Resources
Feb 4, 2022 |
1h 17min
|
868 views
|

Data Science Hangout | Ian Anderson, Philadelphia Flyers | Moving into Leadership & Managing a Team

The Data Science Hangout is a weekly, free-to-join open conversation for current and aspiring data science leaders.

posit.co/data-science-hangout

We were recently joined by Ian Anderson, Director of Hockey Analytics at the Philadelphia Flyers, to discuss the most important things going in data science leadership.

One of the topics that there were a lot of thoughts on during this hangout was balancing that shift into leadership and maintaining technical skills.

Question:
š€š¬ š²šØš® š¦šØšÆšž š®š© š­š”šž š„šššššžš« š¢š§š­šØ š„šžššššžš«š¬š”š¢š© ššš§š š¦ššš§ššš š¢š§š  šš š­šžššš¦, ššš«šž š²šØš® š°šØš«š«š¢šžš ššš›šØš®š­ š„šØš¬š¢š§š  š²šØš®š« š­šžšœš”š§š¢šœššš„ šœšØšš¢š§š  š¬š¤š¢š„š„š¬?

Here are a few of the thoughts shared live from Ian and the community:

* You make time for things that are important to you. If exercise and health is important to you, you make time for it. If coding is something that you want to continue to maintain, you can find the time for that.

* Think about blocking designated focus time to do your individual work

* Be curious. Office hours as a manager to answer questions from your group can also be an opportunity to learn new things/packages/methods together

* Organize opportunities for the team to share any topics that they would like (coding, techniques, etc.) In sharing topics, you can also try limiting the time to ~4min to share. Two benefits of this fast moving, high energy meeting: 1. People will inspire each other with passion and creativity 2. The team gets to practice communicating their ideas with precision.

* Becoming more comfortable with what it means to lead people, it's about enabling them to be really successful. If you measure your success by other people's success, it can also help rationalize that. If you can chop down barriers for them, then they will be more successful so that means the technical time I sacrifice enables them further.

* It's still very relevant to keep your skills at some type of level just for street cred in your team. Bringing in new people, if you can talk that language it just really helps your team and you can have those conversations."

ā–ŗ Subscribe to Our Channel Here: https://bit.ly/2TzgcOu

Follow Us Here:
Website: https://www.rstudio.com
LinkedIn: https://www.linkedin.com/company/rstu...
Twitter: https://twitter.com/rstudio

image: thumbnail.jpg

Transcript#

This transcript was generated automatically and may contain errors.

Thank you all so much for joining today and welcome back to the Data Science Hangout. Happy New Year to everyone. I'm so excited to kick off the first one of 2022 with you all and to have Ian Anderson joining us here for his first time as well. But just to introduce the Data Science Hangout for anyone who's joining us for the first time.

The Data Science Hangout is an open space for the whole data science community to connect and chat about data science leadership, questions you're facing, and really just what's going on in the world of data science. So if you ever want to go back and re-watch these or some of the ones that we had in 2021 or share with someone who's missed it, they'll be recorded and shared up to YouTube as well. We really want this to be a space where everybody can participate and we want to hear from everyone. So there's a few different ways you can ask questions during the session today. So you can jump in live, you can raise your hand on Zoom, put questions in the Zoom chat. Feel free to put a little star next to your question too if you want me to read it out loud. Otherwise I can just call on you to introduce yourself and jump in.

But we'll also have a Slido link too where you can ask questions anonymously and I think Tyler will share that in the chat in just a second here. But without all that I'm so happy to be joined by my co-host for today, Ian Anderson. Ian's the Director of Hockey Analytics at the Philadelphia Flyers. And Ian, I'd love to start by maybe having you jump in and introduce yourself and share a bit about your team and the work that you do.

Hello everybody, first of all, and thank you for having me. And I'm also excited to be here and talk about all things data science, sports, flyers, whatever people want to do. Basically, I've been with the Flyers now since 2014. Had a variety of different titles or roles, I guess, over that time. I guess I don't know how deep you want to get into it or if that was one of your questions if I'm jumping ahead, but I can expand on what I've done. Started as a kind of started as a kind of team of one. I started in 2014 in hockey. I don't know who the participants are today and what their experience level is with the sport, but I was the third full-time analytics staff member in our sport hired at that time. And then there was a rash of hirings during the 2014 summer. But now that's widespread across the sport. So I'm expanding my team currently. We'll be a team of six very soon. So that's the general overview, I guess, for today.

That's awesome. Thank you. And yeah, as we're waiting for some questions to come in from everyone, I think that would be really helpful to expand on how you kind of started this small team and how you've grown over the past few years too. Yeah, yeah. So a little bit of my backstory too is I was working for the Washington Capitals in the NHL in a different role starting in around 2007. So 2007 to 14, I was in more of an operations role embedded within the team. And I just kind of noticed that we didn't have anybody at that time during those years doing this type of work. And it was being done in other sports. And I just kind of realized, like, maybe this is something we should be doing. And I kind of asked. They said, sure, you want to do it kind of thing. And I was like, well, I don't know anything about this, like, technically speaking, right, how to code, or I didn't go to school for that. That's not what I did. So another piece I'll share is I'm a self-taught programmer. I got my master's through Northwestern's Master's of Predictive Analytics, it was called at the time. I believe they changed the program since then. But there might be some people on this call that are from that program as well. Taught myself how to do it. And then, again, that led me to the Flyers. So that's kind of my backstory. But I don't know how much more you want to get into that.

That's amazing. So you just said, hey, we should be doing this. You started to learn it yourself. Yeah, I mean, I talked with somebody. And I was like, maybe I should go work in baseball. And the person was like, wait a second. So there's nobody doing this in the league. There's nobody with these skills. Like, why are you trying to leave? Like, you should be doing that there. It's kind of the advice I got. And I was like, you know what? That's a good piece of advice. So I just did some research. And back in 2014, there weren't a lot of programs, too. That's another thing I wanted to share with people, is like, we're here. We just started 2022. And one thing that's amazing to me about this industry, and especially about sports analytics in general, is like, it wasn't that long ago. Like, we all lose track of how fast this is moving. And eight years is not a huge amount of time. It is in some cases. And in some cases, it's not. And so to see kind of how things have expanded and going through the hiring process now, seeing the applicants and the level of their skills over the eight-year period is like, it's exponential growth. So again, I'm getting ahead of myself. But I'll leave it at that.

I imagine there's a few people on the call, too, who might work in industries that are pretty new to using data science and could possibly be in a similar position. So I'm curious if you have any tips for going about starting that program up or convincing people that you need it.

Yeah. I think what I will say is I have a lot of experience on this. Again, I got dropped into an organization with no infrastructure, no data pipeline, no anything, right? It was just like, here you go. You figure this all out, one person, right? And so what you realize very quickly, as I'm sure everybody here can attest to, is you can't do this by yourself. It's a team. This is all done by a team of people. And anyone that is doing it by themselves, my hat goes to you because it's a lot to be knowledgeable across all areas, right? So I guess my piece of advice I give people because I do have people talk to me, various teams, and are different stages of developing their program, right? Is I just tell them that what has worked for me is just creating a very small reproducible example of work or starting very small with a project that you can deliver, right? And even if it's not the best or what you would consider the best, just give people something that they can see and touch and feel, right? And then they get to start using or seeing how it improves their lives. Like sometimes the products that we make is just making our coworkers' lives easier and automating some tasks or things like that, right? So do something that people could see that makes their life better that they enjoy. And then they say, wow, this is awesome, great work. And then you say, okay, you like that. Like here's where we want to go next. And here's why we need to expand the team because we could be doing more in XYZ, these areas, right? So instead of trying to do like everything all at once is just like start with one small thing and then build on that, like kind of like taking a long-term approach, I guess would be the best.

