to help other people on that forum. Like I wanted to get good enough at this thing that I could help other people. And that actually opened the doors to tons of relationships and friendships and all kinds of things over the years just because I spent the extra time trying to like, not just get my questions answered, but get good enough so I could answer other people’s questions. And so I think that’s part of it is you giving is how you get from those things.

0:46:22.8 MH: And sometimes I think for us old heads, if you will, our ability to give is a little different or shaped differently. Like we’re giving in more contributions to our teams or how we’re structuring it. And so like communities aren’t necessarily our main way to give value. And so like maybe even our way of dealing with it is not as relevant. And so it recedes in terms of our necessity for it. But when you’re a junior in your career, just push in. And I think Val you kind of said that too, like with local meetups, be the person to organise it a little bit or send out the email like everybody needs that help and it’s hard work.

0:47:06.4 MK: And can I just caveat that?

0:47:07.3 MH: Yeah, yeah. Do it.

0:47:08.6 MK: Put your hand up for something and own it. Every single month I am sending the email every single month you offering to do it for three months and someone having to take the time to get you the login and teach you how to do it. And then two months later you go off on holidays and you decide not to send it. That is not helpful. Like it’s actually making someone else’s life harder because they then need to hold you to account. So, if you put your hand up to make a commitment, it’s about honoring that commitment and like even you can be like, I’m gonna put my hand up to do this thing for 12 months. And then in 12 months let’s have a check in and see if it’s working. But like if you put your hand up to own something, you own it.

0:47:46.3 MH: That’s a great point. And doesn’t sound like it’s born out of any personal experience whatsoever. [laughter]

0:47:50.1 PN: No, not at all.

0:47:52.4 MK: It’s so grouchy over here.

0:47:53.8 VK: I don’t think that that’s just for junior. Yeah, I think that that’s accountability from all levels for sure.

0:47:58.6 MH: That’s for everyone don’t flak out.

0:48:00.6 VK: And the one thing just to tie together from one of our earlier conversations of like we kind of self-selected into this because of not feeling fully fulfilled in previous roles that we had. So, there was like something that pulled us together. We’re coming to this role with a completely different set of strengths and weaknesses than the people who come to it from school. And so I also think that there’s like a lot to learn from each other just because we have different built up muscle and strength. And so like where we’re super more savvy potentially like making space for ourselves with stakeholders because we’ve been fighting for our position all along, but maybe we’re not as strong in some of the technical space. Like I’m definitely not speaking from personal experience. That’s where I can share swap stories and skills. And so I think that there’s lots of ways that you can pair up with different people to bring and provide that benefit. ‘Cause again, we’re different archetypes I would say of how we were drawn into this. And so there’s pros and cons to having those different strengths and I think we can learn from each other in those ways too.

0:49:02.4 PN: There’s actually a great article by local Sydney legend Claire Carroll on building a community called Starting With Why, which I would recommend to anyone ever considering writing or starting a community. It’s like the whole point of the article is like she gets these questions all the time. I think because she helped build like the dbt community back in the day and it felt to her, and I think it’s really true, a lot of people saw the success of dbt’s community and said I want that too. And the number one question you should always ask yourself is like well what is the benefit you’re gonna deliver to the community and find that? And then I think if you can do that then you’ll be successful. So, I think the same applies to, if you’re junior and you want to join a community, find out why you wanna contribute and maybe sending an email every month for 12 months is a commitment you’re not ready for. Just go in there and answer questions every day. Answer one question, right?

0:49:54.9 MH: Yeah, this is great. Alright, well we do have to start to wrap up but look at us. We did it and we didn’t even cry one time [laughter], so.

0:50:02.8 PN: I got close.

0:50:05.9 MH: I was ready. I was emotionally available to do that but it’s okay.

0:50:09.6 VK: We didn’t bring you there today, Michael.

0:50:11.3 MH: It’s not required. [laughter], it’s not required ’cause you can be yourself in this community. Alright. But one thing that we do do is we do stick to some norms around here. One of those is doing a last call, something that might be of interest to our listeners. Pedram, you’re our guest. Do you have a last call you’d like to share?

0:50:31.0 PN: I have two really small ones that I’ll put together. One is a Python package tool called UV. It’s PIP but like a million times faster. So, it just came out last week. I forget who. Oh Astral. Astral.sh is the name of the company that picked out. They also do Ruff, which is a very fast Python formatter. So, if you use Python, use that. And then my second pitch for you is a tool called LLM, which is a horrible thing to Google [laughter], but it’s from Simon W if you type dataset E-T-T-E-L-L-M, there’s ACLI. It’s just a command line tool that lets you write to an LLM, ask it a question while you’re in your terminal. And that may seem kind of boring but it actually unlocks a lot of really interesting things. When you’re working in a terminal you don’t want to open up a browser to ask a question, find something. You just wanna know a really quick snippet. How do I do this thing? You can do it right there while you’re in your terminal and you’re working and it changed my life. So, I highly recommend it.

0:51:36.0 MH: Nice.

0:51:38.0 VK: Pro tips.

0:51:40.1 MH: Very nice. Thank you. Val, what about you? What’s your last call?

0:51:42.7 VK: Yeah. So, mine is a medium article from the New York Times Open team and I read it a while back but it’s like just really stuck with me for how cool it was. It’s all about experimenting with handwriting recognition for the New York Times crossword. So, it’s like something that I think we could all have some experience with. But it talks about some of the challenges with like handwriting recognition and I learned all about optical character recognition. Like even some of the tests that they were running, if you think about some characters you have to pick up your stylus or your pen but you have to let it know, oh I’m not done writing this letter, I just need to cross that t or dot that I. And so just some of the challenges as they were going through it and the path to getting to a place where they thought it’s something that they could release. Just really interesting the process and the different experiments that they ran. So, more just like an interesting read but it goes pretty quick. So, highly recommend.

