Skip to content
PodSubs
ShowsPeopleGuestsJobsNewsEvents
More
BlogPodCred
Menu
ShowsPeopleGuestsJobsNewsEventsBlogPodCredNetworksSign in
Join nowSign in
  • Shows
  • People
  • Jobs
  • News
  • More
    • Guests
    • Events
    • Blog
    • PodCred
    • Networks
    Join nowSign in

Energy Bytes

Energy Bytes artwork

EP 21: Sawyer Nyquist from The Data Shop

Dec 19, 20231 hr 19 minAudio

Sawyer Nyquist, founder of The Data Shop discuss his journey from theology to the tech world of data analytics. Sawyer shares insights on how non-technical backgrounds can provide a fresh perspective in tech, emphasizing the importance of blending business understanding with data for effective problem-solving. The conversation explores the evolving role of data analytics and business strategies in driving value and innovation across industries. Join the conversation shaping the future of energy. Collide is the community where oil & gas professionals connect, share insights, and solve real-world problems together. No noise. No fluff. Just the discussions that move our industry forward. Apply today at collide.io Click here to view the episode transcript. 0:00 - Podcast Introduction 1:24 - Tech Career Beginnings 5:17 - Business Problem Solving 11:59 - Balancing Business and Technology 17:30 - User-Centric Solutions 22:13 - Microsoft Azure Data Evolution 26:18 - Azure Synapse Analytics Overview 29:50 - Azure Synapse for Business Users 34:40 - Azure Synapse Co-Pilot Features 43:17 - Founding The Data Shop 43:36 - The Data Shop Origin Story 44:10 - The Data Shop Services 45:15 - Integrating Business and Tech Solutions 46:26 - Business Incentive Strategies 50:10 - Project-Based vs Time and Material 52:00 - Managing Scope Creep 55:40 - Value Identification Techniques 57:55 - The Power of Declining Projects 59:05 - Project Value Assessment 1:01:15 - Understanding Metric Trees 1:03:07 - Making Customers Successful 1:04:06 - Assessing Work Value 1:05:55 - Challenges in Business Outcomes 1:07:48 - Free Bootcamp Promotion 1:09:35 - Quick-Fire Questions 1:10:37 - Open Source Libraries 1:11:24 - Influential Tech Personalities 1:12:17 - Tech Book Recommendations 1:14:13 - Vacation Destinations 1:15:35 - Tech Industry Advice 1:17:30 - Connecting with Sawyer Nyquist 1:20:07 - AI Implementation at DWS 1:24:12 - Episode Conclusion https://twitter.com/collide_ai https://www.tiktok.com/@collide.io https://www.instagram.com/collide.io/ https://www.facebook.com/collide.io/ https://www.linkedin.com/company/collideai

Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.

Transcript

From the show.