What has worked for me is just creating a very small reproducible example of work or starting very small with a project that you can deliver, right? And even if it's not the best or what you would consider the best, just give people something that they can see and touch and feel, right?

Breaking into sports analytics

That's super helpful. Yeah. Zach, I see you're kicking off the questions in the chat and I'd love to turn it over to you to have you ask it. Okay. So first of all, I don't know much about hockey. I've never watched the game, but I do follow the NFL and I'm sure you're aware that they do the Big Data Bowl submissions today, which I have submitted a project if you're interested in seeing.

But my question more about is I see that Megan, I don't know how to pronounce her last name, which is really bad, has a Big Data Cup in March, end of March. And I was wondering if like there's other hockey equivalents of the Big Data Bowl where you can get like normal data that's hidden away from the public that's available to see what the public can do with it. Yeah, this is something I'm extremely interested in, keep track of myself, obviously, because the tracking data that you see in the Big Data Bowl that the NFL is doing, we have, our league has been working on the exact same technology for years. And it's finally this year, so this October of 2021, was rolled out league-wide. It's in every game, every team, everything in the NHL now. So that being said, we're obviously a couple of years behind the NFL, but I could see the NHL potentially doing something similar here in the next year or two. So I'd be on the lookout for that. That being said, there was a Big Data Cup, is what it was called, playing on that same idea run by a company called Stathletes, which is a data provider to various different hockey leagues. And they did their first one within the last few months. I don't remember the date, so I apologize. But within the last, like say, six months, they did something similar where they released their data that they have, their company produces, and they asked people to create a project similar to the Big Data Bowl. So you guys can Google that, and everyone can figure that out themselves offline. But again, that's year one, right? They've only done it one time. I imagine they'll probably try to do it again, and it'll expand and get better. So anyone that is looking to do hockey work similar to Big Data Bowl, the NFL is rolling out, it's coming. I imagine it'll get better and better. So that's my take on it. Yeah, we're talking about the same thing. Yeah, Megan's one of the co-founders. Yep, there you go. Yep. Yeah, perfect. Thank you. Awesome.

Thank you. Ian, I see there's an anonymous question, and this has come up in some of the R and Sports Analytics conversations too. The question is, sports teams feel like such an interesting but difficult type of organization to break into. How do you get into that world, and what's so unique about the industry?

Yeah, great question. I have a ton of mixed feelings on this question. Again, when I started years and years and years ago, people always, the first thing they want to ask you, which is a relevant question, I understand why they're asking is, how did you get in? But the problem I have, and it always makes me cringe answering the question and trying not to avoid it, is this is a totally different world today than what I got into. So my story is so irrelevant now that if I told you, it's not reproducible, to use a data science question. I can't reproduce what I did then, now. It's not possible. In saying that, the piece of advice I would offer people today, and the benefit that they have today, is that there are all these great programs like Big Data Bull, Big Data Cup, Cagle, expanding, obviously. There's so many different platforms, even RStudio, all the different tools that have been developed for people to showcase their work. Shiny apps, you name it. There's so many tables, GT table contests. There's so many ways to build a portfolio today that there wasn't even a few years ago. So it's all out there for you to build your own portfolio. And like I said, this is, I'm passionate about this because I'm literally in the middle of hiring right now, and there's just no excuse not to have a portfolio now. There's so much you could be doing. So I would just tell people, do the job you want, and that starts now. Like, you could be doing the job now, and it's going to require some extra attention outside of work hours, maybe, and things like that, but it's all doable. So that's easiest piece of advice, I guess.

There's just no excuse not to have a portfolio now. There's so much you could be doing. So I would just tell people, do the job you want, and that starts now.

Hiring and building a portfolio

Thank you. And to expand on that a bit more, would you mind sharing a little bit of info on the roles you're hiring for? Yeah, sure. I posted three different roles. So I'm taking my team now, which is myself. I have an assistant director, and he's ahead of my back-end data engineering pipeline, something like that. And I have an analyst on my staff right now as well. We're trying to double the size of our team. Hiring another quantitative analyst who, oddly enough, in this conversation is likely going to focus a lot of their time on the player tracking that I just mentioned that is similar to the big data bull stuff. And then also another back-end engineer developer to work closely with the one that I currently have because that is growing in scope, right? And then the third position will be TBD. It might be a software developer, something front-end. I haven't decided yet, but that's what I've posted as, and those are the applications I'm looking through currently.

Cool. Thank you. And when you just mentioned there's no excuse for someone to not have a portfolio, is that something that you sometimes come across and actively looking for people, or is that just on resumes that people send? I think it's both, you know what I mean? If you want to take that a step further, and like a lot of people have done, put some of that stuff out in the public sphere, then people will find it. If it's good enough, people will find it. We're always constantly looking, and people love to share good work, and it's amazing the network effects of Twitter and LinkedIn or whatever it might be. You know, I was actively reading a paper I saw on LinkedIn from a connection from years ago who's doing some work in the Bundesliga soccer, right? He did something on passing, modeling passing efficiency and stuff like that. So I'm like, I'm reading that paper right now. Like, that could be your paper. You know what I mean? I could be reading it right now. That's something that you've done and some work that you've done or something in the big data bowl. When I have time, I try to go through a lot of those submissions and see what people are doing, what ideas can you borrow and steal and things like that, and also just see the type of work people are doing. It's the easiest way to really find good work.

Definitely. Robert, I see you asked a question you said I can read out loud. So the question, Ian, was does knowing analytics at the level of depth that you do impact your ability to enjoy watching sports? Yeah, that's a good, that's a very good question. The only thing I'll say on that is this is your job, right? And I'm passionate about it. It's all consuming doing this type of work, obviously. There's no days off, right, kind of, because like sports don't stop, right? Like when you do your nine to five, you go Monday through Friday. But like we play games Saturday, Sundays, holidays, you name it. So it has to be something you're passionate about. That's another thing I tell people, like this has to be the thing you think about constantly. I think about in the shower,

I think about dinner, I think about it all the time. Not because I have to, it's just what I enjoy, right? So you should be doing what you're passionate about.

But in saying that, I haven't cheered at a sporting event, and I can't even remember. It's just like, it's not ingrained in me anymore to like cheer. Even when I go as a fan, it's hard to do.

Automation and data science teams

Thank you. I see there's quite a few other anonymous questions that are coming in as well on Slido. And so just a reminder, if anyone wants to ask questions there too, you can do so anonymously. But one was a repeated theme so far in any data science discussion has been automating stuff. Any thoughts on why data science teams do these, not the app development team?

That's a good question. So sorry, can you repeat it? Was there two questions in there? Like one automation and what was the like? Can you separate the questions? Yeah, definitely. And feel free to jump in if anyone wants to add more context if that was your question. But it says a repeated theme has been automating stuff. Do you have any thoughts on why data science teams do these and not the app development team?

