Even Undead Lords Love Data w/Scree

1:14:05
 
Поделиться
 

Manage episode 318512631 series 2798195
Сделано Rob Collie and P3 Adaptive и найдено благодаря Player FM и нашему сообществу. Авторские права принадлежат издателю, а не Player FM, и аудиоматериалы транслируются прямо с его сервера. Нажмите на кнопку Подписаться, чтобы следить за обновлениями через Player FM или скопируйте и вставьте ссылку на канал в другое приложение для подкастов.

Today on Raw Data by P3 Adaptive, our guest is the mysterious Scree from the Undead Lords Guild. Choosing to remain anonymous, Scree brings his data and Power BI experience from his ‘day job’ in an informationally protected industry to his hobby and passion: gaming. Being part of the admin team gives Scree the opportunity to showcase his mad data skills and proves that you never know where a data person is lurking.

Also, in this episode, we learn how a gaming guild uses data to track membership to determine engagement. The data paints a picture of what games are being played and for how long giving the admin team an idea of which games to target for participation. Yes, even gaming guilds work towards goals and use data. After all, guilds are also content providers.

All of this yet they still fit in Power BI and more. You won’t want to miss this episode!

References in this Episode:

Neverwinter Nights

World of Warcraft

Mystery Shoppers

King of Kong Documentary

Office Space Flair Scene

Raw Data with Coach Chase Hargis

CoverHawk

NoCheckDowns

Episode Transcript:

Rob Collie (00:00:00): Hello, friends. Today we welcome as our guest, the mysterious Scree. By day, he is a Power BI consultant, working for a consulting firm. But by night, he's one of the Undead Lords. No, he's not into black magic. It's a gaming guild. But as part of that gaming guild, he does use Power BI to measure member engagement. And folks, the stakes are high because the Undead Lords are gunning for the Guinness Book of World Records for the longest running gaming guild. It's not just fun and games people. And that application of Power BI, or really any technology to one's personal pursuits became kind of a theme of our conversation. The question we had was basically, if you use the technology at work, but you don't use it in your personal life for anything, are you really that into it? And if your answer is yes, that you do use it as a hobby, then he and I share the opinion that that is far more important than any sort of formal training. Passion beats degrees.

Rob Collie (00:00:56): Like the majority of people who've been on our show, Scree took a very, very random path through the world of data to today, starting in an unlikely place. The sales floor of a Best Buy in 2009, where he was on the receiving end of analytics. With no power at all to impact the creation. And that's a little unusual. I think most people in our space, their first interaction with it is almost like organically. They're involved in the creation from the get go. I mean, that's what the VLOOKUP in Pivot Crowd is about, right? But even as a pure consumer of reporting and analytics, Scree recognized the value of it immediately. And in his next job, working in sales at a bank, they gave him access to Salesforce, and he never looked back.

Rob Collie (00:01:40): We also talk a little bit about the challenges of hiring these days. And we absolutely would've talked about Microsoft's acquisition of Activision Blizzard. But sadly, we recorded this episode before that news broke. Might have to have him back on just to talk about that. Anyway, it's not often that you have an Undead Lord on your podcast. So, let's get into it.

Announcer (00:01:59): Ladies and gentlemen, may I have your attention please?

Announcer (00:02:06): This is the Raw Data by P3 Adaptive Podcast with your host, Rob Collie. Find out what the experts at P3 Adaptive can do for your business. Just go to p3adaptive.com. Raw Data by P3 Adaptive is data with the human element.

Rob Collie (00:02:27): Welcome to the show the mysterious Scree. How are you today?

Scree (00:02:32): I'm good.

Rob Collie (00:02:33): That's how I know you from Twitter is Scree. And of course, then I follow the link on the Scree profile and what do I find? I find a gaming guild, right? I say that as if I don't know what that is. I do definitely know what that is.

Scree (00:02:50): Yeah. The Undead Lords. One of the oldest running guilds ever. We're actually going after the Guinness Book of World Records for that.

Rob Collie (00:02:57): No way.

Scree (00:02:58): Yeah. They were formed in 1994.

Rob Collie (00:03:01): What?

Scree (00:03:01): And the current record holder was formed in '96, so we're coming for them.

Rob Collie (00:03:07): Okay. So, '94. What game would've necessitated a guild in 1994?

Scree (00:03:13): Yeah, it was an AOL game called Neverwinter Nights.

Rob Collie (00:03:16): Okay.

Scree (00:03:17): I don't know if you guys remember the old Gold Box Dungeons & Dragons PC games?

Rob Collie (00:03:21): Maybe. You're talking Luke and I are very much old school. An episode that went live at the end of the year, includes a vignette or two about our Dungeons & Dragons, actual role playing tabletop days.

Scree (00:03:36): Yeah.

Rob Collie (00:03:37): But, yeah. I'm aware of the existence of D&D theme computer games.

Scree (00:03:42): They were really big back in the '90s. A studio converted one of them.

Rob Collie (00:03:47): Oh, look at that. Do you see what Luke's showing?

Scree (00:03:49): Yeah.

Rob Collie (00:03:50): The OG monster manual.

Scree (00:03:52): Wow. So, yeah. This guild basically started playing one of the first online graphical MMOs. And back then, guilds were just essential to being able to do anything in any game.

Rob Collie (00:04:04): I didn't even know that there were MMOs until, Ultimate Online is the first one that I was aware of. And I didn't play it. I just knew of it.

Scree (00:04:12): Yeah, they went there after NWN.

Rob Collie (00:04:15): I guess, the idea of a gaming guild, I wasn't really aware of it until one of my friends was playing the Tribes, first person shooter.

Scree (00:04:24): Yep.

Rob Collie (00:04:25): He's actually been on the show, Connor Cunningham. I remember them taking a group picture in the game. I was like, "Oh, that's kind of cool."

Scree (00:04:32): I joined them more recently, earlier this year. So, I haven't played with them since '94. They've just been around since then.

Rob Collie (00:04:40): But the reason we've crossed paths on Twitter is nothing to do with gaming, because even when I do play computer games, I am very much one of the casuals. I'm one of the ones that just fills out the ranks of a guild as ballast.

Scree (00:04:54): Sure. I know the type.

Rob Collie (00:04:56): But we crossed paths on Twitter because of Power BI. Behind the scenes, does Undead Lords use Power BI to run the guild?

Scree (00:05:02): I wouldn't say to run it. It's really kind of in its rudimentary state right now. But I've worked on it as part of a side hobby. So, some of the reporting that I do for my work, we do a lot of healthcare clients lately. I work for a company that was just acquired and they do client contracting for any number of services, mostly tech related services. Engineering, the hard science type stuff. But they've expanded into their data exploration toolset, so to speak. So, we've got people who know data engineering, data science, AIML. And then people like me who do more of the front end reporting work. And a lot of the stuff that I work on, I just can't show. So, if we want to show anything, I have to work on things on the side to be able to showcase my talent.

Rob Collie (00:05:47): Yeah.

Scree (00:05:48): So, for a long time, I was working on stuff for UDL. That's the shorthand we use for the guild. And UDL, we also use a chat application called Discord. I don't know if you're familiar with it.

Rob Collie (00:05:58): Oh, yeah. Very much.

Scree (00:05:59): Just like Slack or any of the chat tools that are out there. But Discord has a couple unique features that make it super nice for a gaming guild. So, one of the things that we've been able to track is when you play a game, Discord shows your status as now playing X game.

Rob Collie (00:06:13): Right.

Scree (00:06:14): So, for probably the last 18 months, I've been scraping that data as it comes across so I can track, one, what games we're playing. How long we're playing them. Who's playing them. And basically be able to paint a picture of activity and organization. Just what games are out there that people are actively engaged with. So, yeah. We have been using it just to kind of get an idea if we're not sure if someone's actively participating or contributing, we can go hop in and go look at their status and take a look at what they've been doing. So, it's kind of a little big brother-y.

Rob Collie (00:06:45): And who doesn't need some big brother in their gaming guild?

Scree (00:06:48): Yeah.

Rob Collie (00:06:49): You have to, right? It's an organization working towards a goal. That's always been one of my tensions with it is, see I'm playing a game to relax.

Scree (00:06:57): Yeah.

Rob Collie (00:06:58): And a lot of people play games. I'm not critical of this. It just isn't me. There's a certain excellence that's trying to be brought to the picture, right? It's very serious.