Welcome back everybody to a uh another episode of of Energy Bites. Rad John Cowan here with my wonderful co-host Bobby Neil. What's up? What's up? Today uh I'm I'm having a great week. We got the the SSO push out to collide this week and uh finally it's taken two months to do that. It's been a process. Yep. Um and then Arkansas beat Duke. So, a huge win for the Arkansas fan base after a terrible uh football season. But today we've got uh we've got Sawyer Nyquist uh joining us as the the uh founder of the data shop. Hey, good morning gentlemen. How's it going, Sawyer? I'm doing well. Uh I know you guys are in the south, but I'm I'm in Michigan and it is snowing right now. Oh, man. It's I mean December 1st snow is beautiful. Um, so I'm I'm enjoying it right now. Warm coffee. That's awesome. December 1st here and it's a balmy 70 degrees. That's Yeah, it is here the last two days. Uh, but that's that's Houston. If you don't like the weather, just wait another day. Yep. Well, we have snow on December 1st and like April 1st sometimes. Very different. Goes goes all over the place. Yeah. I'll be I'll probably be with the kids in the pool in April. Yeah. Right. But cool, man. Well, I appreciate you uh jumping on and joining us. Um we ended up getting connected through, you know, LinkedIn and stuff because you were at the time you were working on uh some of the Microsoft Fabric stuff and and some of that. So, we'll get into that in a little bit, but um you know, the the main thing we typically talk about out of the gate is kind of how did you uh you know, your journey kind of to this point to uh to being the the owner of your own shop. Um, but how did you kind of get into tech? Um, you know, early on growing up, you know, were you taking apart computers? Were you into tech at all? Or did that kind of just happen and evolve over over time and stuff? Kind of walk us through kind of where you uh how you got here a little bit. Yeah, I do not have much of a tech background at all for many years. My education was in um religion and theology. Like I had a liberal arts degrees um from at a bachelor's, and then I had a master's degree from a seminary. And so, like, I used computers; I did had zero technical understanding of of the the guts of them or any technical infrastructure really. A desire for that at all wasn't wasn't uh something that showed up at all in my in my life after finished school. Uh, landed in like a business operations job um doing logistics for a travel company; figured out I kind of liked business stuff. Um, like business strategy, liked operations, like thinking about products and pricing and marketing, uh, and all those different things that showed up in a small business I was a part of. Spent about 5 years in that space and then got a chance to do like a technical support role. So I was like still business side working with business users but for a software company. And all of a sudden my manager said, "Hey, you would be better at your job if you knew SQL. uh you can do troubleshooting, you can do like uh deep dives and kind of like exploratory about what's happening with these issues the customers are running into. Uh and so he he taught me select star from and uh very indebted to to him for that and it that that turned on a switch in my mind. Um all of a sudden it clicked things made sense. I I could think in data. I could think in terms of like structures of how um code would work from at least like a SQL standpoint initially. Uh, and once I figured out that I had a knack for that space, kept diving deeper, and then I realized that there is a a beautiful little niche in the middle between data and the business world. And I figured out I could start to speak both languages, and that often times the technical side of the house and the business side of the house don't talk well to each other. No, not at all. And so, kind of like my liberal arts background, uh, my soft skills, um, allowed me to play that role really well. And that kind of, you know, ended up as like business intelligence and data analytics generally that the type of role that sits between business and data and business and technology and then had a ton of fun once I discovered that that was a domain, that was an industry. That was seven or eight years ago and got into technical consulting. My first, my first real like techheavy job, um, was as kind of a data consultant working with much more senior people than me like beautiful mentors, kind people who put up with stupid questions and helped me really figure out, uh, how to, uh, how to do data warehousing, how to think about data engineering, how to think about data analytics. Uh and really like accelerated a learning curve and a growth curve um for me that um that hasn't stopped. It's been been a really fun journey and I I figured out quickly uh as I got into that that oh this is this isn't just a job like I'm buying books on Amazon about it. I'm watching YouTube videos in the evenings and uh that all of a sudden I was like oh this is something different um and this might be a career path actually something I'll stick with for a while and it has. No that's uh it's funny how many people's uh gateway drug into tech is is SQL or data cases out of the gate. I mean, I'm, I'm in kind of a similar uh, well, um my background is a little bit similar in the the sense that I'm a, I'm a mechanical engineer that ended up getting my my MBA. And, uh, but I, I as well felt that I understood and could communicate both the business side of things and then the technology side of things. And you are absolutely 100% spot on, regardless of what industry you focus on. If you're in tech, there is always a pretty decent disconnect between the developer technical data side of the house and the business users and things like that. And so, you know, like that's one of the the pitches for a lot of the low code, no code stuff that's out there, right? Is give the give the the business user the power to to rapidly prototype, develop, and and push while not having to deal with as much of the uh the technical side as of things. But it's uh no there's that's I mean that's probably one of the most common themes across all of our guests so far is there is not a traditional path to to most people's uh tech tech journeys and stuff. So that's that's really cool and honestly it's probably probably easier to be a high performer when you come with a different background and like you're talking about some of those soft skills again. I think you and I honestly have a pretty similar background in the sense of you know really no technical training and then really similar timeline too. I got into the oil and gas industry but that was also my foray into data uh back in 2014 and I I came out of teach I was a teacher and a coach I mean I had a math degree but I you know my whole thing is like I didn't know what a VLOOKUP was in Excel when I when I started. Um, but I think being able to bring some of those communication skills and learning how to communicate with different people, different types of people and, you know, kind of even the teaching, knowing that people learn differently or see things, you know, different, some are visual, some are auditory, you know, and everything. And being able to kind of tie those themes together and and like you're saying, be that glue or be that bridge between the technical side and the uh Yeah. and the business side is super important. It can't be understated. Yeah. And that's the thing is. It's like the, you know, a lot of the times the developers, you know, they're doing their sprints, they're up against deadlines and timelines and all of that stuff. And so, a lot of the times, if they're not tied into the business unit as well or understand the business or the business problem, they're just, you know, going through and checking the boxes, right? Like, okay, it needs to do this, okay, so I'm going to go ahead and check that box. But that doesn't mean that it's going to be useful or valuable to the end user, because if you don't understand how they, what their workflows are, how they use the tools that they're using, and all those things, then it, yeah. yes you check the box but is that feature actually how the end user wants it or whoever put you know put the ticket in for it yeah or even just giving that a little bit extra I mean like often times I'm able like all right you want this but I know eventually you're going to want this or want this detail associated with it so being able to yeah kind of know what they want and giving them even a little bit extra and like oh I didn't know I could do that but that's even even better um but you have to understand the domain to be able to do that for them as well but let's kind of dive in on some of that though because that's been a lot of some your posts recently uh Sawyer has it been kind of like if you want to know what the business wants, ask them and listen you listen to them. So I mean can you kind of dive in on some of that or even some examples like that have kind of spurred that on? Um, because I think it is important, and I think you've gotten some push back, so we can kind of talk about some of that where it's like,"Well, what if they ask you for something absolutely ridiculous?"And, like you know, kind of the difference between like, the tech, the technical solution but then actually listening to what they really need or want. right. Well, yeah. And what they say they want typically isn't exactly like how you would do it on the back end, too, right? Like it's true. Or, you know, they think they want something, but in reality, the you have to really pry it out of them and dig in to to figure out what the actual problem that you're trying to solve is. Yeah. I I'm a proponent of well, I mean, in the data space, specifically data professionals, we we always harp on like data literacy, and I hear regular complaints about, hey, uh, all my business stakeholders don't understand data. They don't have any data literacy. Uh and the complaint the opposite direction is yeah most of the data people don't have any business literacy um and can't think about solving business problems. And it's it's often I I advocate like it's more important for the data people to understand the business than for the business people to understand data. And when data data professionals data teams can understand and actually want to know what the business outcomes are then we can get some shared alignment. Um, what happens is, like, a, you know, business team is, uh, trying to get their goal, and then they turn to the data team, and they start telling the data team what to do, and the data team starts trying to figure out, like, how can we do this with data? And, like we, we're talking about, like, this middle point of, like what, what data points do you want? Okay, I can get you these data points. I can get you at this um at this level. And then we end up with all these uh crosswires about um which was the right ask and did they actually get what they wanted and they asked for the wrong thing. And it's like what if we were focused on the business outcome? What if we were focused on the end goal and we had a clear articulation of what that was and a clear articulation of how we were getting there and our our markers along the way. Um, if the data team knew what that was, uh, there's going to be much less of the data team or the business team doesn't know what they want or they ask for the wrong thing. We know what the shared goal is. Data gets to data teams get to implement that in the way that they're the experts on the technology. The business is the expert on the business. Um, let's let's try to keep that clear. Yeah. So, as a little bit of of the stories I've heard is, uh, oh, the business asked for something and they didn't know what they wanted. It was the wrong thing. Well, it's because you asked them about their technical requirements and they're they're experts on the business. Ask them what their business goals are. You help them get to their business goals. They told me that they wanted a hundred, you know, spreadsheets that were syncing real time across it. Like, yeah, they were trying to describe technical requirements because that's what you asked them for. You're like, what do you need from spreadsheets? Like, ask them what their business goals are and then partner with them to achieve that goal. That that's been