That's a bad question for me. I guess working in the context of sports, like we're like a startup, right? Like I just got done saying, it's me and two other people. Like we're so small, we're all doing everything. So I don't even know if I can answer that question, to be honest. Like I don't have an app team. I wish I did, but I don't. We're all kind of helping each other out and doing things. So I'm going to pass on that one only because I'm not knowledgeable, not because I'm skirting it, right?

I don't mind jumping in, Rebecca or Ian, to throw you a bone. Like my opinion, I've seen it. It's a very fair question, like extremely fair. And like most things, it depends. And context is really, really important. From my experience in working in a couple of different sectors, I would group it in a few places. One, your typical developers, no offense to anyone on the board, please don't jump through my webcam. You don't know data. You don't know how to program. So those data automation pieces are a lot of times about automating data and processes rather than building front end widgets and building UI and build like that kind of context. They can, how we're positioning it and how I've seen in other places, those developers optimize, they don't create that automation process, right? So it gets into that refinement piece.

I see that as one area where it happens. And in other sectors, again, truth bomb, collective bargaining agreements and unions. So in a lot of organizations, IT shops or IT groups are unionized. So to have them jump outside of that role is a very difficult piece for upper management and executives to play in. So I'd see those two types of factors are really large aspects, which puts data science, data analytics teams in a good position because what we program, we collectively reprogram and we think about process and we think about the data in the automation rather than, well, I have to have this button. It has to like UI software development, conceptual ideas. I would leave that. And then as usual, I'll stop.

Thanks, Ryan. I see there's a lot of great comments in the chat around that too. And talking about how, if you want it to happen more quickly, a lot of times the data science teams will be the one to do that. And Alec, I love your comments there. My proof of concept, shiny apps have been alive for two plus years now.

Yeah. The one thing I will add on that though, like what I'm trying to get out, trying to build towards with my team now is a little bit closer to what Ryan is just got done describing, which is like right now I might be doing, you know, if I'm doing, building an app work or like just use that because everybody understands that. Right. I might be doing like the data aggregation, building the features, things like that. And then I'm doing the modeling, then I'm doing the validation. Then I'm, you know, deploying the model on something or designing the app itself. Then I'm doing UI, like I'm doing all of those components right now.

And so like, I'm not an expert necessarily. I might be an expert in one of those like five things, like a perfect world, a high functioning data science team, in my opinion. And it's something I would aim to aspire to be is that my engineers, the two guys that are two people I all have working in that role, like they're the experts at like feature engineering. And especially when you get big data scale, right? Like making that data, pulling that down and manipulating it as efficiently as possible, because once that loop comes back to production, it's going to have to come back to them anyways, usually, right? As new data is feeding in, they're going to be the ones that can optimize that portion of the process.

And when I do it, it's probably going to be inefficient compared to what they can do, right? So that would be like the first step is having them do it. Then the modeling is a second step. And then the front end UI, that's a whole different field, the people that have expertise in that. So I would say like what I'm trying to get to eventually is have all those things be experts, right?

What candidates and hiring managers should look for

Thank you, Ian. I see there are a lot of questions also going back to what we were talking about earlier on hiring. And there's a great question that came in anonymously that was, what should candidates look for, or evaluate when considering a team or organization to join? And if you're hiring, how do you stand out to candidates?

I was hoping for this question. I know hiring is on everybody's mind. And it's been, I will admit, it's been years that I've been on the other side of the table, but I've done three iterations of hiring myself over an eight-year period on my team now. So I've seen this evolve. And the things I'll say are, honestly, you'd be surprised how many people just don't take the time to like customize their resume. Like if there is some sort of like prompt, like I ask questions and they don't fill them out or they don't take time, they don't answer the questions that I'm asking. So I can't even evaluate you because you're not doing what I'm asking you to do off the hop. So don't underestimate that part of that first stage.

And then the next thing I would say is really lead with what work you have done that will be most relevant to the position you're applying to. So I just did an interview this morning, a second round technical interview with one of my candidates. And she has great experience. Obviously, I'm moving her along. I think she can do the work. And she dropped something on me. And I'm like, wait a second, this is nowhere in your resume. And we've had, this is our second conversation. This is super relevant to the job you're trying to get with me. And you left it completely off. Like why? And if I didn't ask the right question, I would have never learned that about her.

I'm like, okay, you need to put this somewhere very, very high if it's not with me, the next person, because this is so crucial. You almost missed it. And that piece of information you just gave me is like so relevant. So that would be like another piece of advice that I literally just learned this morning is like this person was withholding a very key piece of information that she didn't think to put the highlight. And I was just like, man, this is, that would have been unfortunate for us to miss that.

So that would be one. And then the last thing I want to tell people is that in sports, especially, I'll only speak to sports, because that's my expertise. So I apologize for people that want industry, but maybe it's true for industry, is like, in 2014, when I started, there was a couple of data science programs. And it was like, as I've done hiring, I was like, struggling to find people that had skills, right? And now the applicants I'm getting is everybody has skills, because there's so many great programs. Like I just stopped on describing, there's so many different avenues to showcase your work. So how do I differentiate those people now, right?

There used to be like, I would get 100 applicants and five had skills. So now I'm getting 100 applicants and 40 have skills. Okay, so now that gap is like, there's still how do we differentiate amongst the 40 now should be your question. And the question and the answers I have to that are like projects and things that showcase specific to the our industry, which is sports, right? Show me something that you could apply there. And then now we're going to get into like people that have domain knowledge. And that's probably true of industry too, is like, how can you develop your domain knowledge, you have all these great skills, you might work in industry, but I can't draw the parallel from your industry work to what you know about hockey now. And that might be as we get down, as you get down to the top three applicants, that might be the differentiator for you. So show me that you understand hockey, right?

There used to be like, I would get 100 applicants and five had skills. So now I'm getting 100 applicants and 40 have skills. Okay, so now that gap is like, there's still how do we differentiate amongst the 40 now should be your question.

Right. And maybe I see you just asked a follow up question there, too, which I'm curious as well. So feel free to jump into if you want to add anything. But maybe I think you were asking around what was that information that they left out? Yeah, yeah, like the candidate, your candidate story.

I don't want to, you know, blow that story up. But the point I'm trying to make is that it was, she was working for another, I'll just call it organization. And she was, there was a bottleneck in their process as a team, right? And she got tired of it. And so she said, I bought an iPad, I bought a book, I developed an app, a piece of software, and I run the software on a daily basis. And I run the software on a daily basis, and I'll keep it and do all this stuff. And I'm like, why are you telling me this now? Like, that should be the very first thing out of your mouth. Like, that's an amazing initiative to do.

Like, there's so many things I could go on and on a half an hour about just that. But like initiative, she taught herself to do something, she developed code and software, executed it, start to finish, maintaining it made the organization better. Like, how is that not the very first thing you're talking about? I don't know. I'm crazy. Maybe I think you're crazy.

Yeah. Thank you. I think part of that question, too, is also for data science leaders themselves to how can they also make the roles more appealing to a candidate? Like if somebody has maybe five roles or more that they're interviewing for?

Yeah, that's another interesting one for me, because I did have some people apply across apply across my position. Another thing I did want to get across is now that teams are, when I say teams, I mean, like, analytics team groups within the sporting setting. They're growing now, right? So now it's not when I started, it was like, okay, you're hiring one person, and they need to do everything. And now I'm going to have a team of six. So now I'm looking for people who has like expertise in a certain area, right? So showcase that you have not just a broad knowledge of all things, which is great, right? But I also want to know, if you are joining my team, which I call like my group, where are you an expert? Where do you make us better? Like, where are you better than me?