Scree (00:07:07): And that's the thing. We're a hardcore competitive PVP guild. And maybe the hardcore piece is in question right now just because of how long the guild's been around. And a lot of the original founding members are now 40, 50, 60 with kids and commitments. So, it's kind of hard to say we're going to log in like teenagers and no life it for the weekend. But there's still a level of information that I've been working to bring towards the guild, just so they have the insight. Whether they use it or not is irrelevant. And it's really easy to hide. If you don't want someone to know what you're playing, you just turn off the statuses and my bot can't track what you're doing. By and large, most of our members understand that if they are going to go play some ridiculously embarrassing game, they can turn it off and I can't watch it.

Rob Collie (00:07:54): Discord, by the way, is the best way for me to get ahold of my college aged son.

Scree (00:07:58): Yeah?

Rob Collie (00:07:59): It trumps cell phone by a mile. He's not going to answer a phone call. Maybe he'll answer a text. But hit him up in Discord and you've got a 50/50 chance. You know?

Scree (00:08:10): Yeah.

Rob Collie (00:08:11): I get to know what he's doing. Oh, look, he's playing Halo again. It's like checking in.

Scree (00:08:15): Yeah, in 2022, we're looking to go even further than what we've been doing. And I want to keep track of how many messages people are typing. How long people stay in voice chat so we can see, are they talking while they're playing, or listening at least? And just to get a better sense of the engagement with the guild. And we combine all three of those metrics together is the plan to get a total guild engagement score that we can measure for each member. And if members fall off and aren't as enthusiastic about playing with us, we can reach out and find out why. So, guilds these days are really a content provider for most games. The games themselves have some content. But guilds can provide more to it. And so, just by engaging those members, I think we can keep the reign of 30 plus years going for this organization.

Rob Collie (00:09:01): So, the 1996 record holder, have they dissolved?

Scree (00:09:04): No, they're still active.

Rob Collie (00:09:05): Okay, all right. So, you're on their heels, two years behind.

Scree (00:09:09): Two years ahead. However you want to frame it. Yeah.

Rob Collie (00:09:12): It's like this really long running game of chicken. No, no, no. You give up.

Scree (00:09:17): Right. Yeah, and the record is kind of suspect because it's consecutive running guild. You had to have played consecutively for that entire time. But the actual Guinness record says you can have a six month gap in between two games. No more than six months.

Rob Collie (00:09:34): Okay.

Scree (00:09:35): So, we don't get to see what their record actually looks like.

Rob Collie (00:09:39): I see.

Scree (00:09:39): So, we're not sure exactly if they were active the whole time.

Rob Collie (00:09:43): This is the seedy back alley truth of the Guinness record for guilds. Now we're really getting into the nitty gritty here. You know?

Scree (00:09:50): Yeah. But we want it. It's just hard. A lot of the records from that time are not great. You need screenshots and dates, and back then, screenshots didn't have dates.

Rob Collie (00:10:00): Yeah.

Scree (00:10:01): It's an uphill battle.

Rob Collie (00:10:02): Have you seen King of Kong?

Scree (00:10:04): No.

Rob Collie (00:10:04): You got to watch the documentary, King of Kong. Oh my God. You want to talk about dirty, dirty, dirty shenanigans with regard to world records. And again, in a space that you just wouldn't expect there to be corporate espionage. You know? That is a hell of a movie.

Scree (00:10:23): I'll add it to the list.

Rob Collie (00:10:24): Yeah, oh my God. Yeah, King of Kong. When I asked if you used Power BI to run the guild, that was just a joke.

Scree (00:10:33): Yeah.

Rob Collie (00:10:33): I wasn't expecting there to be anything there. Sounds like there is. You actually are. And I completely identify with this. There's multiple facets of it. First of all, you can't share anything publicly that's actually production data. Like the stuff that you do in your day job is almost by definition, super, super, super sensitive.

Scree (00:10:53): Yeah.

Rob Collie (00:10:54): One of the jokes I make all the time is, "We do two things here. We work with data and we sign NDAs." Those are our two core competencies at P3. And if it's not valuable, if it's not sensitive, we probably aren't working with it.

Scree (00:11:07): Yeah.

Rob Collie (00:11:08): Also, you can't always bring that personal enthusiasm and energy to every last project that you're working on. Most business projects with Power BI actually are pretty stimulating and engaging and fun. That's been my experience.

Scree (00:11:22): Sure.

Rob Collie (00:11:22): But you can't bring the creative passionate flare to its full strength with it until you take it into something that's in your personal life, or your hobby. And this is why, for example, we ended up sinking so much time and effort and money into things like our no check downs site, which is NFL football visualized. And that, in turn, spawned the whole CoverHawk thing with Coach Chase Hargis. It's cool to have public facing examples, in particular, where you can dial up that consumer angle to it. You can make it really, really, really consumer friendly. Does Discord, when you want to go and share one of these reports back to the guild, how do you do that? Are you just taking screenshots?

Scree (00:12:09): I can share the report with the public. It's an early work for me, so I haven't touched it in probably two years now. So, it's not really a great example of my current work.

Rob Collie (00:12:19): Sure.

Scree (00:12:20): So, when I originally got started, I was making a website for this game that we were going to go in and play, and I learned how to do backend dev work with JavaScript. And there's a JavaScript library for Discord. And that kind of facilitated learning how to scrape Discord. Make a bot that works in Discord so you could run commands and do things against that bot, the guild could. So, I don't do programming for my job. I'm not professionally trained. I have a high school comp sci AP test, and then maybe a couple courses in my undergrad. But I never professionally trained on any of it. I do programming on the side, again, for my hobby. If I'm going to work on something on the side, I'm going to challenge myself to learn something new. And so, this was kind of that angle I started on a long time ago, was learning how to program. And then, how do I visualize this?

Scree (00:13:11): And at the time, I was learning Power BI for my job, and decided, why not use this? So, I made a database that all this data's being trapped and captured into. And then, a front end API that connects with that data and then Power BI scrapes from that. I've written out this whole infrastructure basically for this report that maybe three or four of us use right now, at most. Again, it's just one of those learning things where I'm not just the guy who takes the data and does things with it. I was involved deeply all along the way of gathering the data and storing it. And how do I share that? And how do I make that into a visual front end using Power BI. And I've learned a lot of valuable lessons there that I use in my work because I've done all those things. So, I know where we can do certain things to make the data cleaner or where you should do them. And where you absolutely shouldn't.

Scree (00:14:03): So, 400 million entries of people logging in and logging out of Discord's not the one you want to do transformations on in Power BI. So, just small things along the line that lessons learned and always index your databases. That was a fun one.

Rob Collie (00:14:18): Yeah. I want to pump something up here, which is that these sorts of side projects, like what you just described, those are the real thing. This is where real programmers come from. For example, you had to opt in to take an AP computer science course in high school. What I've come to realize is that, that sort of volunteerism is much more important than any sort of formal training when it comes to programming. I've had the formal training. I'm a card carrying comp sci major. I've got a bachelor's degree in computer science. Look at that. Mm-hmm (affirmative). Yeah. And I'm just not into it. I'm just not into programming at all. It doesn't speak to me. VBA, okay, fine. Give me some VBA.

Rob Collie (00:15:00): But all of the good stuff that I've ever done, all of my shining programming projects, all of them have been personal projects. And right now, if we were running a software org where we were primarily a bunch of programmers, if that's what P3 was. Now, we do programming for sure. But we don't put ourselves out there as a software development shop. That's not our core competency. But if we were doing that and I was involved in the screening process for potential developers that we were going to hire, I would be looking primarily for what they do with their spare time. If they go to work and program and then when they clock out, they don't do any more programming, I don't know. I'm going to decrement their value. I'm going to say, "Nah, that's a check minus."

Scree (00:15:40): Yeah, but at the same time, if someone comes to you with a COVID Power BI dashboard, are you doing a jig around the table? Because I've seen enough of those to vomit.

Rob Collie (00:15:49): Oh, yeah. There's plenty of those. In fact, we're in the process right now of retooling our interview. And it definitely does not involve a show us your best COVID dashboard.

Scree (00:16:01): And I get it. What dataset are you going to work on that's not proprietary?

Rob Collie (00:16:05): Yeah.

Scree (00:16:05): You're going to use the most prevalent one that's out there. But, oh man.

Rob Collie (00:16:08): Yeah. And talk about data quality issues, right?

Scree (00:16:10): Yeah.

Rob Collie (00:16:10): Depending upon where you got the data, your model can conclude we're all dead already.

Scree (00:16:16): Yeah, or China still has only had 6,000 COVID cases, or something. Like, it stopped. They don't have COVID anymore.

Rob Collie (00:16:24): I had a high school teacher in history class that told us, "There's so many problems with history, folks." At the very beginning, he went through all the different flaws in the story. He's like, "People forget. People write things down wrong," whatever. And then he goes, "And also, people lie. They just lie." And you never really know what you're getting. We weren't there to witness it, so we're trying to piece together what actually happened. That's certainly true with all the various different places you can get COVID data.