kind of the work I've been hoping to do at the data shop where, you know, I'm helping teams define, identify, and deliver value. Like, hey, we're trying to find things that are valuable. And value means making the business work well. Um, is the business working better? And that looks like goals and strategy as well as sometimes technical infrastructure and processes and delivery methods. Um, but it spans across lots of soft skills as well as like the hard skills. And uh the the mindset that I've struggled with a lot is when it becomes an us versus them with the data and the business and the when it's uh when the phrase is the business doesn't know what they want. Um it's kind of a a patronizing or paternalistic view of the business of like they don't know what they want. Um or what's underneath the curtain of that is the business is stupid. Um and they're not asking the right questions. It's like the business actually knows what they're doing a lot. um they're just not speaking your language or you're not interested in learning their language and yeah and they're actually generating the revenue that you know like is paying yourself you know like so some things are going right they just don't know how to ask you or you you haven't you know created the right pipelines for them to engage with you or or to you know speak your language so yeah and you bring up a good point that I know Bobby is like a pretty big proponent of and that is you know making sure that uh that these things are defined on both sides. And you know, the the business users don't care how the sausage is made. They just want the sausage, right? And most of the time the tech guys don't care what the sausage looks like in the end. They're hyperfocused on building, you know, the thing. And I completely agree with you that, you know, the the typical approach because that's what devs are used to is:"Hey, okay, we're going to go get all these technical requirements and blah blah blah, and then we're going to break those down into tasks and sprints and all this stuff."And the business users have no idea what any of that [ __ ] is. They have no clue what a sprint is or what agile development is or any of that stuff. And so it's you know it's uh I completely agree with you that a lot and it's tricky because I kind of fall on the more on the business side because that's most of my background but you know the the business users I don't believe the business users should be responsible for defining the like objectives and stuff. It should fall on the project manager or the technical teams to Unpack, like, understand to your point: what is the problem we're trying to solve here? Not how do you want to go about solving the problem and all of the steps in between that, because that's what the technical folks are there for. And you don't want a business user saying,"Yeah, I think we should use, you know, SQL for this unstructured data set." No, that's a dumb idea, you know? Like... and well and what what partnership looks like between data teams and business teams is uh we both win together or we both fail. Like there's there shouldn't be a scenario where like the data team's like,"Well, we did everything right. We're good."And the business like, "Yeah, but we didn't hit our goals or we couldn't achieve what we wanted to do." Like that's not partnership. Um that's that's everybody has different incentives, different goals and competing alignment. Partnership means I don't win if you don't win. And so I'm with you, data team to the business. I'm with you until we hit those goals and I'll move with you. If we got something wrong the first time, that's all right. like we're still moving along toward that objective and we win when we're or yeah we don't win until we both win. Yeah. I mean like again if you were trying to work on something that was going to reduce cost and then your profits weren't any better at the end of the year. It's like, well, we built the tool that helped them, you know, identify where they could have saved. But, like to your point, you didn't have the partnership with them where they had buy in that they were actually using it potentially, or it didn't deliver what they wanted. But then there weren't again, I've seen things where there's just not the communication of,"Oh, I built this and you said it was good," but then two months later you there was no follow-up and it's like,"Oh, we're not using that."And like, well, why not? Well, this I mean, honestly, like, you know, and it's been good. We started at the kind of at the idea of one of our previous guests. I started doing some like office hours, you know, every couple weeks where like people can just come in with different questions like,"Hey, how do I do this?"or"Hey, this isn't working. Can you help me?"And we had one yesterday where they were using a dashboard and they're like,"Well, this thing hasn't worked for a while."And I went in and like someone had saved it with a filter and it was the filter, but it wasn't obvious that the filter was on and it was filtered to May. So it probably stopped working around May, which sounds about right because about five, six months ago. And I was like, if if we open this filter out to today, it'll work. And sure enough, they're like,"Oh, this is great."You know, but even just little things like that where it's like when there's silence on both sides and you don't have those lines of communication, like, you know, you don't you lose that buy in or Yeah. No, I think you're spot on, right? Like, because a lot of the times it's like, okay, you go, you develop whatever you're trying to develop, and then it's on to the next thing; and then there isn't a lot of that followup, like, hey, a month, two months, 3 months, 6 months in, like that just checkup. Because typically what happens I mean, we see this all the time, whether it's in the field or in the office, right? like, if it isn't working the way the end user wants it to work or expects it to work, they just quit using it. Yeah. And or they're going to work around you and Yeah. or they'll work. That's another very common problem in our industry is they just find ways to work around it. Shadow it. Yeah. Well, I mean again like you work with enough we work with a lot of engineer. John is an engineer, but like you work with enough, you know, like we're talking like petroleum engineers or whatever, but like they're problem solvers and if you don't get them, you know, give them a good way to get it, they're going to find a way and they're going to work around it. Yeah. Um and you also see it even and maybe we can talk about this in how you've seen it work at bigger and smaller organizations. Sawyer is like, if when it puts too many roadblocks or, I mean, sometimes they have necessary layers of, you know, security or protection but like, for each of those kind of roadblocks, people are going to have to find a way around it. Like,"Oh well, I can't do this on the database; now I got to learn Power Query,"or now, like, and then that business logic gets pushed further, kind of downstream, and you know, then you have monstrosities of Excel documents because that's the only place that they can do anything, you know? Um, so you, maybe if you want to talk about that, like if you've been at bigger, smaller shops, and just what you've seen that works, or um, or doesn't work, or... I'm also curious, while you're talking about that, because this is another theme we've seen all across pretty much every guest. But, you know, from the dev side, the most successful implementations of pretty much anything I've ever seen in adoption is typically meeting the business user or the end user where they are most comfortable, right? And so, like, I used to hate the fact that our industry does, so you know billion ion, dollar publicly traded companies and you still have [ __ ] running on spreadsheets, right? Like multi-million dollar decisions being run on spreadsheets and I was like,"Oh great, spreadsheets where, you know, data goes to die, right?"But the but I've kind of flipped on that and it's, you know, if you have data and the end user is an accountant who is used to using Excel and knows all the shortcuts in Excel, you should meet the end user in Excel because that's where they want to that's what they know and where they're most comfortable. So find a way to make that work and you can do that now where it's you know a lock down spreadsheet or power you give them reproducible workflow still but you know yeah right they just need the data to be updated and refreshed into the spreadsheet and then they can do whatever magic they want on the back end with it but it's never going to die so like your data architecture your data strategy like it needs to take into account Excel and there's y all sorts of idealistic visions from data professionals around like you know having clean and pure like business intelligence reporting and analytics that's governed and managed uh and don't export my report to Excel. Um but that's what people do and that's what they need to be able to do often times. And so how do we take into account meet the business user where they're at there? There is always going to be a push and pull. Thinking about that larger organization question of of having lockdown controls on like centralized and like require like all sorts of approval levels um at Fortune50 companies, all sorts of approval levels before you can get any data you wanted to or get any changes made or um access to uh pieces. And so you end up doing workarounds. Um there's like the lock down and centralized uh ideal and then the other side of like kind of like open chaos or open free flowing information where you can get what you want but who knows the quality of it. And it's always a tug and pull between those two pieces. Like the organizations that are super locked down and have all sorts of like uh layers of approval to get access struggle a lot to move quickly and to make anything happen. And that's there's just bureaucracy at large companies and that just exists when you've got 17 enterprise architects who have to approve a change. Um, but at small, nimble organizations, like, yes, people can get access to what they want, like, in the same day. But you have all sorts of other, like, quality and governance challenges that show up, and it's just a it's a push and pull and. it and your your company changes as it matures and it moves, and as it grows you, you end up with those different layers of structure. You move from being really innovative and nimble um to being much more efficiency and uh control driven. Yeah. Process driven. Yeah. You're either you're either a preservationist or you're um you know progressive and it's always a push and pull between those and organization tends to move along that spectrum from one side to the other. Neither's neither's bad. You just you just got to realize where you're at and and work within those constraints. No, I think that's spot on, right? Like, even, you know, I've worked at a number of companies and stuff, and like, when I started here, you know, we migrated over to HubSpot, for example, and uh, like, I initially struggled quite a bit with, like,"Okay, well,which who, you know, we only have 10 people in the company, in the office at least,"and I was like, "Well, which of these users needs admin privileges?"Right? And like, that terrifies me as the one who I've done CRM, you know, administration, basically, for the, well, almost a decade now. But it's one of those things is like you can't just leave it open because then your data just becomes complete garbage. But you also want to give it, you know, you don't want to bog down your small organization by having too many limitations in place, but you also don't want people going in there and just creating wreaking havoc not realizing what they're doing. It's also easier to loosen the reins than it is, you know, that's same with parenting, you know, with, you know, an organization like you can, if you have a Titan, it's a lot easier, but once it's the cat's out of the bag, it's really hard to pull it Pandora back into the box. Yeah. Once they've been using that new data field that they created on their own six months ago, and you didn't know about, and you're like,"Oh, well that, we already have a field that does that, and you've now screwed up all the filters and all the reports that are off of that."But