I want to hire people smarter than me. This is my philosophy is like hire somebody smarter than me in a different area so that they can be the expert and tell me what to do, or take ownership of that piece of our group, right? So showcase that you can become an expert in something. I don't know what that is. That needs to be something, you know, the person, the applicant decides. But if a lot of your work is just like general, that's going to be harder, I think, as we move along, and the industry grows and grows and grows to be a generalist. When we have a lot of people that can do general things. Now we need people that can specialize in something and be impactful in an area.

I want to hire people smarter than me. This is my philosophy is like hire somebody smarter than me in a different area so that they can be the expert and tell me what to do, or take ownership of that piece of our group, right?

Moving into leadership and managing a team

That's really helpful perspective there. I see there's quite a few other questions in the chat too. And Seth, you just asked one. So I'd love to pass the mic over to you, if that's okay. Yeah, sure. Hi, Ian. Happy New Year. I think at this point, it's like, right. But yeah, I was just curious about, I see that your progression within the organization is really to kind of, you know, now you're director of analytics over there. You know, what are some of the challenges that you find as you kind of, I guess, leveled up? And if you don't mind sharing some of that, that'd be great.

Yeah. Again, this is like my 2022 goal for my myself is like, and I'm being vulnerable and honest here when I say this is like, I started as a team of one. And my prior role, I kind of worked for myself a little bit independently as well. And then I've hired two people that I have a group of three. And to be honest, like I'm learning the management side of things. As I'm going, I'm doing it on the fly kind of thing. And I do believe I need to get better at managing a team, a data science team. So that's what I'm going to focus on this year is going, how do I go from three to six people that are a part of my group, including myself, right? How do we organize ourselves and optimize our group, right?

And that's something I think is what I need to progress at next is if anybody has good ideas, please drop me a note. I'm looking for high-performing data science teams and how to manage them and make sure everybody is pulling in the same direction and feels empowered to do their job. So does that kind of answer your question? I'm sorry. I might have misunderstood. No, that's great. Thanks.

Okay. And we can also make it more of a discussion too, if anyone else wants to jump in and have some feedback too. Yeah, I guess just to expand

on that stuff is just like my progression has been like going from an individual contributor to now going to like a manager, right? And what that entails. And then I'm also part of like a management team that is doing more strategic type work and like, you know, long, short, medium, long-term planning, like things like that. So it's like distinguishing those roles. That's where I get split in two is like, it's not just the contributing data science work. It's the managerial strategic work as well.

Thank you, Ian. I see there was a question earlier going back to the idea of the portfolio too, if someone had asked, what is more impressive to you for portfolios, a machine learning model or exploratory data analysis with tables, plots, different visuals, storytelling? For me, I think it's, it depends, you know, it really depends. I've seen people just do excellent work in a variety of different domains. So I don't really have a great answer for that. I think it's just really showcasing like what you are passionate about and making it clear what sort of impact your work is going to have. Like where do I fit you on the team, right? If it's modeling, like here's my modeling work. And if I join your team, I'm going to be the expert in these areas. And here's the domain knowledge I have about hockey to apply that problem to this, you know, these tools, use these tools to apply to that problem, right? So you got to make, show me that you can make that bridge, right? Is it like the technical stuff is all great, but I also need to know that you understand how to apply that to the right problem within the game of hockey, which, you know, not being said, the game of hockey is very complicated, more so than almost every sport. It's way more complicated. And I can go on about that, but it takes, it takes a lot of domain knowledge to understand our game. So I'm not saying everybody here needs to be a former player. I'm not a former player. I have no playing background at all. I'm just saying you need to understand all the components of the game itself to be able to do the job.

Moving into leadership and losing technical skills

For sure. And Trevor, I see you had a question you just put into the chat and I'd love to have you ask that live. And if it's okay to maybe also ask about some of the process that you're going through internally and trying to show showcase importance of data science too.

Oh, happy new year, everyone.

Happy new year.

Happy new year. So my question is a continuous question, right? So as you go up the ladder of moving from executing day-to-day projects to actually managing a team, it's just something that lingers in my mind because in your current situation, that's how it is, but next year it could be different. So I've always wondered if I moved so far away from the code, so far away from managing, actually executing projects, do I lose my coding skills? And when I'm in the next environment where I need that skillset, is that going to be a difficult process to refresh myself? You know, it's just something I always toy with mentally.

Now that is a great question. And I'm actually going to drop, start with everybody, I want to drop a post that I was reading last night into the comments right now. And again, this is on top of my head. So I was literally reading this. You guys can, anyone that wants to read, can you guys all see that, everybody? It's an excellent blog post. So you guys, it talks a lot about what your question was, so I'm not skirting it, but when somebody can say it better than you, I tend to just allow the better person to say it. But Trevor, I will say that's what 2022 is going to be for me is sorting out what you said. Am I moving further away from the code? Am I managing the team? Am I supporting people in the sense of creating tools for them to do their job better? Maybe that's some of it. Like whether that be for the R users out there, are we creating more packages that are more internal packages that help everybody do things? Is it just setting expectations and managing the group through like processes, right? Like get version control type stuff. Is it templating all of our workflows so that it's clear that we're all working collaboratively, but we have a standard structure, right? So those are the really relevant questions to building a high functioning team, I think. And we've made strides in that over the last couple of years. I give a lot of credit to the two people on my staff for helping to build that, because it's not just me, it's we're doing it collaboratively. I guess the piece of advice that I'm trying to follow is like, listen to the people that are working with me, right? Like, what do they want to see? What is relevant to them? Like ultimately I have to have some non-negotiables, but other than that, like they're doing the work, so I need to check in with them and see what makes sense too.

Yeah. So Ian, what I really want to find out, are you nervous about moving away from coding? Nervous? No, I mean, I think you make time for things that are important to you, right? We all do that on our daily basis. So if exercise and health is important to you, you make time for it. You make time for it. You just do, right? Like it's either important to you or not, right? So for me, if coding is something that I want to continue to maintain, I'm going to do it because I enjoy doing it. I'm going to find the time, whether that's at night, that's in the morning. I have two young kids. I mean, I'm juggling a lot of things, right? So it's just time management. And when you have a staff, it's, you know, I've talked to my, with my staff also about maybe having like, toying around with the idea of like office hours, like, okay, 9am to 12pm, I'm like blocked off. Like I'm, that's where I'm going to do my work, my individual work, right? And everybody's going to do their individual work. And then 12 to 1.30 office hours, come with me to any questions you might have, anything we need, we need a decision on or like there's, it's not a clear path, we need to, we need to choose the best way to go. So I've toyed with some of these ideas. Again, I'm open for anybody that has experienced this and has great things that they've seen in practice. Like I'm all ears. I don't, I'm not an expert. I'm, I'm just a guy trying to do my best here.

I think you make time for things that are important to you, right? We all do that on our daily basis. So if exercise and health is important to you, you make time for it. You make time for it. You just do, right? Like it's either important to you or not, right?