Scree (00:16:51): Yeah.

Rob Collie (00:16:52): Look at Florida. Wow, they're doing great.

Scree (00:16:54): But I do agree with your original ... Like, if you're not working on the stuff outside of work, is it really something you're passionate about? And if you're not really passionate about it, are you really better than someone who is?

Rob Collie (00:17:05): Yeah.

Scree (00:17:06): And I question that. Again, I don't expect it out of teammates and peers. But it definitely sets them apart when they do, right? They've probably run into more things. It's a matter of experience. The joke right now in the Power BI LinkedIn is everyone's looking for five to 10 years Power BI experience, when Power BI's only been out for really five years. Six maybe.

Rob Collie (00:17:27): Yeah. What a silly metric by the way. Years of experience. You know? It goes back to the gaming metaphor, right? You need someone just to sort of put in some time and grind and grind, and grind, that's different than someone who's actually really good at it.

Scree (00:17:40): Yeah.

Rob Collie (00:17:40): Off and on, gosh, I think I took a 12 year break, or an 11 year break from gaming, really of all forms. 11 years clean is what I call it. 11 years clean of World of Warcraft. And when I went back, I went right back to classic, which is the thing I'd been playing before when I quit. And I'm no better than I was, right?

Scree (00:18:01): My wife would be lucky if I was 11 days clean. So, kudos.

Rob Collie (00:18:05): Now, my life doesn't support, at all, the idea of being part of a really active guild that's doing these 25, 40 person deals that run for hours, and you're not really sure when they're going to end and all of that. That is not compatible with my life anymore, at all. You were saying you weren't formally trained and all that. And I want to say, who cares?

Scree (00:18:25): Yeah, even Power BI. Who gets formally trained? Five years ago, there wasn't any formal training. Back then, you just did it, and if you didn't, you found someone who had done it before, and you learned. I was there at the ground floor and there was no formal training. It would've been irrelevant because the application has grown and expanded and evolved by leaps and bounds. Every six months there's a huge release or feature set that comes out that just changes the game on its heels. I mean, it helps if you involved five years ago, and now you understand all those changes and what they were. And maybe nowadays, the formal training's probably necessary because the app has grown to that point where I don't understand how you pick it up on your own. There's just so much to it.

Rob Collie (00:19:10): Well, surprise, surprise. You're not going to find me to be a big believer in formality. Let's say we looked at all of the different training and instruction options available in the entire universe for Power BI right now, how would we go about defining whether one of them was formal or not? When P3 runs a three day training that is open to the public to sign up for, as opposed to the ones we do privately for individual companies, is that formal? I don't know. I mean, it's certainly real.

Scree (00:19:40): To me, I'd say formal training would be, yeah, anyone putting on an actual training that has some sort of professional experience. Yeah, I don't think you have to get more specific than that definitely.

Rob Collie (00:19:52): Because the other end of the spectrum is something you might not have been exposed to. I call it the training factories. There's a small number of people behind the scenes that are just churning out training modules on topics that they have no experience in whatsoever.

Scree (00:20:06): Yeah. Yeah, the Power BI grifter. Yeah.

Rob Collie (00:20:08): This is by no means limited to Power BI. It's everywhere. I watch these people operate. They're like, "Oh, what's new and hot today? Okay, I'll go make a training for that." I go look at the training. I'm like, "Well, this person's never touched this in any sort of real environment." However, they make a ton of money selling those scripts, those modules to all of these training centers throughout the country. And so, if you want to go get some worthless certificate that says that you've completed a Power BI training course, you can go get it, and you would've gained a lot more with your time there if you'd just been screwing around with guild data, or NFL ... Whatever. Some sort of personal hobby dataset.

Scree (00:20:45): Yeah.

Rob Collie (00:20:45): Whatever time you spent following the flow chart. And it's not even a flow chart, right? It never branches, these training scripts. I guess for me, I start getting the alarm bells. Oh, well I went to a training center and I took one of these training factory classes and learned nothing. In fact, probably learned some really bad things that were net negative for my understanding of the tool.

Scree (00:21:05): Yeah. I mean, I've run into some of those just doing what I do, consulting. In larger companies, it's more obvious where they've held a company-wide training on Power BI and you get people who are making reports and breaking capacity all the time. It's usually those people who are to blame because they took that training and they think they know what they're doing, and had no understanding of the underlying stuff because it was just a cursory glance at what Power BI does. But that's common. You're always going to have different levels of skills.

Rob Collie (00:21:33): Let's rewind and say as much as 10 years ago, I was providing formal training on Power BI. Instructor led formal trainer. I just didn't call it that. I didn't wear a tie.

Scree (00:21:46): Yeah.

Rob Collie (00:21:48): No tie required for this. So, along the same line, I was recently asked to help a recent college graduate who's sort of a distant relative of mine, with his job search as a CS grad, computer science grad. And started with his resume. And as I talked to him, the thing that I was most interested in, that I thought was the best signal that he could send to potential recruiters who were reading, scanning through resumes at 10,000 per minute with OCR, or whatever. He had done this project. He had built a Discord bot for he and his friends to make real time wagers against each other like who was going to win certain games and things like that and keep score and everything. I'm lik, "Oh, no, no, no." And it wasn't even on his resume. That needs to be first. Not in your project list, right? Not the thing that you were forced to do in your senior project course to get your degree, because you look at that and you're just like, "Oh, this reads like every other senior project course ever." Is just so vanilla.

Scree (00:22:46): So, I've actually heard that a lot of software engineering firms will actually scan resumes for GitHub addresses.

Rob Collie (00:22:53): Wow.

Scree (00:22:54): And if you don't have one on there, then it shows that you don't really have a portfolio of work that you could show off, which is how you would basically show your familiarity with subject matter off of a resume.

Rob Collie (00:23:07): I need to circle back.

Scree (00:23:08): So, all side projects should go into that GitHub repository and link that in your profile. You're going to find hiring managers who are going to want to go, "How does this guy program? Does this code even look good? Is this someone I can train?" So, he should probably spend some time cleaning it up because I've been in his shoes. Does he comment his code? Probably not. Is it one big file, or should it be multiple files? Those are the kind of things that maybe he could spend some time on while he's looking for a job.

Rob Collie (00:23:35): Hey, you know, that's why we do these podcasts, right? I believe all of those things. Those things you just said, they make total sense to me. He submitted his resume to a bunch of places, the big places, and didn't get any follow up. I'm thinking, well there's got to be something, either a red flag in your resume that's disqualifying you. Or a lack of a green flag, something that they're looking for that's not there. GitHub address.

Scree (00:24:00): Well, it could also be school.

Rob Collie (00:24:01): Yeah, that's true.

Scree (00:24:03): Maybe. I'd be really interested to see how many local school people they hire at Amazon. It's probably none. Well, Amazon maybe does it. But Facebook and Google.

Rob Collie (00:24:14): It's so weird. It's such a departure from the Microsoft that I joined. Famously, Gates didn't even graduate. Right? They never asked for my transcript. Never cared about my grades. Never validated that I had graduated. In fact, my office mate when I first got there, hadn't graduated and hadn't told anyone and was just still finishing his coursework. And it was more of a prove it mentality, as opposed to pedigree.

Scree (00:24:41): Yeah.

Rob Collie (00:24:41): I really like that. We have gone with that mentality at P3 completely. Pedigree means nothing. What you can do is everything. And that cuts both directions. You can have a great pedigree and be relatively not competent.

Scree (00:24:55): Yeah.

Rob Collie (00:24:56): And not curious. So much of what we do relies on empathy and humility and emotional intelligence and real world connection with other people. And also, confidence and humility at the same time, the perfect blend of it. And these are things that there are some 22 year olds that absolutely possess these qualities. Hats off to them. I didn't possess these qualities at least until my late 30s.

Scree (00:25:24): Yeah.

Rob Collie (00:25:25): We have people who work at the company who are definitely younger than that. But it just took me a while. I think refinement and experience is crucial to us. And frankly, people at our company, we support them, we onboard them. There's plenty of support behind the scenes for anybody who's running into difficult problems. But there is no shallow end of the pool at P3. We don't get to dictate to our clients how difficult their problems are.

Scree (00:25:50): Sure.

Rob Collie (00:25:51): And we don't even really know going in. So, basically everyone's got to be a black belt on their first day.

Scree (00:25:57): I think it's just more to the bigger picture of there's not 100,000 five year veterans with that kind of skillset like you guys have.

Rob Collie (00:26:08): Yeah.