I'm very curious to kind of jump into some of the stuff you were you were working on uh previously at uh at Microsoft and and what you guys have been doing. I mean obviously I I think people sleep quite a bit on what Microsoft has really done over the last decade plus, right? But like Sure. I mean growing up you know Windows was always there, Microsoft was always there but you know Apple was always like the progressive innovative or viewed that way right and I feel like at least the last decade now once they they started integrating all of the the team stuff and you know just all the cloud based stuff um then the power apps and PowerBI and you know Microsoft is just so good at spotting new trends technologies softwares going and buying them and integrating them into their stat, you know, very similar to they become good at it. Well, yeah. But I'm I'm curious just to to kind of hear a little bit about what you were working on over there and and some of the stuff that uh that that you know you were you were playing with and seeing and stuff like that. The Yes. Yes. So, I spent about three years at Microsoft, and mostly in like a professional services, like helping customers use Microsoft products, um, doing development and build out for them. And so, got a chance to work closely with product teams and try to like help figure out like,"What is the directions product? What are we doing with this? How are we?"Because we're trying to implement this at a at an innovative customer who wants to be on the cutting edge of this, you know, Microsoft has a a reputation for you, from naming conventions to changing their names, like that kind of stuff, like valid reputation. They've had products that have flopped. Um, we can talk about what that looks like at times, too. Uh, they've also, as as you pointed out, have been probably quietly remarkably successful in the data world. Um, in certain types of organizations, there are still certain types of organizations that, you know, if you work in the startup world or the SAS world, Microsoft has not really like it's it's it's AWS stacks. it's modern data stack tools. Um but if you go enterprise or any sort of like legacy or established industries, then you end up with like heavy Microsoft stack both legacy as well as um Azure focused and and PowerBI has been a gateway drug for a lot of companies on Microsoft tools and and that's something that Microsoft has specifically uh capitalize on with with fabric and how they moved that direction. Um because well so Synapse was the the last big Microsoft data launch uh you know that was four years ago maybe sounds I would say because before that was Azure data warehouse technically right and then they kind of sort of rebranded a little bit and then yep rebranded pulled some things under the same banner and it was it was a lot of hype around and Synapse and I think there's valid criticism that Synapse probably didn't didn't didn't achieve what everybody hoped it would um And so people were pretty skeptical about anything new and what's next. Uh and so when fabric was was hitting all the bells earlier this year, it was hey this is just maybe valid skepticism coming out. Yeah. Um Yeah. And so I I you know Synapse did a good job of incorporating data factory and data factory has been a successful product and Synapse did a good job of introducing Spark and having Spark notebooks and Synapse having its own spark runtime which is again a really powerful incorporation. The Synapse data warehousing piece did not live up to the what what it needed to be and Snowflake and data bricks continued to gain steam and take over huge market share in that world. Um, so fabric has kind of been like the answer to that in some ways or their goal of like repositioning to hopefully catch up with like large um, and I'm speaking as an outsider. I'm not like inside Microsoft. So this is like I have I see the pros and cons and the and the flaws uh, as well as some of the the diamonds that are in there. Yeah. No, that's I mean that's what we want on this podcast is the real story, not the PR version. Um, for people who aren't as familiar with some of the Microsoft stacks products and stuff like that, can you kind of just briefly touch on what fabric is, what they're presenting it as that, you know, how that compares to certain, you know, what other things in the market that compares to or is an analog to that? Well, I mean, from an energy perspective, like I'll just I'll back up a second just thinking about like industries and where they're at. Uh, and do a history. something like you've got your kind of like your is solutions or the infrastructure solutions that were um SQL server on prem SSIS SSRS um SSAS those like the legacy products that are still probably that are very prevalent in the market still um a lot of people are still running SQL servers boxes server room yeah I mean that's oil and gas number one 101 yeah they all so like that's the foundations like hey like Microsoft started there um in terms of like their data stack and I still a very rigorous data stack. Uh there's the cloud versions of those. Then you've got like your paz solutions in in Azure. Um which is you know your Azure SQL DB data factory. Um and then PowerBI became like the the main successor SSAS basically. Yep. Yep. Under the hood like the same engine down there. Uh and so that you kind of move from like IAS solutions of like here you get your own boxes and um install these these servers on there or these software on there to p solutions hosted in the cloud. Um fabric is the is the SAS solution the next progression. So, from infrastructure to platform to like software as a service, and it takes away...the goal of Fabric was to essentially, like, it's going to package up those those data tools, um, and it's going to put them into a managed environment, uh, where the where the all the configuration pieces, all of the, um, a lot of like the governance and a lot of the knobs and switches that are under the covers that you're used to configuring if you had a SQL Server or even if you had an Azure DB, um, all that gets abstracted and is available in a managed offering within Fabric. So it's a SAS SAS service and it's the first of its kind in terms of like um a SAS spark runtime. Spark has never existed as a SAS platform. Um it's you know the it's it's a SAS data warehouse which is the most similar to what a snowflake would be like in terms of like ease of use like in in 30 seconds you've you've spun up a data warehouse deployment. Yeah. Um and so just like the the ease of use from moving from like legacy technology of SQL Server on prem to um what you have with fabric is is worlds different. And even back to like early in our conversation when you talk about the the the the gap between business and IT that gap existed for a long time just cuz there was heavy IT infrastructure that required specialized skills and then business users and it was harder for those to merge together. They live further away physically in the building. Um, but that's slowly been closing as technology has moved from IAS to PAS to even SAS solutions where I don't need IT infrastructure in order to deploy a service or deploy a data warehouse or a report. It's a little bit of like fabric is the is the SAS solution um to to package up all of Azure all of Microsoft's data legacy data products and all their path and Azure data products um into a single banner. Very cool. So with some of that, can you talk a little bit because I mean, I think, say on that business user side, because I think one of the interesting things too is even like you can almost connect to or mount your Fabric data like through Windows Explorer I think I've seen, right? You can actually see it there, technically. really um but then I think it it is cuz it's really similar in a way to data bricks where it's a a it's using what Delta Lake files right um And you can hit it with Python or R engines, ductb, and so on. But say more on the business user side, because people could and I guess they kind of sort of could before but now, like, they'll be able to use essentially kind of Power Query or data flows and stuff like that right on the data, like kind of downstream. So, again they, they won't even need necessarily a data engineer to write some data aggregations for them if they know how to use some power query I mean can you dive in on some of that yeah so I mean if that's right or wrong too It it is it is directly correct. So, like Fabric,they um, the Microsoft team like used PowerBI as, as I said, the gateway drug to all of a sudden. They're like,"Every, we have, I don't know, the millions of users I don't[the] exact count it is millions of users of PowerBI, organizations that already have PowerBI installed."What they did is said,"We've got all these people who are already have this PowerBI tenant, already have reports out here already, to use it; we're just going to put, get like open it up, we're going to add all sorts of additional tools inside that same."If you've used Fabric at all, or if you opened up PowerBI in the last couple of months, it's now the Fabric emblem. All those tools landed inside of PowerBI as opposed to having to like, you got to go buy something else or head somewhere else. So they basically took PowerBI a lot of significant updates to PowerBI itself. Um but then they added in a bunch of other tools that integrate very tightly into that PowerBI ecosystem that you're that's that SAS platform that you're already familiar with. And so yeah, like you know, one way to think about it is like they just added an abstraction layer on top of all that. So you don't spin up Azure storage accounts anymore. you have one link and you don't spin up a data warehouse anymore. You or you don't have to in Azure. Um you now click a button and create a warehouse. And so all of those abstractions are on top. It they say it's it's fully rearchitected from the bottom in terms of a technical standpoint. But from a user infra perspective instead of individual Azure resources that require an admin um you're creating all that within the PowerBI portal which is a very friendly place and you can do your data engineering tasks and your data manipulation in very PowerBI friendly tools. the PowerBI winners, the PowerBI users were the big winners with fabric because the tools that they were familiar with Power Query they know how to use and the PowerBI interface all that just got a huge upgrade and all sorts of additional tools put at the fingertips of a PowerBI user. So if I understand correctly I so I love the the ETL backend Power Query side of of PowerBI because I'm a very visual person and I'm also not detail oriented enough to be good at coding. Um, and so like you know I can understand how I need to transform the data but having that guey on top of it and then of course the ability to go in and actually do any kind of raw scripting is so nice from a user perspective. But if I understand correctly I'm now able to do the exact same process but I'm able to actually do that on my data lakeink from PowerBI or is it just another there's another tool within PowerBI that allows me to do some of those ETLs. Yeah. And and this is so like we're we're in process here, but it used to be you you do PowerBI desktop development. Everything was in desktop. And so like you did you had Power Query and you did ETL kind of in the desktop. Um they started introduce data flows which do some of that in the cloud. Um for the most part all of PowerBI development is going to now be available in the cloud. So when you're developing