So thank you. When you have a lot of good ideas, you're doing a lot of great brainstorming. The next step for me, this is Valerie Bataille, I'm doing a lot of remediation and facilitation, I would suggest is having a strategic planning session for some of these. And it's great because you get buy-in, you get deliverables, you get timeframes, you get prioritized. But you're right at that point where that's kind of what I would recommend if it's more of a scientific study and you have a lot of different partners, then it's really getting everybody where you have a lot of different expertise. Everybody should learn what each person's role is and responsibility. And then from there you work on goals, deliverables. And it's really important that you're all answering the same question when you're doing data. Because if you get, that to me is the crux of quality with data, quality control. If you don't know what your question is, you're not going to get the answer you want. So for scientific studies, it's more let's get everybody on board. Does everybody agree this is our question? And how are we going to get there? And what's everybody's role going to be? When it's more management and the team and what does everybody think is the priority for the year? And what can we work on? That's more strategic than recession.

Yeah.

Thank you very much. I see, Ryan, you have your hand raised as well. If you wanted to jump in on this topic.

Yeah. If I could just give a little piece to Trevor's question and add on to a bit of what Ian has said in the comment from Frank in there. This was top of mind for me like 12 years ago. It's when I really start to transition from a heavy technical role into management. It was geospatial, but technical is technical. It kind of doesn't really matter if it's engineering, if it's coding, if it's whatever. And I struggled with it for about two years. And I think it was because I was comfortable at it. And when you're really comfortable and you're successful or anything like that, but when you're good at something, it's kind of hard to want to leave it behind. But as I matured into my role and I became comfortable of what does it mean to be a manager? What does it mean to lead? I have a bit of a philosophy that I like to carry around. It's the shield and spear. My role becomes less of tactfully putting out widgets. It's more enabling the people to be really successful. So if I try to measure my success by other people's success, and I know that might sound selfish, but it helps rationalize in my head. So if I can chop down and break down all the barriers for them, then they will be more successful. So that means my time, right? There's only so many hours in the day. So I sacrifice that technical to enable them to be further. However, I think it's still very relevant to keep your skills at some type of level just for street cred in your team, bringing new people on, having other people within there. If you can talk that language and they see you're like, oh, that's not just some dude in a suit. The barriers come down and you can not say that we wear suits anymore, but it just really helps your team and you can have those conversations and your communication goes up in a field that is technical. So I would say do it. And I fully agree. If you value it, you're going to make time. I totally agree with that.

Hiring people smarter than yourself

Yeah, I like to expand upon that too. I think my approach also is to use a little bit of humility with my team and just go to them and say like, hey, what are you doing here? I see you've done something here, teach me that. I want to learn that. And I tell people that actively that the guys on my team right now and people that I'm interviewing, I'm saying, I said this briefly earlier and I mean it, I want to hire people smarter than me because if I'm hiring people below my skill level all the time, then I'm constantly having to be the teacher where I'm trying to lift up six people myself. Whereas if I'm constantly hiring somebody that's an expert at something above me, then they're lifting me and my group up, the other people in that one domain. And then the next person's doing that in their domain. And so we're spreading the expertise across and we're teaching each other and working collaboratively as a group. The buck still stops with me and I'm still responsible for, like you said, being the shield, as you said, for management or moving obstacles out of their way. And sometimes the obstacles that are in people's way, I've realized is like me or processes. I'm putting things in their way in terms of tasks or making things inefficient for them to do their job better. So how do I learn to move those things out of the way? So those are things that I would like to add, I guess, on that.

I want to hire people smarter than me because if I'm hiring people below my skill level all the time, then I'm constantly having to be the teacher where I'm trying to lift up six people myself. Whereas if I'm constantly hiring somebody that's an expert at something above me, then they're lifting me and my group up, the other people in that one domain.

Awesome. I love this discussion that's happening now, so I don't want to jump away from it. Ian, I think this answers your question from earlier, Ian Hunter. There's a lot of Ian's on, I think. But around upskilling your team and are you going to require any continuous learning per week or per month or specific upskilling goals per year?

Yeah, that's a good question. I think each case is unique, but I don't know if I personally feel comfortable mandating things. I just don't think that ever goes well. Any aspect of life is like mandate things. You want people to want to do that. So again, to me, that would be like, what is that individual's goal? What are they trying to accomplish? What do they want to get better at? And then how can I direct them or put things in their path to say like, hey, do you think this would be helpful for you to achieve your goal? Are you aware that this exists? I'm a little bit of in a unique position because we work, our organization is owned by Comcast Corporation. So we've been integrated into that corporate environment. So we're still operating as a team, but we're also corporate now employees. And so there's going to be some opportunities now moving in 2022 and beyond for us to access some of the corporate environment where there is a lot of training and classes and certificates that you can get through various different things. So that's something new that we've never had. And now how can we leverage that? How can anybody that want, I just make, I need to make them aware that they can do that and then provide them the time and the space to learn, right?

For sure. And it looks like there's a lot of great suggestions too, in the chat around what other teams do too. And Evan, if you wouldn't mind jumping in and sharing that example, I think it'd be really helpful too.

Can you hear me?

Yep, I can.

Okay, good. I got to apologize to the team. I'm on my work computer and it's government, and so I don't have a web camera on it. So you don't get graced by the mess that's in my office right now. But obviously I'm in the government. And so we kind of have to kind of balance meetings because government, nothing gets done without meetings and red tape. But as I was kind of looking through the blog posts that you had in there and trying to find a balance between how

do you kind of inform the team, both on some professional development kind of things. So that's kind of, in my post in there, I kind of mentioned we do a brown bag, which is up to an hour. And that can be on, hey, I found a new cool way to do something too. Hey, here's this new

example from like three years ago. Hey, XT Boost is cool. And one of our guys that has spent a lot of time on there kind of went through and kind of did a quick discussion on that for us to some additional work that we've done with some customers actually was our topic yesterday.

And then in our case, we do what I call it a project update. And so that's like a quick 30 minutes. We've got two time spots a week on there. Sometimes we use them, sometimes we don't.

And that's kind of at all levels of a project. You could just be starting something, you're like, hey, team, this is what I think the task is. This is who I think the customer is. This is the data I think we need to be using all the way to, hey, this is a closeout. Like we officially handed this over. It is whatever the desired end result is. It could be an app. It could be a Flex dashboard. It could be a report. It could be, hey, the answer is 42. Hey, this is kind of what we've closed out here. This is what we learned. In my position, I'm our project management chief. And so my question then is like, okay, cool. So have you documented the code? Is it the final information is on the GitLab page? Have we closed the loop? Is there anything else that we need to kind of circle back on with the customer? Have we answered all of the questions? And more often than not, then it's like, oh, hey, can you, from the customer, can you also do this? Like, well, we probably can, but now we probably need a second project on here. Cause that's something called scope creep and sorry, this is a cool project for you, but maybe this team has to pivot to something else. So hopefully that was a little insight on kind of how we're doing things to some level. Definitely. Thanks Evan. Frank, I see you raised your hand too.