Scree (00:26:08): And what's your solution if you can't find any? Just not provide that service? I think companies have to find a way to provide that service if there's a demand for it. And so, I think it's going to require a big shift. I mean, Power BI has been growing leaps and bounds. The fact that it's tossed in now for clients who already have a basic Office package for their corporate environment are finding it's a more cost effective solution than something like Tableau.

Rob Collie (00:26:34): Oh, yeah.

Scree (00:26:34): Why would you waste the money paying for two different packages when you already got one of them included. So, I see a lot of companies switching to Power BI and then struggling to have any meaningful talent in it. If they weren't working on it before, and these people are long time veterans in their company, well they're certainly not going to have the experience ready to go when you guys pull the trigger to move. So, that seems to be where I'd say, I don't know, out of my last five clients, there's probably four of them who have switched to Power BI and just don't have a huge amount of talent in it.

Rob Collie (00:27:03): And of course, this is central to our business model, is that we have a concentration of talent that is rare. And increasingly, we're rethinking the way that our marketing activities work. We call it grow dev internally. I'm a big believer in verb over a noun marketing. Well, we could just sit around and do marketing. What does that mean? Let's go grow the company, right? And for the longest time, that has meant client growth. Finding new clients that we can help. It's increasingly become both. It's a supply and demand. We also need to be turning our attention to recruiting and it's the same sort of digital funnel with different twists and turns in it for sure. But we're in the envious position, or the good problem to have. We are now finding new work quite a bit faster than we've even been hiring. I think we have the riddle cracked. I think we can continue to hire the right kind of talent at the pace that we need. It's just that we need to invest more specifically in that activity than what we have in the past. And so, this is an exciting new problem for 2022.

Scree (00:28:13): It's just a tough time to find talent. I mean, we've had an open call for Power BI talent and Tableau talent for months and months, and months, and we just haven't been finding anyone. I think we brought on someone a couple weeks ago and I was kind of flabbergasted when I interviewed her that someone like this was available because we've been looking and we couldn't find anyone.

Rob Collie (00:28:32): Yeah.

Scree (00:28:33): It's just a weird time. And I think it's probably more true of software engineers, because there's so much more competition for being able to move anywhere, remote only work. And it's just kind of turned the field on its head because why live in San Francisco and pay $4,800 dollars for a closet when I could go live in Idaho in a mansion and pay the same amount, and make the same amount of money doing it. So, I think it's turning the whole industry on its head, and it's making job searching and hunting for talent so much more competitive. And I'm sure that's true of every industry. But some of the more technically inclined fields, it's probably tougher than others.

Rob Collie (00:29:07): Yeah, COVID, obviously is a bad thing. But we were forever remote only. And that was a big advantage for us for a long time. We were a 100% remote workforce. Before it was cool. Before it was COVID cool. If you looked at a map of where our people lived, we certainly had employees in big cities. But relative to population density, our employees were not concentrated in big cities the way that you would typically expect, just based on population, right? And so, yeah. People who were, sometimes even two hours from a major airport were on the payroll and they're awesome. And those same people who live in those non metropolitan locales, there are a lot more employers competing for those same people, for sure.

Rob Collie (00:29:50): All right. So, this usually one of the very, very, very first questions I ask somebody. But we started with games and guilds and Undead Lords. We had to go there first, right? So, computer science AP class in high school. I did not do that. It was available to me, but I didn't. I showed my stripes early. Mm-mm (negative). No. Uh-uh (negative). But when did you start working with data in a way that you would consider where you first discovered the itch, like, "Oh, I could do this?"

Scree (00:30:18): So, I've worked in sales for a significant portion of my professional career. Both at Best Buy and then at a bank that's based out of Chicago. Best Buy was developing into this as I was working there, and I worked there for almost 10 years. But they started coming out with metrics and stats of sales level, employee level performance, kind of showing you what you've sold over the last month, and attachment rates, et cetera, et cetera. And as a sales department manager, I started tracking performance of our teams and just getting a little bit of a taste of what it looks like to actually turn register transactions into some sort of data that gives me ... I mean, for context, before this existed, you basically had employees writing down what they were doing and hoping that that was accurate or even truthful at the time. There was no way really to verify it short of printing out hundreds of receipts for every single sale that an employee did. And it was just too monotonous to measure and maintain that.

Scree (00:31:20): But we had several times throughout my career where they tried doing that. Sales manager come with this brilliant idea. So, when it finally became automated, it started to make the whole process a lot easier. And it just opened my eyes to, okay, now I know who's really bad at selling X. I can go coach and train them to get them better at that. And I don't have to worry about the data integrity piece, which is what we probably spent most of our time worrying about, which wasn't even something that should have been something we wasted a moment of effort on. But again, the data integrity piece was so critical to be able ascertain who's doing what, I think we just lost the road through the forest there.

Scree (00:32:00): So, near the end of my career at Best Buy, we did start automating the process and the data became so much more eye opening to me as to, "Oh, yeah. Now we can actually use this." Course correct employees and train them in the right ways. And when I went to the bank, we did a lot of the same things there. All the personal bankers were weighed and measured based on performance. And you were given a ranking out of the whole company, and everyone was trying to compete to get to the top. So, again, that data piece was just always constant in my peripheral vision. So, when I moved on from that and went into E-commerce, that's where the data metrics piece, it was overwhelming. If you work with Amazon at all, like a seller, they have data on things that I'd say 90% of the people have no clue what it is or how to use it correctly. So, I worked for a firm that actually did that kind of work with them and said, "Hey, here's the important metrics you need to understand. Here's the things that you have to focus on." Running of reports and stuff like that. So, it's been an evolution throughout my career seeing data and seeing the value of data.

Scree (00:33:04): And then, obviously my side projects have gone into the whole how you gather data and collect it and store it. So, it all kind of came together there, I think at the end as Power BI was really kind of rocking and rolling.

Rob Collie (00:33:17): Going back to the Best Buy story for a moment, if you can remember. This might not be the resolution that you record at those moments. But, maybe. When you go from a subjective measurement of your employees or a team in terms of how well they're doing, it's so heavily weighted towards, as you walk around the store, you witness a customer interaction, and that one interaction weighs really heavy in your picture of how this employee operates, whether positive or negative. Right? It becomes this out sized dominant piece of information. And then also the way that people talk to you, the way that they are able to portray themselves to you, of course, you're vulnerable to being mislead there. And then on the flip side, people who don't represent themselves well, right? So, do you remember being surprised at what the objective data said and how different that was from your subjective opinion at the time?

Scree (00:34:08): Yeah, especially when the automated system flipped in and we were getting real time data as to who was doing what. I remember that first month just being blown away by it because you're right. When we would coach and train an employee, we'd follow them and shadow their interaction with a customer. And so, we'd be listening to the whole thing, kind of secretly and not so secret. Because you'd see this manager hovering around the interaction. And it wasn't a good indication of how the employee operates when you're not there. They would always do everything that they were required to do, short of them forgetting something, which inevitably, they always did. That's the red flag, right? If they forgot to do something when they're being watched, chances are they're not doing it when they're not being watched either. But you didn't think about it at the time. You just chalked it up to maybe being nervous that they were being observed.

Scree (00:34:55): But, yeah. The datasets were completely different. We'd have people that were turning in numbers that didn't align with what was showing up automated. And I remember there being a lot of turnover at that point, in the coming months because the people who were casting themselves in a better light weren't actually doing that good. They'd have one interaction per day that was stellar and the rest were complete garbage. And they'd show off the one really good thing. And then you'd run around and brag about their performance on that one really good transaction, and then they'd secretly be on the other side of the store just ringing people up without offering them anything, like all the Best Buy stuff that you're supposed to offer.

Scree (00:35:30): Yeah, it was kind of eye opening. And it was great. It was like the big brother effect, right? You're able to see the total picture, and before that we couldn't. It was eye opening to me. And it helped, honestly, a lot of our employees who weren't good at sales admit that because they couldn't hide it anymore, and then they would come forward and ask for help. But then there were some who didn't really care to improve. They were just there for a paycheck. And so, those employees have to go too at some point.

Rob Collie (00:35:59): Yeah, it's like the old prisoner's dilemma thing again, right? It's in everybody's best interest to simultaneously come forward and be vulnerable and admit that they're not very good at things. But if only one of them does it, then it's not going to go well for them. You know? And so, no one does. But then, when you level set the objective reality for everyone, there is that moment of relief for some people. I can imagine that they are able to improve and be helped to improve.

Scree (00:36:26): Yeah.

Rob Collie (00:36:27): Yeah, I wouldn't have thought about that. So, sometimes the big brother thing isn't all bad.

Scree (00:36:32): It is and it's not. So, I was there at Best Buy during the 2009 recession.