PowerBI reports and um doing ETL and Power Query work that's now natively developed inside fabric and it's it's it's connected to like data fab or data factory which is your integration tool. So like yes it is like you your PowerBI sits directly on top of your data lakeink and you can transform data directly from the data lake with an ease and integration that just like didn't exist very well before all all natively in the cloud like you're you're working within the browser to do all this work and connecting to data factory pipelines in terms of like creating overall orchestration flow of data transformations the transformation engine in data factory now looks like power query Got it. And scales like power like an enterprise data integration tool should uh but with the ease of use for someone like John who like wants to see gooies like wants to see how it shapes and like visually and I'm that way too a lot of times it's easy to look that way. Yeah. And let's talk also then because I mean I know fabric being released was also coincided really well for them with uh GBT and co-pilot. I mean, can you? Because, like, I mean, there's some of that promise with it too now. It was where you could actually define pipelines by prompting, correct? And then that it, you know, they're working to where that can build your pipeline just from a prompt. Um, or even then, I mean, obviously they've kind of had it in a way, but probably is even more amplified now is: being able to visualize data in your PowerBI reports and asking, literally asking it questions, not, yeah, not having to create charts or anything that. or even like it suggest make suggestions now too which is really nice like, hey, do you want to see it sliced this way or hey, this basic little regression shows that this is really important. Yeah. Yeah. Which some of that stuff has like been in PowerBI for a little bit like show me insights on my data. Uh and it was like hey sales go up by month like okay not not super helpful. Um, but I mean we we talk about people sleeping on Microsoft. Uh, the the Open AI integration and the pieces that are coming and the co-pilots that are showing up in Microsoft are really uh showing up moving very fast. Um, and so yeah, Fabric is there's not like one co-pilot. This is what I always I'm trying to get my head around. There's like there's a a dozen co-pilots. There's a co-pilot that lives here and a co they're all called co-pilots. They all operate a little bit differently. Um, so the yeah within fabric you now are having co-pilots that uh and this this changes week by week. So, I mean, even as listeners are hearing this, like, goodness, I can barely keep up with when co-pilots are released in different pieces. But the PowerBI co-pilot's like,"Design me a report around this data model,"and it'll give you and with, with a few directions like,"I'm looking for sales analysis," or"I'm looking for, um, you know, performance downtime,"or pieces like that. It'll it can build out visuals. There's the, you know, the GitHub co-pilot. It's been around for a while. Um, yeah, a while. It feels like a long time. User. Yeah. I mean, I've been using about a, I guess, close to a year now. I mean, I guess it all came out about a year or so ago, but it's it's really impressive. Yep. So, like that that that lives within like fabric notebooks um in terms of like Spark runtimes and you're like, hey, I'm coding up notebooks. I want to have like copilot assistant alongside that. It's showing up in Data Factory, like "describe the type of transformations you're looking for,""describe the data movements,"and it can start to again mock up, build out the initial skeleton and framework of what you're looking to do. Uh, so yeah, providing a lot of, um, a lot of levering to the work that users would normally do uh in Fabric because almost every piece of it now has a co-pilot assistant that's going to be available that's in private pre in previews now um and we'll slowly be rolling out to more general availability. I mean, they could have had Clippy, you know, in the last five, but it makes me so sad that they I wish that Clippy would have been Yeah. Yeah, that would have been beautiful. What a I mean, what a great nod, right? Like to the past, but also making it actually useful. Like, that was in my mind, that's what Clippy was trying to kind of, of course. Yeah. 20 [ __ ] And we talked about this recently. Like, I mean, none of these ideas are new. I mean, people have been trying to do computer vision or AI for more than 30, 40, 50 years. I mean, but now we finally have the technology and the capability to do it at scale and relatively cost effectively incredibly quickly. Yeah, I think that's one of the big like the the technology enabled the speed, right? Like having, and you know people, a lot of people don't have any idea what federated learning is. But, like, that's what all these AIs are. That's why they're able to move so quickly; is because with GPT, you've got a million plus users, you know, in the first whatever, week, month, using it, and they're all training it every single time they use it. And the more people that use it, the better it gets trained, and the better the mother model becomes, and it just keeps going through that. cycle and so it's uh like I'm super bullish on on especially with just how fast everything's moving. It's terrifying because it's like like what's it going to who knows what it's going to be like you know a year from now but well and then all the rumors with with Altman. Yeah. But uh that's a fun Thanksgiving weekend, right? Yeah. Yeah. That made made things uh pretty interesting. But, you know, the the it's not going away, and the speed at which it can move is just so incredible that I think a lot of people still sleep on that aspect when they're just thinking about it, or, you know, like casual tech folks don't really understand just like how powerful all of those users using it at the same time are to its development speed and stuff. it and it starts to change the types of skills and problems that you want to solve as in your in your domain in your skill in your career. So if you're you know if if you've specialized and developed your career and skill around solving a type of problem that GPT can solve or LLM can solve, it's going to be tough sledding over the next 5 to 10 years as GPT solves the problem that you used to be able to solve. Um, or if you're a coder and you pride yourself on being fast and efficient and effective coding, like you'll get faster and more effective, but some of those skills are going to be commoditized by LLMs. Yeah. And so, you got to think strategically about like what are the types of problems that need to be solved uh that GPT isn't solving right now and won't be solving for any time in the near future. And those are a lot of like people problems, those are process, those are patterns. Um, yeah, those are like larger frameworks. It's a level of abstraction above, because a lot of the, like, the raw typing on keyboard work, um is, is going to be things that LLMs get to, um, provide a lot of advantages and leveraging too that. that's going to optimize out some of the work that you you were really good at, that you did have a unique place in the market for. And you talk about earlier on, just about you know coming from that more liberal arts background I mean, you came even from a theology background but you're starting to see even now, with some of this prompt engineering, like they're looking at journalism majors or English majors, and those are the people that are getting offered $200 $300,000 to be prompt engineers. I mean like but knowing how to ask the computer questions is going to be a big deal, you know, and then in upskilling people on that side of it, you know, it's not now is it going to be do people need to go learn SQL or should they learn how to prompt a co-pilot, you know, or an LLM? Like I'm very much so in that camp, right? Like uh, I backed up our our app's database to uh to SQL explicitly using GPT without any lines of code written, and I was just amazed at that the fact that it did it so well and quickly. And then I was like,"Well [ __ ] I guess I don't have to learn SQL anymore."Like, that's been on my short list for the last 5 to 10 years now, and it's like,"Well, I can get GPT to do it, why do I need to?" I I need you know like I think that's the other thing there, right? It's every math class we've ever taken, right? Like, hey, you need to understand the proof and the bounds so that you can leverage this. And that's I think that's a big proponent, too. But I completely agree with you. Like a lot of people don't they the casual user doesn't understand how valuable the prompt itself is, right? And like even internally I've every time I prompt it it's act as and your goal is to and like and just including those two things can completely change the output of what it gives you for the better a lot of the time and stuff. And so like do not my advice for most people is do not sleep on on the prompt. And there's so many tools out there now where it's like, hey, here's all these crowdsource prompts that people upvote and you can just copy and paste them, you know, and like or store them now within GPT directly. And so it's it's wild. Yeah. And we see how powerful like that like general generic open GPT is like, hey, no specific use case, you can tell it do whatever. We think about like having co-pilots that are embedded and designed around a specific use case like designing a PowerBI report or helping me write SQL. Um, some of that prompt engineering becomes even less essential because, like, it knows that you're writing SQL queries right now and it knows it's pro because it's within the SQL guey within Fabric or it's the PowerBI and it knows that it's working on top of your data model. And so, even like those hyper tuned co-pilots become really powerful that. yeah I mean all the stories about you know GPT making mistakes or not understanding what you're trying to do but when you start putting co-pilots that are designed around specific tasks and use cases uh it it yeah becomes more powerful as well. So, let's let's pivot a little bit, and maybe we can talk about how, maybe how some of this weaves in. But let's talk about your journey into the independent consulting and the data shop, and maybe first talk about just, you know, why you kind of move that way and what your what the mission is, and then, you know, maybe how some of this technology is enabling that as well. Sure. So I had an entrepreneurial itch for a while and I've worked at big companies. I mean Microsoft is huge. Worked at large within large consulting organizations or uh worked alongside like large enterprises doing data platform builds. And I, I just didn't work well as a small per a small cog in a very large wheel with huge systems. And I uh felt the itch to go do this myself and saw a lot of things that I liked about consulting, saw a lot of things that I didn't like about consulting, um, and wanted to think about designing consulting practices a little differently, um. So, Data Shops is solo, solo shop. It's going to stay that way. It's me. Um, this is not scaling up to a large team of people. Um, but my my goal, my passion really is to help teams, data teams specifically, define, identify, and deliver value for their businesses and communities. And so, I've got a specific interest in working with nonprofit organizations, um, but but I've worked alongside government organizations, retail, plenty of for-profits as well. And so I, I think about, uh,