Yeah. So as Evan was speaking, he had mentioned meetings and got my mind thinking and something really effective that I've done with my team in the past six months that I think is probably relevant for everyone here. Every month we've done basically a progress update. So there's an expectation on my team that you have to be upscaling. You have to be learning. I don't care what it is, honestly, right? Do you want to learn JavaScript? You want Python R like whatever it is, you just got to be in that learning mindset. Cause I think that's when you uncover like kind of unexpected benefits of this stuff. So every, every month we have an hour long session. I have 12 people on my team now. And right. If you do that math, that's not a lot of time for each person to give an update. It's really quick. So each person gets like four minutes to update everybody else on what they're doing. So they have to have a pretty concise visual. Their story has to be pretty concise. And I found a lot of benefits coming out of this. First of all, everyone sees that everyone else is learning. So the laggards are like, Oh no, I really got to get my act together. The other thing is they have to practice being concise. Cause when you go to stakeholders to tell a story, you don't have a lot of time or attention, right? So they get to practice that. And then it's probably one of the only meetings on my calendar through the month where I see people getting really excited about it because we have that speed rather than being on one, another one of those calls where you're like silence, silence, everyone feels a little bit bored. It's like one of those high energy, like quick turnover meetings. And we're even like using that height, the quickness and trying, testing it out with some of our stakeholders and reverse showcase meetings. So that's been like in terms of upskilling one of the most successful routines I've put in place. Well, thank you, Frank.

an expectation on my team that you have to be upscaling. You have to be learning. I don't care what it is, honestly, right? Do you want to learn JavaScript? You want Python R like whatever it is, you just got to be in that learning mindset. Cause I think that's when you uncover like kind of unexpected benefits of this stuff.

I'm doing a bad job of managing the comments in the chat right now. Sorry.

But I just want to go over to a few other questions earlier from Slido too.

And one anonymous one Ian was around the kind of data that you work with. Can you elaborate on different data and work that you focus on in sports? Is it specific to ticket sales, player statistics, or free agency, money ball-esque?

Oh, I'm sorry. Can you repeat that one more time?

So it's just, the question was just elaborating more on the different data and work that you're focusing on right now. So is it more specific to ticket sales or player statistics or free agency?

Yeah, yeah. That's a good distinction. So a professional sports team is kind of split into two business units. You have the business side, which is ticketing, market segmentation, marketing, all that stuff you guys, you know, most companies have, right. And then we're the hockey, I work in the hockey side of the business, which is my boss is the general manager. I work with the coaching staff, scouts, you know, people on that side of things. Really the easiest way I tell people that ask me, you know, what exactly do you do? It's like, we look for any way to use data, you know, to make a better decision. So that could be, you know, something on the coaching staff. It could be something with our strength and conditioning group. It could be something with, you know, drafting players, it could be player acquisition, it could be the player on our team, how is he going to age? You know, what's this production going to be like moving forward? So it's kind of endless. It's just anything related to the team that could potentially make us better or help us make a more efficient decision, you know, add another piece into that decision making process. We are not the only piece that don't, I don't want to make that, I want to make that clear. We are just one piece of a cog, right, that there's scouts involved, there's other stakeholders that are all putting in information to one big piece that ultimately decision gets made from.

Collaborating with scouts

Can I ask to expand a bit on that? Because you just mentioned there's so many different pieces. How do you all work together? Like, how do you work directly with the scouts?

I guess it's just being involved in in the meetings, I think we have like designated meetings right throughout the year. Like next week, we have our midterm amateur scouting meeting, where the scouts will meet and they'll just say, Okay, we're halfway through the season, heading into this upcoming 2022 draft. What are the players of note? Who's trending up? Who's trending down? Which players are we concerned about? Which players need more coverage?

From a scouting standpoint, it's always like, it's an eyeballs business, right? And you have to see them make an evaluation. Sometimes a person gets hurt. Sometimes they don't play. Sometimes they're suspended, you know, so sometimes it's like, which guys do we need to go get more on? So those meetings being a part of the, it's helpful for me, I don't necessarily contribute much to them. I can at times, but what I what why it's important and relevant to my role is that I need to know what the scouts are saying, what they're interested in and what their problems are. You know, I'm really having a hard time on this guy. He got hurt last year, the 16 year old, so we don't have any book on him. He's playing as a 17 year old in this league with these players, and he's struggling with that decision as a scout. So what I can do as a data analyst is be, go to him after the fact and say, Hey, if I could provide some information on similar players, doing similar things at similar times in that league, would that be helpful? Yeah, man, that would be awesome. Can you get that to me? I can help them in that manner, right? So that's where the collaboration is. And a lot of times it's, I think the consulting aspects, like some consultants that are probably on this call is, is it's like listening, like what are people talking about? What are the issues, right? And then using my knowledge of the data we have within our possession is where can I go to that archive and pull that piece of information down, package it up and get it back to them so they can feel more comfortable in their evaluation. Is that helpful?

No, that is really helpful. When you say package that up to them, how do you share that with them? It can be a variety of ways. Sometimes it's a phone call, sometimes it's a text message. Sometimes it's a report. Sometimes it's, you know, a lot of it's, when I talk with my staff and I'm interviewing people, like a lot of times, a lot of our job is preparing for meetings, if that makes sense. So we get in these meetings, right? And all of a sudden, somebody's questioning, okay, should we acquire this guy? Or do we want to make this trade? Or what do we think about this player? We need to answer that question right then and there. What sort of tools can we build prior to, to enable us to answer those questions, at least as best we can in that moment. So a lot of the work is like preparing. So that might be an app, that might be a dashboard, that might be part of a web, our website that we have, it might be something like that. But you might be working three months, three months, three months, why am I doing this? Nobody's looking at it. And then we get into a meeting, and somebody's asking the question, you're like, oh, I've been building that, here it is, put it up, and everybody's looking at it, and we're interacting with it. So for us, it's, I had this mistake originally, thinking that like, I'd be in a position where I could just give this tool to people. And oftentimes, they don't want, they don't want to do that, right? They just want to see it, they want to see you, the expert, kind of work through it and showcase all the different aspects of the information to them, and sometimes explain it to them, what they're looking at. And then they say, oh, that makes a ton of sense. And then they have questions on that. And you can work through those. So it's just opening up that box and having more productive conversations. Because if everybody remembers, like back in the old, the old days, you know, somebody would ask a question, right? And then you'd say, oh, I could answer that. Let me get back to you in a week. Right? Like, so how valuable is that to that person? They've forgotten the question a week later. They have or they've moved on or whatever, right? We want to answer that in the moment is the best way to describe a lot of what we do also.

that might be an app, that might be a dashboard, that might be part of a web, our website that we have, it might be something like that. But you might be working three months, three months, three months, why am I doing this? Nobody's looking at it. And then we get into a meeting, and somebody's asking the question, you're like, oh, I've been building that, here it is, put it up, and everybody's looking at it, and we're interacting with it.

Thank you. I appreciate that perspective, too. And it's important to think like, I guess in my head, I'm like, oh, people just want to create this dashboard or this like application and just give them that. And that's the answer. But that's not really how it always works.

I'm going through just to see which questions I had missed from earlier. But Bruno, I know you had a question towards the beginning around, like data and marketing as well. And asking if Gritty was a result of this. Do you want to jump in, Bruno? Yeah, sure. So Gritty jumped on the hockey mascot theme and basically destroyed every other mascot. So did you have any hand into like crafting this piece of marketing flyers, propaganda, using data or social media analysis or?