Rob Collie (00:36:37): Ooh, fun.

Scree (00:36:38): So, you've got sales goals that were generated a year before. So, completely ignorant to the whole the economy sucks now. Everyone's fired. And they're like, "Well, you're not hitting your goals." And I'm like, "Yeah. Have you looked at the store? There's only three people in the store." Well, go sell to them. Just cognitive disconnect because the upper leadership's wondering why all of a sudden we're not hitting goals. And it's like, "Are you paying attention to the news at all?" It was harsh because then you're able to actually see line level employees who aren't generating sales. But is it really their fault? It was just a weird time. I wasn't a fan of being held responsible for economic situations. And that's really what we were doing, is holding employees to that accountability, which is nonsense.

Rob Collie (00:37:22): That's no good at all, no. Almost like a new principle just came to me as you were telling that story which is thoughtlessness gets magnified by layers of management.

Scree (00:37:32): Yeah.

Rob Collie (00:37:32): It's just so easy. Default state is thoughtlessness. Just hold the wheel steady if that's the mentality. As that mentality flows down an organization, it only gains momentum.

Scree (00:37:43): Yeah. And one of the offshoots of this whole thing was that we had all these middle managers who kept visiting our stores. Well, Best Buy decided that, maybe the middle managers don't really serve a purpose. If we're going to miss our goals, then they're not really useful. So, basically got rid of most of the middle managers. And I remember when I learned this, I was wondering why they hadn't been coming at us as hard about missing goals. And it was because the people that they were accountable to weren't there anymore. So, the pressure stopped flowing down hill when they made these changes. But never truly went away. And then, shortly after the recession, it was the whole people were using Best Buy as a showroom. They'd walk in with their cell phones. This was the big new thing. They'd look at what they wanted and then go price shop and you'd inevitably find this stupid site, BNH of Maine or something. And it was every TV was 50% off with no taxes. And then people would be like, "I want this for this price." And we're like, "Get lost." Because you'd have to pay like $1,200 in shipping to get it. So, it was like go for it.

Scree (00:38:46): It was so prevalent that we were coached and trained on how to sell to people who had their cell phones out. And I'm like, at that point, are they really going to pull the trigger from us? But, we were able to convert some of them.

Rob Collie (00:38:59): I had a client years and years ago that one of their lines of business was shopping at Best Buys and places like that on behalf of the manufacturers. Does this sound familiar?

Scree (00:39:13): Sort of.

Rob Collie (00:39:14): They would go in there to see if all the Best Buy stores, for instance ... Best Buy wasn't their only place that they deployed these anonymous shoppers, right?

Scree (00:39:22): Oh, yeah, yeah, yeah. Secret shoppers, they call them. Yeah, yeah. Yeah, we had those. Yeah.

Rob Collie (00:39:26): But this was very specifically aimed at verifying if the, I don't know, the new Sony TV was being displayed properly.

Scree (00:39:36): Yep.

Rob Collie (00:39:37): Was it playing the right video? Was it on the end cap as agreed to in the contract?

Scree (00:39:44): Yeah.

Rob Collie (00:39:44): Were the fliers out or whatever, right? And then, eventually, all of that human intelligence flowed up into spreadsheets that these people I was working with were having to report on. And then, they needed to use Power Pivot at the time to help bring that whole process under control. Were you ever on the receiving end of a manufacturer complaining about the ...

Scree (00:40:05): When I first got hired at Best Buy, I was actually their marketing manager. So, I would set the end caps and stuff like that. So, I was directly exposed to these people. They'd come to me and say, "Hey, this isn't supposed to be here. Can you remove this?" Because they couldn't touch anything, so it was always, "Can you do this? Can you do that?"

Rob Collie (00:40:21): Yeah.

Scree (00:40:21): And then I'd have other people who were like, "Can you just sign off that I was here." I would just have to sign their stuff. So, yeah. Those people were fine, because ultimately they didn't care. They were just collecting a paycheck. They weren't super picky. If something was wrong, they'd just write it down and we wouldn't even hear about it. And to be honest, we would never even hear from Sony saying, "Hey, your end cap is improperly set." So, even if it was wrong, we never heard anything. So, I don't know what those people were getting paid for or where that message was going to.

Rob Collie (00:40:48): Oh no.

Scree (00:40:49): But we did have secret shoppers, people who would come in and act like they were buying stuff from us. And then, they would buy stuff and their entire motive was to find out if you offering everything that Best Buy should be offering. And so, were you wearing your name tag was the big one. Were you able to identify this person by their name tag? And you got reamed by corporate and senior management within the store if you didn't have your name tag on. Did you offer them a warm greeting. It was just like spot checking. And we'd always get a kick out of it when a manager or an assistant manager got interviewed by these people and didn't do, and then they'd throw it in the garbage because they didn't want anyone to know that they didn't do what they were supposed to.

Rob Collie (00:41:33): Reminds me a little bit of the flare in Office Space.

Scree (00:41:38): Yeah.

Rob Collie (00:41:39): 12 pieces of flare is the minimum. But this guy wears 20. Well, why don't you just make 20 the minimum? Ah, I don't think ...

Scree (00:41:48): And it was always so random too. We'd get two visits of these a month. Sometimes we'd have the same employee get interviewed twice and you'd get a score at the end. And it was just a kudos piece, a coaching piece. Hey, this person sucked. Can you go coach and train them? But if that employee was our number generator in revenue for the store, based on the actual data we had, then we just basically ignored those. I don't know. It was one of those weird, cognitive dissidence pieces where you'd look at the data versus this random one off interaction, because a lot of times these people didn't know how to actually role play with a customer. So, it was just an awkward interaction to begin with.

Rob Collie (00:42:24): Oh yeah.

Scree (00:42:25): Because they're faking like they're buying something, and that they're generally interested in what you're trying to sell them.

Rob Collie (00:42:30): That's right. They might not be interested in that thing, and it's not their money that they're parting with.

Scree (00:42:36): Right.

Rob Collie (00:42:37): It's very tricky to make it real.

Scree (00:42:40): You'd get some very non social people. And this wasn't a high paying gig.

Rob Collie (00:42:45): Well, that's realistic.

Scree (00:42:46): I think they got $20 bucks per mystery shop. I don't know how you make a living on that driving around to all the stores.

Rob Collie (00:42:52): I mean, this is the economics of operating at scale in a people-centric industry, is amazing. And this is why I am so blown away by Starbucks. In the way that McDonald's is often talked about, and they'd even self identify as more of a real estate company than a fast food company. I think Starbucks is an HR miracle. Think of the complexity of what the team at a Starbucks has to juggle at all times. Are you a Starbucks client? Do you frequent Starbucks?

Scree (00:43:28): No, I'm a Red Bull client.

Rob Collie (00:43:30): Red Bull client. Okay, that's a choice. We just choose different delivery mechanisms.

Scree (00:43:38): Basically.

Rob Collie (00:43:39): There are, factually speaking, probably like a billion different combinations of things that someone can order in a Starbucks. Even if you're just limiting it to one drink, there's got to be a billion possibilities.

Scree (00:43:52): Yeah, I believe it.

Rob Collie (00:43:53): And some of them are on menu. Some of them are off menu. And the pace at which they have to operate, sustained pace, it's unbelievable. And I think with COVID, it's even gotten worse. I think Starbucks are more busy now than they've ever been, in a weird way.

Scree (00:44:09): I believe it.

Rob Collie (00:44:09): And the clientele at a Starbucks, let's just say that it skews towards entitled.

Scree (00:44:17): Yeah, I'd say so. You're paying $14 dollars for a coffee, yeah.

Rob Collie (00:44:23): The incidents of jerks that you have to deal with. And also, by the way, these aren't just people in their natural state. They're also people who are in a hurry. And it's just so easy to get it wrong given all the combinations and everything. And the complexity of all the orders. The intensity of the customers and their expectations and everything. And every time I go in a Starbucks, I just stand in awe. Starbucks baristas I think are ... I don't know what they're paid. Ignoring pay for a moment. The fact that it's even possible to have this many high performing people deployed in this manner at a given point in time in the universe is just incredible. I think there are entire businesses that could be formed just on the idea of going in and poaching baristas from Starbucks.

Scree (00:45:14): I looked it up.

Rob Collie (00:45:14): All right, what do we got?

Scree (00:45:15): Less than $13 dollars an hour, national average.

Rob Collie (00:45:19): Really? Even now? My son went and got a job, his first job ever this summer at Home Depot, and they paid him $15 an hour.

Scree (00:45:26): And it's national average.

Rob Collie (00:45:27): Oh boy. There's got to be better benefits or something.