defining value not by inputs and like, how long are we spending on this job? And even that's how consultants bill. Um, and that framework of how, like, most companies are are buying services and deciding outcomes based on how much, or deciding, uh, value based on how much input went into it. How much money did we spend on that? How much time did we spend on that? Um or even based on outputs like how much stuff did we make? but really like defining value based on outcomes and that's this goes back to the beginning of our conversation around like um business outcomes being the driving force for the data team. Uh so yeah that's that's what I do. I love the technical stuff like I'm happy to dive into like the weeds of data warehousing, data engineering stuff. Um, I love the people in the processes side, and I like to partner with data teams, uh, to think about those holistic pieces of partnering with the business to, uh, define business outcomes, um, and helping them get those pieces correct, um, helping them think about, uh, think well about business literacy and, and business problems, um, and then dive into the weeds around like,"How do we do that well from a data, process, procedure, infrastructure side of things?"Uh so yeah, that's that's been kind of the mission, the vision of the hit shop. Um we're not talking about tracking time and hours and inputs of how much things cost us or how many times we put into it. Uh we're really focused on like what are the results that we we deliver at the end. Um I want my customers to deliver business results. Um and so I partner with them and similarly like I want to deliver business results for my data team directly that I'm working with. And so hopefully there's there's a lot of alignment shared incentives which I think has been lacking in a lot of how IT consulting services are done. a lot of misaligned incentives around uh the the consulting firm and the the end customer there. Yeah. Because I mean I think one thing I've seen a lot with you know it or data consulting it's like how can we embed ourselves and stay longer you know I mean you come in and like it's about oh oh now that oh this but oh they've actually got a problem with this too. So how can we make ourselves sticky or you know again really like I mean like they they brought you in for a reason and like you know identify what that reason is I guess is what you're saying and then deliver on it and yeah you know and again by doing it the right way and that should you know build on itself and snowball. Yep. When when you get paid based on how many hours you spend at a company then you are looking for ways to be sticky and to spend more time there. Yeah, when you're paid based on delivering an outcome or a business value, uh you you know when you're done, like you know when you delivered your outcome, maybe there's another outcome you work to later. Um but it's less of less of a misaligned incentive around like I need to spend more time here. What other things can I what other work can I crop up? Yeah. And there's also the thing too where it's like the just because they spend a lot of time doesn't mean it's better. I mean like I mean usually the the value is can you can you deliver a good solution quickly? I mean, like, and that's why people should pay a lot more for that even because, I mean that time to value, or that, you know, even like NPV, you know? I mean, there's a lot more value in getting to that solution you wanted quicker just because this person was really good and it only took him 10 hours to do it instead of 40, like, that's a good thing. Yeah. It's more valuable if it comes quicker right like ultimately like hey I can have a solution in six months or I can have it in one month right I'm paying more for the one month. Yeah. Yeah. Yeah. Or even with...well yeah, that's like in uh, in project management, right? You have three knobs you can turn, and it's speed, it's quality, and it's uh cost, right? And so, if if you're focused on keeping your cost down, your speed and quality probably aren't going to be very good, you know? You can't, you can't turn all three knobs; you can only have two maxs, right? And so, I think that's uh that is a lot of problems with. With development, right? Like, I've started using a new developer for our app who is a standalone, you know, solo developer. But and, I was using a shop before, which wasn't a big shop, but I had a project manager, and then they've got a dev team behind them that, you know, focuses on back end, front end, etc. And, you know, the the old shop that I was using was technically cheaper on like an hourly basis, but then half of those, you know, like don't even get me started but there were, uh, time reports that I got where it's like half the time that I'm paying you for is going to the project management. I'm like, I'm doing the project management. I don't know why we're doubling this here. Like, I'm not coding. I'm and I'm technical enough to be able to tell you the requirements and everything that I want. I just want it done. I don't need it to go through three different people and then have to be pay for your time to copy and paste it into Jira. Like, it's just ridiculous. So, now doing this with the solo developer, I can just tell him put it into his ticket. He goes and he does it and he calls me or shows me a loom and says, "Hey, is this how you wanted it to work?"And it's so much faster and it's actually cheaper even though he's more expensive on an hourly rate because it's done faster, right? Total cost of ownership. Yeah. Well, and like the other thing that you point out too which I also completely agree with is that you know it's the incentives like people are so people everything that humans do it boils down to incentives in my opinion, right? How are people incentivized? Whether it's within a company or whether it's, you know, transactional like what we're talking about with developers. And yeah, if you're incentivized on an hourly basis, guess what? You're going to want to try and spend as many hours doing that [ __ ] as you possibly can, whether it was good quality or not, right? Or whether you could have done it faster or not. And that's of course completely counter typically to your customer who wants to save time but also wants to speed it up, right? Like and so I, I really like, respect, and appreciate how you've kind of structured that, because I think that needs to be much more uh, that needs to be a lot more common. And I also think the flip side of that, from the company's, you know, the client's perspective, there need, like, clients need to realize that, you know, hey, this should be, like, a successful project with a good team should be based, like, projectbased, not time and materials based. and because your incentives are aligned I mean, all the way back to your original point, right? Like, if and how many projects also, like, could have had so much potential? But they ran out of budget, they were delayed; like, there were all these cost factors up front that prohibited it from ever getting across the finish line. And then they've abandoned that when there could have been a lot of value extracted from it, but they just never got it across the line, right? We run into problem all the time in our industry with people, you know, the last 5 years, especially the last year or so, like, hey, we we're going to use ML. We need to use ML, blah, blah, blah. Every [ __ ] press release and earnings call and all of the energy companies are talking about ML. And then it's like, "Yeah, we've got all this data, but it's 20, 30 years of data that is across spreadsheets and PDFs and SQL servers, and none of it's structured, none of it's, you know, like you can't you can't just jump to, 'Oh, we're going to automate everything with AI and ML,' if you don't have good data structures and stuff put in place."And so it's like you're going to have to spend the money to go do that. Whether you do it internally or externally, you have to have these foundational pieces to achieve those long-term big picture things that you want to do. And it's going to cost time and money to like it. It requires resources to get there. Whether you want to uh whether you want it to be that way or not, like it's it's just a thing. And I mean the going back to the original thing we started with like business outcomes. The reason companies come in and just start billing as a consultant and just start working on stuff sometimes without end um is because they don't actually know what they're trying to get to. They're just doing lots of work and we're headed towards some ethereal goal of having a data warehouse or having some sort of analytical solution. We don't actually know what the end goal is or we don't have markers of how to get how do we know to make progress towards it. um from a business outcome standpoint. So that's why scope creep is a problem and everybody's terrified of scope creep um because we're like, well, we're we might we might end up doing extra work on this side over here and that'll be more hours over here. We didn't the business was probably lazy up front when they hired the consultant in terms of like they didn't know what their business outcome was. They maybe had a bunch of technical objectives. The consulting company didn't figure out all that either. And so they come in the door and be like,"Well, we don't know how long it's going to take, so we're going to bill ours and we're going to roll with that."Um, like that's it all really comes full circle to like the point of like we're goal is to deliver business outcomes and value. Um, do we actually even know what that means and how do we figure that out first? First and most importantly, like what are we actually trying to do here? Because I know like you've run into that all the time. Like everyone does that, hey, I want I want this. It's like, okay, but what does that really mean? Like what impact is that just going to make your life a little bit easier or is that have a direct business impact that we can then scale out across the entire organization instead of just you know it makes your it saves you 10 minutes a day for you know whatever. Yeah. It requires you actually care like so for a lot of people in a lot of people in their work like hey like I haven't thought about this enough to like define an outcome. Um it's the people who are passionate and actually care about what's happening. They're the ones who want to drill into this outcome stuff and figure out like what's the end result we're getting to. People who are checking in, checking out, punching time cards. Um that's that's where a lot of this comes out of is like you have to put in some work to figure out what are the outcomes we're driving to instead of just like I'm doing my day job. Yeah. Well, and it ends up being a much better and longer term relationship too, right? like this new developer I'm using. I wrote him a couple reviews yesterday because of how good like we were we're implementing alt zero for our user management on our low code platform.