0:52:34.8 MH: Nice. Alright, well what about you Moe? What’s your last call?

0:52:40.7 MK: Mine is a little strange today. So, I think everyone knows this. I’m obsessed with Adam Grant. I am a very big fan. I read everything he writes, watch everything he records. I am all up all his books, everything all up in his jam. And he came to Australia to speak a while back and I had this inner turmoil of do I go or not. Like the tickets are not cheap, it’s a significant amount of money when you have like a set budget for what you can spend on your education in a year and like especially for someone that has such a strong online presence and a series of books like am I gonna get more value out of going to an in-person event than just absorbing what I could absorb online? And I decided to go and it was phenomenal even though some of the content was stuff I had already heard before.

0:53:29.5 MK: There was something about number one being in a room with other people that cared about the same thing as you. And number two, like not just trying to rapid fire, skim an article as you’re falling asleep in bed at 10:30 without the time to process. Like I was out of the office for half a day, I didn’t even bring my laptop, I refused to take notes because I wanted to be really present. At one point I picked up my phone and my girlfriend from work that was with me was like, Moe put your phone down. This was important to you. Be present. And I was like, yeah, it’s a good thing you’re here to keep me in check. And I did get so much value from it and it’s funny like weeks later I’m still processing a lot of the things that we discussed that day and how I wanna implement them with my team and the reason I bring this up is like I wanna encourage you for number one, like he deserves the money for the ticket because I do absorb all of his content. That’s my way of giving back and making sure that he keeps producing the kind of content that I wanna see and engage with. But also just to remind you that sometimes like in person is worth it. I know there are so many resources online but just being there with the same, the people that had the same shared goal is worth it. So, yeah that, like I said, it was a bit of strange one today and over to you, Michael.

0:54:45.3 MH: Well, mine is also a bit self-serving but still community related which is we’re going to do an advice show on the podcast and so we’re still collecting your questions and you can go to our website at analyticshour.io/advice and learn how to submit a question. Send us an audio file at advice@analyticshour.io and maybe we’ll cover your question on the show. It’ll be a lot of fun. And so there you go. So, that’s my last call just to keep plugging that so we get more questions from everybody. Alright…

0:55:21.9 MK: Can I say how excited I am to do those questions…

0:55:23.4 VK: Over the moon?

0:55:24.8 MK: Like give us…

0:55:25.0 VK: Cannot wait.

0:55:26.3 MK: Give us some curly ones. Yeah, give us some good ones.

0:55:27.1 PN: Dating advice.

0:55:28.6 MH: Yeah, well as long as it’s not four paragraphs in your whole life. No I’m just kidding.

0:55:34.7 S?: [laughter] Speed reading [laughter]

0:55:35.6 S?: That’s right.

0:55:36.4 MH: Alright, as you’ve been listening I’m sure you’ve been thinking, hey, yeah, I’ve got a comment or a question or I’d like to hear more about this. Well, we would love to hear from you and the best way to do that is through places like LinkedIn or X or the Measure Slack group where we’re really active and locally optimistic. I didn’t know Moe that you were on there. So, now there’s at least two of us and I know Pedram you’re on there ’cause I see pictures of your dog all the time. [laughter] and/or used to, I don’t know if you post them as often in the…

0:56:03.4 PN: It’s been a while.

0:56:04.6 MH: Pet’s channel. But that was actually a great channel and I loved keeping up with that one. Just a little shout out to that one anyways. But please reach out. It’s always great to hear from people and we like to get back to you as best we can. All right. And yeah, Pedram, thank you so much. Thanks for coming on the show.

0:56:24.3 PN: Thank you. This is fun.

0:56:24.4 MH: This has been great and a great discussion. And I think something that’s like good to think about is we’re a step back from the sturm and drang of the day to day in analytics and the talk about like the broader community. So, this is a good discussion and no show would be complete without a huge shout out and a thank you to Josh Crowhurst, our producer. Josh, thank you very much for all you do on the show. And I know that I speak for both of my co-hosts, Val and Moe, no matter where you’re at in your career stay involved in the community, get involved, do something and commit to it for at least 12 months [laughter] And no matter what you commit to, remember this, keep analysing.

0:57:10.0 S3: 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.

0:57:27.9 Charles Barkley: So, smart guys want to fit in. So, they made up a term called analytics. Analytics don’t work.

0:57:34.7 Kamala Harris: I love Venn diagrams. It’s just something about those three circles and the analysis about where there is the intersection, right?

0:57:45.4 MK: Rock flag and be a giver.

0:57:48.9 S?: I like it.

0:57:51.7 MK: Well, to be honest, I just realised with Five Seconds to go that it was my friend to do rock flag and I was mortified and I was like, screw you guys. You’re putting my name down [laughter] I don’t know who did this.

0:58:04.1 MH: I don’t do that. That’s probably…

0:58:05.1 MK: Yeah, I know it’s Tim [laughter]

0:58:08.3 MH: Yeah.

0:58:08.8 MK: I figured. I feel like he does it intentionally ’cause he knows how much I like viscerally hate it.

[laughter]

0:58:13.8 MH: Well, I mean culturally it’s the least associated thing to you too ’cause it’s like such a random one off.

0:58:21.6 MK: We should probably tell, Pedram.

0:58:22.9 PN: I have no idea what’s going on.

0:58:28.1 MK: You poor thing.

Podcast: Download | Embed

Subscribe: RSS