Scree (00:45:31): Yeah, I imagine they have decent ... I mean, it seems like they have got benefits. But still, yeah, you'd expect them to be paid more.

Rob Collie (00:45:37): And whatever the number is, I'm sure it's not enough. I don't think it's just a training thing. I think it's also their ability to screen. They find just amazing people that, I would say that half the people that I interact with at a Starbucks are people that I wish I could hire in some capacity. That's the impression I come away with. Just really, really, really impressive. Maybe that doesn't apply to secret shoppers at Best Buy.

Scree (00:46:00): Probably not. But it is an impressive achievement that they've been able to find people who are willing to do that for as little as they seem to be making.

Rob Collie (00:46:08): Well, Scree, I think you should temporarily suspend your Red Bull habit just for a week, and develop yourself some sort of really obscure boutique order. If you need me to just come up with one for you, I'll give you what I usually order. Go in, slap your money down on the barrel head, and tell them, "I want a quad grande one pump mocha, please." That's a relatively simple order. It doesn't have that many words in it. I think it's a six word order. I saw on Facebook the other day a meme that if your coffee order has more than five words in it, there's something wrong with you. And I'm like, "Damn, I've got six." Just barely over the threshold. But then just watch the system spring to life and watch all the other things that they're doing. Watch all the things that they're juggling, and even avoiding running into each other. And it's quite an operation. And I think most people take it for granted, but I just stand there in awe every time. Just, wow.

Scree (00:47:01): Yeah.

Rob Collie (00:47:02): So, back to Best Buy for a moment. Those reports, they were just coming at you. You weren't part of preparing them?

Scree (00:47:07): Not at all. Yeah, it's all internal behind the scenes stuff. We would just refresh them.

Rob Collie (00:47:11): But it would still give you, a light bulb would go off for you. Like, "Oh, this is different." This objectivity and automation. Did you have any idea what sort of tech they were using, or was it just built into the point of sale registers almost?

Scree (00:47:26): I don't know what backend they were using. I couldn't even have told you. It was all obfuscated to the point where there was no recognizable anything on our end.

Rob Collie (00:47:34): There was probably reporting services or some equivalent.

Scree (00:47:38): Yeah, I'm sure. But, yeah. I wish I could've seen it. But, no. I never actually got a chance to dive into that piece with them. I jumped into a gaming role with them. They launched a gaming department, and within six months it had flopped. And so, they offered all the gaming managers severance packages and I took it and left. I fled. I was like, "I'm done with Best Buy as a career." And it helped me get out of that whole retail sales industry. It was awful.

Rob Collie (00:48:04): Escape velocity.

Scree (00:48:05): I had started dating and then was about to get married, and I was like, "I need to get out of here." I can't do the whole random schedule every day and not see my wife most nights. So, I'm like, "I'm going to go find a job in banking because they have the greatest hours." You know? And it took me four or five months. But it's still a sales job, effectively. I mean, banking is still sales. So, got in and did great there for several years.

Rob Collie (00:48:31): Having business accounts at banks exposes me to a level of sales coming at me from the bank that I'm unaccustomed to as an individual banker.

Scree (00:48:44): Yeah. It's just a different type of sales, and you just had to know your clients well enough. There were times you didn't want to sell to them, if you were just building a relationship. If they're just opening an account, chances are it's not the right time to offer them the entire gamut of services that we offer. Get them what they came in for, set them up, contact them again a week later, and then work from there. There were tips and tricks that I learned through my time there. And honestly, in some ways, banking was easier than retail sales. I didn't have to waste time answering what an HDMI cable is, or a UBC cable because everyone would mispronounce, and I'd just look at them confused for 10 minutes. The tech support piece was removed from my plate and it meant more honest conversations with clients. So, I preferred it in the long run.

Rob Collie (00:49:33): Yeah, it's a bit more straightforward transactional rather than upsell. That makes sense to me.

Scree (00:49:37): Yeah.

Rob Collie (00:49:38): So, at the bank, you were also, I would assume, were on the receiving end of reports, rather than the construction end.

Scree (00:49:45): Sometimes. When I was there, they actually launched Salesforce. We transitioned our entire client management system over to Salesforce. So, I was a super user that piloted this rollout. And Salesforce had all sorts of interesting information embedded within it that we didn't have access to before. And I kind of data mined the shit out of it. So, I knew when mortgages and home equity lines of credit were coming up and needed to be renewed. And so, I had lists of every loan coming due in a hundred miles of our location and I had reached out to all of them. So, just stuff like that, that I knew no one else would have any inkling of how to do it because they just aren't tech savvy.

Rob Collie (00:50:30): That's so cool. That had to have been a big transitional moment for you. A number of years on the receiving end as a consumer of these reports and seeing the value of them. So, therefore, developing the appetite. Developing the sense of business value. And then, one day someone shows up and says, "Oh, look, here's the menu."

Scree (00:50:51): Yeah. I do remember getting my hand slapped down, though, pretty hard, because I was stealing loans from other branches.

Rob Collie (00:50:59): Too effective. Too effective.

Scree (00:51:00): The bank manager started catching on that I was calling on their clients and were like, "What the fuck?" So, they had conversations. They're like, "How are you getting these lists?" And I was like, "What are you talking about? These people naturally just wanted me to call them." The gig was up at that point. And so, I showed them how I pulled my list. And then every branch manager had to run their own list and distributed it to their employees. Said, "You get this client, you get this client." And so, after a few months of me stealing half the sales in the region ...

Rob Collie (00:51:34): On net, they won out from this, right? Because they learned new behaviors as a result of what you were doing. Well, we should all be doing this and making rules around it.

Scree (00:51:44): Yeah. And I think ultimately, being able to see loan data inside Salesforce was an oversight on their part.

Rob Collie (00:51:51): Yeah, probably.

Scree (00:51:52): But it was meant so that if you went into a customer's profile on Salesforce, you could see all of their accounts. So, you would know that they have a business account or a personal account, or a credit card, or et cetera. And so, you wouldn't have to go cross reference it with a different system every conversation you had. You would have a customer call in, you'd pull up their Salesforce, and you'd be able to see everything at a glance. That was the intent. Now, the sad secondary effect was that all the accounts were in Salesforce and it made it really easy for me to run queries against it. Salesforce was built as a super simple tool to run searches and find clients that meet certain patterns. And so, I was just ahead of the curve there. And it helped that I was kind of tech savvy going into it. You had a lot of older banker types who were 50, 60, who didn't want to use computers in the first place, and I'm eating their lunch month after month because of the insights that I had at that point.

Scree (00:52:49): But at the same point, there were also some really funny moments where I would have a conversation with a customer and they were nuts. Like someone's calling in asking about UFOs or something, and can a UFO get into my account. My modem's making a noise. Does that mean someone's hacking my bank account? And I'd be like, "What did I just walk into?" And I'd go into their Salesforce profile and there's no notes. So, I would leave these really detailed hilarious comments so that if anyone ever went to that client's profile again in the future, they'd get to see the nut show that the person was.

Rob Collie (00:53:20): Yeah.

Scree (00:53:20): And I would get random texts six, eight months later of people cracking up reading my notes. So, small little pieces of joy like that, that come back tenfold. He was like, "That guy is nuts. I was reading your note, and it made me laugh."

Rob Collie (00:53:36): Can UFOs get into my bank account? Well, if they want to.

Scree (00:53:41): Yeah. I was like sure.

Rob Collie (00:53:43): If they can traverse the galaxy and get there-

Scree (00:53:45): Well, you're not a good bank then if you let UFOs into my account. Well, yeah. You know, you should probably take your money elsewhere. Sorry.

Rob Collie (00:53:51): Yeah. Go find one that advertises as UFO proof.

Scree (00:53:54): I mean, you could just tell them you wrapped the safe in tin foil or something, and he's probably fine. But I'd say Salesforce was probably the first time where I actually had the opportunity to interact with the data on my own, on my own terms. And that was just further evolution of my course towards data. And that's where I started my MBA when I was at the bank.

Rob Collie (00:54:12): Okay.

Scree (00:54:13): And I kind of knew that that's the direction I was going to go, was trying to pick up courses that were data based or data oriented to the extent that I could. I mean, you only had like four elective classes out of the whole program. So, three of the four were in that field.

Rob Collie (00:54:27): Just so I can background this against other technologies and where they were in their evolution, roughly what timeframe are we talking about here when you were at the bank and doing your MBA?

Scree (00:54:37): You know what? I'm going to cheat. I'm going to go look at my LinkedIn profile and tell you.

Rob Collie (00:54:41): But see, go to the objective source. None of this subjectivity. This is what we've been talking about. This has been a theme of the whole conversation.