Anyway,

you all right? You can make it.

Anyway, like there's all these people have implemented good lord people have implemented all your own stuff on this platform before, but never with existing users. And so like I I talked to so many developers, so many dev shops. And I could not find a person that could confidently be like,"Yeah, this is exactly how we're going to do it."And the guy that I found, sure enough, he actually wrote one of the all zero plugins for the platform out of the gate, which is how I found him. But then he as we were developing updated his plugin based off our workflow specifically so that it would work for what we were doing. And that's a true partnership mutually beneficial relationship instead of a hey you know I just needed to check these boxes and hopefully it works kind of thing, right? Like people come back and they keep using you and they want to use you and they'll tell your friends about you and write [ __ ] on the internet about you and like it's a it's a much better way of doing things and just trust in general. you just when when both again I guess it's come comes back to the aligned incentives and just you trust the other person that you know they're giving you what you need and that you're delivering what what they need. Um it's a beautiful thing but I mean I don't want to get because like billing and pricing can be pretty a very it could be at its own topic. Um, but like, but I say with this, I mean, like, how do you do you have kind of a framework that you use, like, for, you know, working with clients to can identify what that value is and then, again, making sure that there's value in the project for you too? Because, I mean in, at the end of the day you are still a business, and you know you, if you got to make money and provide for, you know, yourself, family, you know, whatever else,like so, I mean, how do you kind of protect yourself, you know, with, like, identifying that value and making sure that, you know, the value that they're going to receive is also you, generates value for you as a as a business owner. Uh