No, that is the brainchild of someone else, unfortunately. But it's amazing how he's taken on this whole life. At some point, someone will write a book about it. I'm sure. Let's just I'll leave it at that. And it'll be I'm sure people will be reading about it in case studies and Harvest Business Review or something at some point. And I'll be reading it, too, because I have no idea what went into that. And I'd be interested to hear that as well. So I have the same interest as you. Thanks, Bruno.

Mike Smith, I see you had asked the question earlier as well around if it's any easier to influence decisions in sports rather than or harder compared to other organizations. Would you want to jump in, Mike, with your perspective there? No, go ahead. That's great. I'm just interested to know if you're I guess all organizations you're trying to influence up to managers and managers are either going to accept the data driven analytics or go with gut feel. But just if you have any experience of differences between doing that in a sports context versus any other.

Yeah, I can't really speak to like outside of the sporting world, because I admittedly don't have a lot of industry experience outside of this world that I'm in. But what I've arrived at as my own mental model is like people are people. This is like still a people's business. You're still interacting with team members. So whether you're in the sports or you work at Target or Walmart or whatever Fortune 500 company, the people are still have the same motivations. They still have the same personality traits. So I don't know if it's easier or harder. I think it's just people. It's teamwork. It's getting people, making people in the group feel like they're heard and that their input is is valued, even when that decision might go against what you're recommending. It doesn't mean it's not valued. Sometimes it's on the contrary, right. But in sports, there's also an element of like, there are things and I would put this out to all the modelers, right? Let's make this about data science, okay? When we build a model, what does everybody know is that we have things that are in the model. So there's like the known knowns and there's the known unknowns. So there's things that we know exist, but they're not in the model, right? And no model accounts for everything. So when somebody is making a decision about how I say this, they make a decision that is influenced by something else over here. Sometimes it's not in the model and I know it's not in the model. So like, what would that be? Like somebody might be saying, what is he even talking about, right? So something like that might be saying, hey, here's this player. We want to acquire him. He has all these great results. He's going to project out nicely. And then my GM goes down and gets the guy's medical file from the team that he's currently playing on. And he has some sort of degenerative issue and that there's a high likelihood that this

person's going to have health problems, you know, and that might affect his performance. So like, that's not in my model. That's not in my report. That's not in my evaluation. That's not in anything I'm saying. It's a known unknown. And when you come to me and say, hey, we're worried about this, I'm not upset about that. I'm like, oh, I didn't know. I don't have that information. Otherwise I can model it, right? So there's an element of that. Like, what are the known unknowns and being okay with that, right?

I realize an hour just went by very quickly. So I just want to be conscious of your time if you had to jump to right to another meeting. I'm okay. If you guys want to, if everybody wants to keep going, it's fine. I'm probably tired of hearing me talk. No, no. There's a lot of other questions. Thank you.

Technical interviews and hiring

One question that came up earlier, I just want to make sure to ask was around like, what does a technical interview look like for your team or how do you approach interviews?

Yeah. So we've, I've tried to do different iterations on this, but the one I'm currently using right now, and again, this is evolving over time and I'm sure there's better ways to do it. Like my personal opinion is like hiring is so difficult that I don't think any of us are trained in that. I don't know if there is training. So we're all doing our best in a complicated process. But for me, what I, what I boil it down to is like, can you do what you say you can do, right? I need to have some sort of test that shows me that you can do what you say you can do. So it doesn't need to be any more complicated or intensive than that.

And I'll give everybody an example. So not this time, but last time I hired, I got to a second interview with a group of people and I had a coding challenge that I gave people, I don't know, 24 or 36 hours or something to do, return it back to me. And there's a variety of reasons why I do that. I got down to the time I was supposed to speak with this individual and he, the person sent me an email one hour in advance and just said, I have nothing. I could not do the exercise. I told you I could do X, Y, Z, and either I overestimated myself or I, you know, he didn't say he was lying, but he basically couldn't do what he said he could do. So I avoided a very large and costly mistake. Had I gone on to just trust that person and I'm not saying it was done intentionally or that this person was trying to fool me, but I'm just saying we have to have some sort of checks and balances that people could do what they say they can do.

Right. And then I just want to see the depth of your work. Like how far do you take things down? Are you just a simple, add two numbers together? No, that's fine. We'll get to that in a second. But do, can you show me some techniques that I think that you could bring to the table? And then the third component of that is there's no perfect candidate and everybody's going to come in with their different strengths and weaknesses. And by doing this coding challenge, I'm learning about you as a candidate to say, where do I need to upscale this person? Where do I need to spend time with this person on day one, prior to this person starting, I'm going to have a plan. I'm going to say this person needs to get better in XYZ or after a certain number of months, I want to circle back with them and say, hey, I think you can utilize these skills a little bit better and make you more efficient or can excel your code writing ability or whatever it might be. Or maybe it's somebody on my team, like, hey, this person on my team, I think would really help you in this area. So by showing me a little bit about what you can do, that's not to eliminate you as a candidate. It's more or less to see what are your strengths? What are your weaknesses? How can I help you? How can I upscale to you eventually?

Thank you, Ian. That's really helpful. I know that comes up quite a bit in terms of like technical interviews. And sometimes when people are asking maybe far too much of someone, we need to make sure that they have that component.

Yeah, I've read the, you know, the tech world is in the whiteboarding and all that stuff and putting people. I don't think that's for me, coming from a sport background, the whole idea of practice in sport is to simulate things that you will see in a game, because that's your job, to play games. So if we expect you to perform at a high level in a game, we need to simulate that situation, slow it down, help you practice and get better and get the synapses moving quickly and firing quickly, right? That's the whole idea of practice is to slow things down and then gradually build up and up and up and translate exactly what you did in practice into a game. If I'm having you do something in practice that you're not doing in a game, why is that relevant? Why should we be spending our time doing that? So I kind of see the idea of making somebody code something on a whiteboard while I see some benefits to it. At what point in the game, in this game as you working on my team, are you ever going to stand in front of a whiteboard and do something like that? That's just not how we work. So why would I make somebody do that? Just one man's opinion. I know there's going to be people that are going to argue this to the death, but just one person. No, I like that. Thank you.

So if we expect you to perform at a high level in a game, we need to simulate that situation, slow it down, help you practice and get better and get the synapses moving quickly and firing quickly, right? That's the whole idea of practice is to slow things down and then gradually build up and up and up and translate exactly what you did in practice into a game.

But in saying that... They might get agreeing in the chat too. Yeah, but in saying that, when I do give them a coding challenge, I do put them under a time limit and I do tell them it's coming ahead of time, because that is the job. It's not me not being fair to them or putting them under pressure just for my own fun. That's not it. The job is, it's Friday night and we're playing Saturday and our coach says, hey, I'm struggling with lineup combination or some problem he might have. I'm wondering who should be going. This person's hurt. I'm struggling on who I should put in or lineup decision. What can you give me on that? I have 12 hours to get him that information, the best of my ability. So that is the job. So I have to put people also under that pressure of saying, okay, I need to work through this problem. I need to come to a minimal viable product to give this person that can give them information quickly. Because sometimes we are put under the gun in terms of getting somebody something before a game. There is a deadline. That might be different than industry where we're working on quarters or long projects and things like that. No, we play tomorrow. So I need this before the game.

Democratizing data across an organization

Definitely. I appreciate how over the past few months, this has really started to feel like a community of people and seeing people come every week is awesome too. Trevor, I know you and I were just chatting earlier today. I think you said you have to leave, but Trevor, you were talking a little bit about how to present to your team on the benefits of data science and maybe learning from someone else in the community who has shown this to their own executive leadership. And I just want to see if you wanted to pose that question to the group before you run.