Scree (00:54:46): 2013 through 2017.

Rob Collie (00:54:49): Okay. So, in 2013, I was full on in on Power Pivot. In the early days of the blog. I was writing two articles a week and sustained that for some number of years. I was already way, way, way in. But the market didn't really know about it. Just going back to Salesforce for a moment, just out of curiosity, how far were you able to get just using the Salesforce UI to construct queries and reports, versus how often did you have to dump it to Excel to get where you needed to go?

Scree (00:55:19): I created spreadsheets that we could share with our team. So, eventually, like I said, when the gig was up and we actually had to start tracking this on our own, it was up to the individual managers to assign and keep track of when loans that were maturing needed to be called on. And so, I converted a lot of those reports into Excel file formats. But that was merely just out of ability to share them.

Rob Collie (00:55:41): Just portability.

Scree (00:55:42): Yeah. Yeah. But in terms of Salesforce, we didn't really interact on any level beyond just using their UI.

Rob Collie (00:55:50): Yeah. It's when you get into, "Oh, I need to cross reference this report with that other report. Or there's a calculation I need to perform that that field isn't in Salesforce yet." Those would be the times which you'd have to dump for further analysis.

Scree (00:56:03): And it was funny because at the time as a super user, I had access to probably more features than eventually we got paired down to. And I was working with this one guy who was the manager in charge of this transition, and he ended up going to work for Salesforce. So, I thought it was interesting, his evolution, watching him go off. I never kept up with him. But it was just interesting to me to see someone who was responsible for rolling out Salesforce to our company, end up going to work for the bigger team. But he always helped me run custom reports if I needed them. If I ever wanted to say, "Hey, is this something that's doable?" He'd be like, "Yeah, sure. Let's do it." And we'd run crazy stuff to just get an idea. We were trying to isolate when people didn't have accounts, which was slightly harder to look up in the Salesforce than people who did have accounts. The not wasn't really a function baked into the UI.

Rob Collie (00:56:56): Yeah. Yeah.

Scree (00:56:57): So, he was able to help us limit it by geography and location.

Rob Collie (00:57:01): Just dying now to use the DAX EXCEPT function against these two lists. I've really gotten into the EXCEPT function lately. That's one of those sentences you'd never really expect to hear. Let me tell you, I'm really liking the INTERSECT and EXCEPT functions these days.

Scree (00:57:20): You know it's funny because I see lots of time and effort spent on DAX and I've been doing this for five or six years now. And with one exception, I've never really had to do complicated DAX.

Rob Collie (00:57:34): Interesting.

Scree (00:57:36): If you focus all your time on making a great data model, the DAX kind of writes itself. Now, there's situations where you can't write the data model because it's already made and some of the DAX can get a little messy. But I've never run into a situation where the DAX was so convoluted that I'm scratching my brain for something.

Rob Collie (00:57:55): Well, it might be, without seeing the business problems and the datasets that cross your desk, I really can't know. Right?

Scree (00:58:02): Yeah.

Rob Collie (00:58:03): First of all, I completely agree that get the data model right and the DAX isn't hard. Get the data model wrong and the DAX is impossible.

Scree (00:58:10): Correct.

Rob Collie (00:58:11): That's certainly true, 100%. Secondly, something that's been coming up a lot in recent episodes of this show anyway, is just how much of a turnoff I would find on the internet today if I was going to try to get into DAX. The stuff I would run into on the internet today would scare the hell out of me. I would want nothing to do with it. It's crazy. So much of the DAXs on the internet right now is this hyper optimal, almost like ivory tower flavor that its whole philosophy is go in construct all of these virtual tables using DAX as a sophisticated query language. And then at the last second, collapse those virtual tables down with arithmetic. It's kind of like violating my noun and verb principle.

Scree (00:58:55): Yeah.

Rob Collie (00:58:55): The way I write DAX, the way that I learned to write DAX was very verb oriented. Go do this for me. Right? A calculate of a date's YTD, right? Go do this. Go give me the year to date sum of this. It's very command oriented. It's a verb. But it's taken on a very noun dialect lately. Go build this thing. It has a loose corelation with the answer that I want. But then, in the last minute with a flourish, collapse this thing that I've built, this glistening structure, collapse it down to a number. If I were completely new to DAX, I would go, "Oh my God. This is not for me." So, there's all of that. And then the thought I'll leave you with on this topic, the thing to chew on is that sometimes the reason why I find myself in complicated DAX ... Again, my DAX is not going to be as complicated as the ones I run into on the internet most of the time. But the reason I find myself using things like the EXCEPT function and the INTERSECT function are because my ambitions have been teased.

Rob Collie (00:59:59): If you're not hungry in the morning, you can walk around a long stretch of the day not hungry. Then if you take a bite of something, invariably, you're going to be hungry, really hungry 30 minutes later. You've awoken the beast. I think there's something like that with DAX as well.

Scree (01:00:15): I think it ultimately comes down to how the clients are ending up using the reports. And DAX is sort of a derivative of a desire to show something that's not immediately obvious from the data itself, the raw data.

Rob Collie (01:00:29): Yeah.

Scree (01:00:29): So, most of the clients that I work with, they're really fundamentally just getting started with Power BI. And so, they're trying to show their data in an efficient way. That isn't to say they're trying to show insights that that data is telling. They're just trying to show the data. And once you get into the insights piece, that's probably where the complicated DAX starts to flow out. And I just don't see that very often because we're getting a lot of those learning to walk people first.

Rob Collie (01:00:59): Yeah, yeah. Yeah.

Scree (01:01:00): The ecosystem that gets them to a Power BI report is also pretty convoluted. It's not just the Power BI site, but the data stores. How are you connecting? What is the transformations? How does that data get in there? And so, a lot of time and development gets spent on that. And that's what I think a lot of our clientele tend to spend that initial contract piece with is getting the data into a place that's usable and then converting that into a report. And the insight piece is one of those nice to haves down the line, but it's not something that a lot of clients have an ability to speak to right off the bat because they haven't seen the data in a presentable way yet. My guess is DAX comes into play more towards that second half.

Rob Collie (01:01:41): I agree. That very much is in line about what I was saying about ambitions. As your ambitions grow, the depth and the complexity of the DAX that you start to wield goes up. And it's an amazing language. We're fortunate, I think, in that we tend to attract the kind of clients who really are looking to ... They're not just looking to replace what they currently have. They're not looking at a new way to visualize numbers that they already have. They want to get to the metrics that they've never had, and they know that they're out there. They know that there's something, that they could be operating off of a much smarter way of visualizing everything. And it's not just about automating all the things that they're already doing. It's not just about getting the manual workout. Now, there is a lot of that too, of course.

Scree (01:02:25): Yeah. I think I've just had the downside of not getting the really fun stuff yet.

Rob Collie (01:02:30): Well, you will. You will, especially if you maintain an ongoing relationship with some of these clients over time. They will start to as they get the initial problem space as they conceived of it. As that gets addressed, some of them anyway, will naturally zoom back and go, "Oh, there's a frontier here." And they didn't even perceive the universe outside of that border until they populated it. Until they got it handled. For example, at some point, I will produce and share on Twitter a word cloud of the wordigamis from this episode. The words on this episode that had never appeared in an episode of Raw Data before today. And those words will be sized, by the way, in the word cloud, based on their rarity in the English language. Their usage in the English language. And the number of times they were used in the episode. So, now's the time to drop some vocab. You know? Let's get it into the cloud. You know?

Rob Collie (01:03:31): And of course, I could probably do that in Power Query if I was good at it. But I look at it and go, "Oh, I can do this in DAX." And this is where my good friends the except and intersect function come into play. And believe it or not, even on approaching million word dataset of all the words ever spoken and transcribed on this show, it's just lightning fast. It just chews it right up. There you go. There's your word cloud. It's so cool.

Scree (01:03:55): So, I've only been in the consulting area of Power BI since May. Before that, I was in a solutions firm that did E-commerce work, right? And so, I worked on the same product for years. And so, the needs of that were very different than client based. So, I haven't really worked with tons of clients yet. So, I think you're right that over time, we'll see them evolve from learning to walk to running, to the point where I need to get there. So, it's mostly my newness to the whole consultative piece that's limited by exposure to that.

Rob Collie (01:04:29): Yeah, and that's nothing but good news. You know? You've already got a toolset that's doing amazing things for you, and you're good at it. And to know that there isn't a ceiling on what it can do, or if there is, that ceiling isn't crowding in on your head already. There's head room to grow in terms of your capabilities and in terms of the value you can bring to people. It really is just a hell of a toolset that can address so much breadth and depth. I've said it before, I'll say it again. It belongs in the software hall of fame, if there were such a thing. Maybe they should open it up right next to the Rock & Roll Hall of Fame in Cleveland. I'm sure they would get the same number of visitors. Everybody loves the software hall of fame.