Yeah. So just to let you guys know what is happening on my side. So we are in the process of building out a company-wide dashboard. So I'm in Jamaica, so we call the dashboard Jamaica 360. I think it's 360 degrees because the idea is to give users the ability to see the entire business instead of looking at just their side of the business, because that leads to silos. And we actually see that impacting the results that we get in a negative way. So that's one. Two, what we're trying to do is democratize information or data, because what we want to do right now is to ensure that everybody is data literate and that everyone can analyze their own information to an extent. Because what we have now is a bottleneck where when people need data, they have to come to my units. And of course, that is inefficient. It slows down decision making process. So what we're trying to do is upskilling the entire organization and also help people in terms of consuming their own data. The fortunate thing for me is that, which is not typical with data persons necessarily, is that I sit on the senior management team. So all I need to do is to convince the general manager and she's definitely on board in terms of where we need to go. But what I would really welcome is your thoughts in terms of how best to do it. I've been reading a lot in terms of what we need to do and some ways of democratizing information, some ways of sharing information. So any thoughts on that would be welcomed.

Is that a question for me? I'm sorry. Or is that to the group? It's in the entire group, actually. That would be great.

I'll jump in a little bit. I had that problem at a previous role because our audience was the city of Toronto, so three million people plus around the world. Now I'm leading the open data team. So how do you upscale citizens to use data to be able to do transparency and things like that? And I talked a little bit about before to other people. So fast forward so you don't have to go through that painful conversation, is we decided to make it easy for people based on where they work. So rather than trying to force people to speak our language, the data science language, like machine learning engineers, all that kind of stuff, let's not be elitist and say, well, you can't work in Excel. You can't work there. You have to come to us. No, no, no. If it's all about people making decisions and not discrediting their domain knowledge that they bring to the table, we'll let them work wherever they want. Don't tell me how to work. I won't tell you how to work.

So what we did is we started to make data access, that being the raw material, that being the gold, so to speak, easy for anyone no matter where they were. So we made tool sets, plugins for really simple things that most people used in our, which would be our customers, education people, and Google Sheets so they could access all the data that we had by a Google Sheet plugin. We made an R package for our R package people. We made Python libraries for our Python users. And we made a plugin for geospatial users in open source. So one thing that we did about that was making it easy for people to get it, because what we heard from hundreds of people is, I'm not a data scientist. I don't have a stats degree. I don't even know what a CSV is. So if I don't know what that is, the barrier of entry is might as well be a billion dollars. They just can't do it. It's too scary. So if we make it easier for them, that was one way to get them up the ladder. And then we heavily invested in writing and writing blog stories that were fun. Now, I air quote fun. It's not Disney and Pixar fun, but not being like a PhD thesis. It's not being jargon. And taking that away, taking that burden and cost away, making it so that the entry level for them was easier, and using fun words, analogies, connecting it to stuff, swearing with symbols instead of letters. So just stuff like that. So I think it really boils down to knowing your audience and having empathy for them.

Thank you.

Yeah, I'll jump on Ryan's end there. One thing that translates to what he said, he was talking about data science to citizens, I guess, or less data, sophisticated people, right? It's like using their terminologies, right? And so like in sports, the one thing that we try to do also is by being a part of those meetings with the scouts and with the coaching staff at times is like, what are the terms that they are using for certain things, right? Like I might have a metric or something that we've developed over here. But if I can't, if it's as simple as just putting it in a language that they're already speaking, instead of me having to teach them a new language, isn't that like an easy solution to, we all talk about like buy-in and getting people. Sometimes it's just we create an art, we meaning like the data professionals, we create an artificial barrier, an unnecessary artificial barrier between us and the end user. And then we blame them sometimes for not meeting us, right? Where I think the easy solution is like understanding and being aware that that could happen and try to head that off ahead of time and say like, can we put this in their words so that they'll instantly get it, they'll pick it up a lot quicker, right? And sometimes just changing that terminology is really the key to getting quote buy-in or people to understand it, right?

Sometimes it's just we create an art, we meaning like the data professionals, we create an artificial barrier, an unnecessary artificial barrier between us and the end user. And then we blame them sometimes for not meeting us, right?

Thank you. I love that. I know we're quite a bit over on time now. So I just want to ask one last question, if that's okay. So you shared that awesome blog post, which I realized is from Alex Gold from the RStudio team as well. I forgot it was too. I was not plugging RStudio. No, I appreciate it because I didn't see it. Now I have to reach out to Alex about that. But I was just curious, are there other resources or if it's books or podcasts or things that you'd recommend to us all? There's a few.

You can send them to me later too and I can include them in the recording as well. Yeah, there's a couple, but I'm kind of blanking on the whole list right now. I'm like, yeah, no worries. But yeah, I'll put a couple in there if people want to get more involved. There's obviously a lot of great work in the public domain too. It's amazing the work people do, again, showcasing their skills or doing blog posts kind of things like Ryan was talking about is just show how data can be fun and how can it be useful and things like that. Showcase your thinking too. That stuff is interesting for people to see. And those are obviously, like I said, the missing components. When you get down to everybody has the skills, what's going to differentiate people?

Definitely. Well, thank you so much, Ian, for chatting with us. And if people ever have follow-up questions or just want to connect with you, is LinkedIn the best place?

Sure. Yeah, you could do that. I think I have all my stuff on there. Another piece of advice too, for maybe the younger people on this call is if you're going to reach out to somebody, just be direct about what you're asking. Sometimes I can't figure out even what people contact me and I just move past it because I don't know what they're asking. Instead of going back and forth six times about figuring out what they want or what their question is, I'm like, okay, or what their question is. Take some time to think about what you're asking. Be direct, be clear, and then I can answer you quickly. You might think you're one person emailing, why isn't this person responding? But if I get a hundred of them, it's hard to get back to a hundred people. The more direct you could be, the quicker I can get back to you and you could be one of the ones out of those hundred that I actually can try to help. I do want to help, trust me, but I need to know how and without having to weed through innuendo, I guess.

For sure. That's really important too, especially as you get, I'm sure you get a lot of messages on LinkedIn too. Yeah. Please know stop and chat. Please know, do you have 10 minutes? Because again, it's 10 minutes and that's not a lot of my time. It's not, but if I get a hundred, 10 minutes, that adds up. It might be better to just ask one simple question that you want to know. If that's something that you prefer to do, I can help that in that way. But outside of that, it's just a time situation. Sorry. One other thing is I was actually going to repost some of the job postings and I think I maybe have old links or it says they're no longer up.

It might be better to just ask one simple question that you want to know.

Yeah. I'm down the rabbit hole on that. I've turned them off just because it gets overwhelming. I haven't hired anybody. So if you are really adamant and you want to insert yourself, I'll make a pop culture reference here. My wife sometimes watches The Bachelor, right? And you know, like they're 10 shows in and they introduce somebody. I think I always think that's a terrible idea for that show. But if you want to be that person that comes on mid-show, send me an email. Or LinkedIn. All right. Well, thank you so much, Ian, for chatting with us. We really appreciate it and good luck today. I know, or I think you have a game. Yeah, we do. Yeah. Thank you very much. I appreciate it.