Scree (01:05:08): I like the GitHub repository notice that your data's going to be stored in a vault in, where is it? The Arctic. Have you seen this?

Rob Collie (01:05:18): No, I haven't. Is it right next to the seed vault to save humanity, all the seeds?

Scree (01:05:24): I guess. But if you ever commit work, it says, "Hey, your code qualifies to be stored in the Arctic." I think it's the Arctic.

Rob Collie (01:05:33): Really?

Scree (01:05:33): It's been a while since I've actually logged in. But it always cracked me up. I'm like, my code for the Undead Lords is being stored in the arctic because that's going to be hugely critical for mankind to bounce back.

Rob Collie (01:05:45): That's right. That's right. Yeah. I wonder what criteria they use.

Scree (01:05:50): I think it's just about everything.

Rob Collie (01:05:52): Really? I mean, I have one add on for Warcraft that I didn't even really write. A friend of mine is a VP at GitHub and wrote all of the guilds. Or he and one other partner in crime wrote very, very, very elaborate EGPG client server add-on for WoW. It's something else. Integrated between the guild website and the endgame add-on. Real time bidding. Real time scoring. It's crazy. And I was like, "Hey, I need an add-on so that I can make snarky comments in the raid." And the comments are all of a particular formula. There's only certain forms of these snarky comments that are just boring. It's just tedious to type the same comment over and over again. So, I need a push button interface that you can drop the right snarky comment at the right time. Again, that's the value that I brought whenever they took me on a mission.

Rob Collie (01:06:48): That code is probably in the arctic too. All that really was, was a string table that I filled in.

Scree (01:06:53): Humanity is saved.

Rob Collie (01:06:54): I know. I know. If you ever need, Rogue Chat 9000 is what I called it. My chief contribution for sure. What have we not talked about that we should have? Is there anything that we haven't gotten into? This is like the elective phase of the conversation. Oh, I never asked you, what games is the guild into right now? You mentioned there's a PVP guild. So, for people who don't know, player versus player guild.

Scree (01:07:20): Yeah, it's mostly people killing other people in virtual games. They're super aggressively playing New Worlds right now.

Rob Collie (01:07:29): Okay.

Scree (01:07:30): Amazon's New Worlds. I don't know if you've heard of it.

Rob Collie (01:07:32): Nope, I hadn't.

Scree (01:07:33): It's Amazon's first actually released game. They've had a few that got scrapped along the way. It's had a tumultuous launch.

Rob Collie (01:07:40): Well, I mean, any good game does.

Scree (01:07:43): Seemingly, it's the norm now. But yeah, we still got a good 30 or so people playing it.

Rob Collie (01:07:48): None of you are running around Warcraft.

Scree (01:07:50): I am not actively playing it. I had too much on my plate for work the last couple months. So, just no time and no energy.

Rob Collie (01:07:58): Just like me. Work life, professional life, personal life, and then, what? No, I am not treating this game as an IT problem, which is what you kind of have to treat it as. The number of interface modifications and macros and things like that, that my friends use in Warcraft is just insane.

Scree (01:08:18): Yeah. I used to play it and I remember spending almost as much time on my UI as I did playing the game.

Rob Collie (01:08:25): Yeah.

Scree (01:08:26): So, I was in the same boat.

Rob Collie (01:08:27): I don't need a development problem in my gaming.

Scree (01:08:31): You know what's interesting though about WoW and its history is that the game's mechanics in the game actually had to evolve and become more difficult because of the UI capabilities that the designers gave the player base.

Rob Collie (01:08:46): Yes. Yeah.

Scree (01:08:47): So, they had to make the end game problems more challenging because the players were creating UI solutions to all the problems that they threw at them.

Rob Collie (01:08:53): Yep.

Scree (01:08:54): And so, it became this ...

Rob Collie (01:08:55): An arms race.

Scree (01:08:56): Arms race.

Rob Collie (01:08:57): Yeah.

Scree (01:08:58): And I always found that hilarious, because normally a game developer would just say, "All right, well we'll just turn this off and then you have to play the game our way." But they didn't do that. They let the player base create tools. It's smart. I remember it came to a head during one of the expansions that I played in where someone wrote a mod that told you where to run to in the 3D space.

Rob Collie (01:09:23): Yeah.

Scree (01:09:24): So, it said, "This is safe. Run here." And that became so egregious because it basically played the game for you. It wasn't that something was happening that you needed to run from. It was, "Go here and you won't get hit." And that was so egregious that they ended up turning off some of that arms race functionality.

Rob Collie (01:09:44): Mm-hmm (affirmative). They're smart, right? I used to tell people on my team that worked for me at Microsoft that every feature take back, everything we took out of the product had an absolute value of impact. It's like 10X of every add, everything you added to the product. So, you've got to be really, really careful about take backs. Blizzard took the right approach. They opened Pandora's box, and rather than try to slam the door shut by neutering and negating and disabling KPI capabilities, they just made the game harder. And it's the same thing with as the play balanced over the years when they discovered that a particular class of character had an advantage over everyone else. The knee jerk response is to take away that advantage. To dial down or nerf the capabilities of that one class. And so, when they could, they didn't do that. When they could, they just introduced capabilities in other classes. They would balance by adding rather than removing.

Rob Collie (01:10:42): It's a really delicate, delicate balance. It takes master craftsmanship, essentially. It was something to watch. And then, some time after I quit playing, apparently the wheels completely came off of the whole enterprise and the whole game went in the tank. And I never got to witness any of that. I sold at the top apparently and got out.

Scree (01:11:01): I did too.

Rob Collie (01:11:02): But now we're riding that same rollercoaster back up to the top of the hill. I'm in the level 70 expansion right now, second time around. I'm looking forward to the level 80 classic expansion. And then we're not sure where they're going to go next.

Scree (01:11:18): I can't believe that they found as large of an audience for this as they did. It's the most unbelievable thing.

Rob Collie (01:11:25): It's insane, isn't it? Just imagine a game released in 2005 and people are still playing it today in basically the same form.

Scree (01:11:34): More people played it today than did at launch.

Rob Collie (01:11:38): Really? Oh my God.

Scree (01:11:40): I think that was what I saw.

Rob Collie (01:11:41): That's crazy.

Scree (01:11:42): But, the audience is bigger now, right? So, they grew the audience by leaps and bounds.

Rob Collie (01:11:48): I guess that's true. The people who were playing in 2005 weren't 80 years old, right? So, we're all still around. I mean, almost everyone I play with is so much younger than me that if they were playing at first release, they were like 13 years old.

Scree (01:12:02): Or not even born yet, which is the case with a lot of the people who are playing Classic right now. So, maybe that's their audience is the people who've been ripping on WoW being so easy, "Back in our day." And these people wanted to go get a taste of that. I don't know, maybe.

Rob Collie (01:12:21): Have you seen that they've now released WoW Classic: Season of Mastery. Have you heard of this?

Scree (01:12:26): No. I tapped out of Classic about, I don't know.

Rob Collie (01:12:30): So, that's like classic, classic. Season of Mastery is like all the rating counters have been slightly buffed up.

Scree (01:12:40): Oh, God. Yeah, no thanks.

Rob Collie (01:12:41): So, basically that they have the same level of challenge overall that they did at first launch before everyone figured everything out. Before all the interface mods got so good and all of that. Again, it's an add. They didn't take away the capabilities. They just made the raid bosses harder.

Scree (01:12:58): No thanks.

Rob Collie (01:12:58): You don't want to go back and wipe on veil over and over, and over, and over again.

Scree (01:13:03): We did. We did. I just don't want to do it again. I've already done it twice now. So, a third time, not in the cards.

Rob Collie (01:13:12): This was a gaming heavy conversation. It's not for everyone. But we got a lot of good, good stuff in about data along the way. Of business data. If you're just here for the business data, we covered that ground. Hopefully that gave us an excuse, air cover for us to talk in this gibberish that five people listening are going to understand. Like, veil. What is veil? You don't want to know. Endless frustration and in-fighting. All right, man. Well, I've really appreciated this. I've enjoyed it. Nice to connect with someone who all I have none of them is an avatar on Twitter. I love that kind of experience. Thanks for being open to doing this. You're putting yourself out there, and I appreciate that you did it.

Scree (01:13:55): Thanks. I had fun.

Announcer (01:13:57): Thanks for listening to the Raw Data by P3 Adaptive Podcast. Let the experts at P3 Adaptive help your business. Just go to p3adaptive.com. Have a data day!

84 эпизодов