yeah, I there's not a clear uh formula that you follow. So it depends on the nature of I'm not going to like start work on a problem or on on a if I don't know that I can deliver the solution. Yeah. Um at the end. And that that's a big part of it right there. like if I don't know what the end result's going to look like and I'm not confident I can get you there, we're we're we're pulling out. I'm very happy to say no to work in that sense. Um if I'm not the right person, I'm going to send you somebody else who who is. When I'm confident about the solution that we can deliver, um and like I know that like, hey, I know that outcome, I know the measurable steps we can get there, and I'm I'm confident that I can help you get those directions. Um then it's a matter of figuring out what does that do for your business and what is that like what kind of value are we are we developing there? Um, and then and then what makes sense from a um from a cost perspective for you to invest to get that result. And so there's a lot of art and science in that. Um, and it's it's it helps everybody get on the same page a little bit cleaner because I'm I'm here until that result is done. Like that's that's the way I talk with customers. I'm here till the results done. So, uh, I'm in line to making sure that we define that really well and that I know that I can get you there. Um, because I'm not getting paid more if it takes longer. Yeah. Um, I'm getting paid more the same amount to deliver the result. So, it's a it's it's a it's a hopefully partnerships are much more art than science and that's kind of what it ends up being like at the beginning when you talk about like cost of a project uh and then throughout and even like the termination like when things wrap up. Yeah. Yeah. No, I think I think that part where you talked about being able to able and or willing to say no to work is is a huge thing because I mean I think there's some people that and I think it's tough when social people get started, you know, they're like I need any money, right? Like they they want to get going to eat. Um but yeah, but identifying say what your niche is and then also what your skills are, what you know you can deliver, I think is is a huge deal. And then being willing to say like look, I don't know if I'm the person for this is huge. But I think that also builds trust too. And like those people come back when, you know, they all right, we understand better what you do and like maybe you're not the right person for this, but you know, you've built a level of trust because you shot us straight. You didn't just try to milk us, you know, for two months out of out of some money. No, I mean that that that's huge in our industry, right? Like if you burn somebody, our industry is international, but it is still very small and people talk all the time and if you screw somebody over, it's going to make its way around. Or if you are unsuccessful and promise one thing and you couldn't deliver on it, it's going to make its way around. And so like people absolutely respect you being able to tell them no, I can't do that. You should go talk to this person or whatever. Um just I mean you're right. It builds so much trust. But like one thing you also mentioned though too that a lot of people don't focus on is it's like hey I want to do this. How much is it going to cost? But they don't ever talk, at least directly, with the developers or development team at all about what what the value of it is, right? Like, quantifiably, what is the value? And that's where a lot of the friction in software development within businesses comes is: business user has this problem, and they present it to whoever for funds or budget for that solving that problem, but then they have to now sell their boss on why it's worth that input, and they don't know what the value like what the tangible value of that is on the back end. Sometimes you can't, but you should at least try and estimate, you know, to with reasonable inputs and and you know, constraints and constants to figure out like, hey, it's going to cost 50,000 and it's going to save our team of 100 people 30 minutes a day, which equates to, you know, $50,000 a week or whatever the hell it, you know, may be. And so therefore, it's a very good obvious benefit, right? People just come in and they say,"I've got this problem. Can you solve it?"And they don't get that layer deep of,"Okay, what truly is the problem? And what is the value that this problem solving this problem will actually provide to figure out whether we actually need to approp figure solve this problem or not?"Really? Yeah. Yeah. Because the higher up the the leadership chain you go, the more and more they're focused on business objectives. They see like end goals of like what we need. Like if you talk to the CEO, they're not concerned about activities happening within the company. They're concerned about like key objectives that is set by the board or that's related to like the share price. Like there's on those objectives. And so if you try to sell tasks as you go up the chain like well we're doing this work done here. Um it communicates less and less as you step up the chain of leadership because they're less and less concerned about busy work or activities that are happening. They're much more concerned about what are the results that it's doing to like hit that men. Is it making me more revenue or reducing my cost? Yeah. Or IP standing out different from my competitors? That's the Well, I think it's and I think we were talking about something earlier and um and again it always comes back as a partnership, but you know the business needs to identify KPIs, but sometimes like they need coaching. Yeah. 100%. you know on How what those are, or you know, there needs to be a working session with the data people. And so, I mean, because... and there's a kind of a framework that I like that I saw. Um, I think it was about within the last year or so. I think the guy Abby came out there, but it's like the idea like metric trees. I don't know if you've seen that at all, but like... I haven't um but it's kind of like all right let's start from the top from a company and like let's say that our north star is you know let's make it really simple profit yeah you know so profit all right what's the formula that profit it's you revenue minus expenses. All right, so now that's a tree here and now all right let's now what say for us what generates revenue well it's barrels of oil times oil price you know and then there's different things you know that feed into those but you can break those down and then at different layers like now the data team can actually deliver on like how do we track those things that are actually working their way back up but we have to identify what that north star is and like that's what you're saying like I don't know maybe your CEO is more cash flow you know right focus or there's different, you know, metrics that matter. No, you know, but then really breaking down and making sure that whenever we're working on something, it's part of one of these parts of the tree that actually adds value back up to the top because I mean, I think, you know, we see it a lot even like on our data team that you get people and and they're in the weeds and they you want to help them do their job, but is it really value adding to the business like you know that they were able to do this thing 10 minutes faster again like personally? Yeah. And and again like and then there's building you know where you can help. That's good. Like, you know, making sure that what you're doing is adding value whether it's as an internal data team or, again, even as an external consultant making sure that it's, it's driving the value, again, and especially for the people that matter the most. And I think maybe another thing we can talk about is, you know, making regular talks. I think you've talked about it, good, but I think Ethan Aaron talks about it, you know, quite a bit too, is like: go talk to the executives you or VPs, like, and like understand what's keeping them up at night or what, what they see as objectives. Because, like, then you'll know, you know, when you're working on certain things, that's actually adding value towards what they see as being valuable. No, I mean that's just a pro sales tip out of the gate, right? Like if you want to sell really well, ask your potential customer what their problems are and what their incentives are, right? Like how are they compensated? Figuring out if they have bonuses and how like if you can make your customer an all-star, you can get them a promotion, you can get them their bonuses that year, whatever, you're going to keep calling you. Like it it makes a lot more sense that way when you just boil it down to, hey, how do I make you look like a stud? How do I get you your your bonus or get you promoted or whatever? Like that because that's ultimately again going back to incentives. That's really people I win when you win. Yeah. No. um that that tree idea makes a ton of sense because the other thing too, right, is it's like it's not even, you know, okay, yeah, it might add value, but how much of that value? Like how much value is it? Because that's one of the hardest parts about development is priority. There's so much there's always a backlog and there's always a laundry list of [ __ ] that needs to get done and it needs to be done tomorrow. But which one of those things has the highest priority? Which one is the most impactful? which one has the easiest route and fastest route to success versus the other ones that might be more complicated. You might have to do some more architecture stuff on back end or whatever. And so like all of those things are very important. Yeah. And also what parts of that can you control? I mean like say that part I I got into revenue and say for us it's you know barrels times oil price. I can't really control oil price. Yeah, you can get pads man. Yeah. Well yeah. Yeah. I mean in a way until prices double while you're headed I mean no to your credit I mean I I did I kind of skipped past that. I mean, you know, I mean, the finance team or whatever could come up, you with a hedging strategy that, you know, ensures, you know, you can ensure a floor, but then you you also cap your profits on the top. That's the double-edged sword of the the hedging side of it. But um I mean John, you're right about like there's when we talk about this it sounds really simple like oh like just do things that are valuable and like like this is this is hard work because there's lots of competing ideas, lots of competing priorities, lots of objectives that you could go after. Um it requires a it really boils down to leadership. Um does leadership have a clear vision of where they're headed and can they articulate that and communicate that well? Otherwise like among the troops like the teams are going to be chasing whatever priority they think is shiny for that week uh and you will end up all across the board. So like it's sounds simple achieve business outcomes um it takes a lot of work and it takes a lot of good leadership to to do that well. Yeah. Well, and to your point too, like even if the company leadership understands it, if they can't effectively communicate that down or like people, you know, we haven't really talked about this a lot, but um you know, when Bobby and I were working together full-time, we did a ton of strategy stuff. We did I mean, OKRs and we we built out we completely revamped our mission and our vision for the like the company and all those things. And it's like people kind of, you know, see the very generic mission statements and vision statements from companies on the websites and stuff, but like if you have a good one, that is what you make your decisions off of, right? Like our our mission at digital wildcatterers is to change the way the world thinks about energy. So anytime I'm like, hey, we're going to we do we need to spend money on this? Do we need to put our resources towards this? Do we need to focus on something that's I can lean back into that to say, hey, does this check that box? If it doesn't, then I probably don't need to do it because that's not the mission or the objective of what the company is trying to do. Yeah. Right. And like when Bobby and I were working together, I was in product development, new product development, and it's one of those things where it's like, hey, I've got all these really good ideas. I'm getting all this feedback from our clients on things that they would like to see, but how the hell do I prioritize them? Like, which one brings the most value? Do these align with, you know, some of the things that because that's the problem is like, especially when there's... you're in an industry where there's a lot of opportunity, like oil and gas, not just from a money perspective, but the fact that they're 5 to 10 years behind tech-wise, just like every other major old school industry is. Um, that you don't you can chase a million rabbits, but you need to make sure that th those rabbits fit with the organi the organization itself and the mission of what the organization is trying to accomplish and what leadership is trying to accomplish and stuff like that. Because if it doesn't, it's not going to really matter and you're probably going to get a lot of flack for it on the back end. Yeah. And I think I mean and it kind of plays in to your theme of actually saying no, even like to projects. But there I, think and I'm going to, maybe I'll nail it, maybe I'll paraphrase it but I think with Steve Jobs, but I think like, focus isn't saying yes to one thing, it's saying no to like a thousand things like. um so I mean like but but but then having the processes in place or even just the permission like in a business to say no, I'm not going to do that. like and have that trust to be able to have that candid conversation with someone say no that's not actually important to the mission this is what we need to do or not right now like the urgency of it isn't yes it's a good idea let's put it let's document it let's capture it let's put it somewhere so that we will get to this at some point but right now it's not a priority and that happens all the time like it's a good idea but it's not right the right time for it yeah sure if this is I mean the type of people listening to this are probably the people who uh think about this a lot. I mean, they're listening to a industry podcast probably on their own time. Um, so if if like data teams being valuable and like these type of conversations interesting, I have like a a free 12-day boot camp email course on my website. If anybody's interested, they can data.co/value. Um, and they can sign up. It's like 12 emails that basically walks through this sort of paradigm of like how are we defining value? What do we not want to measure? What do we do want to measure? Um, what how do we know when we're getting it wrong? those type of pieces. So, if that's useful or this conversation is interesting about creating an invaluable data team or teams that deliver value, um I'll throw that out there as a resource for for the Definitely. And we can Yeah, we'll put link it when we send stuff out. Put it in the show notes and and on the YouTube video and stuff, man. We've already blown through that hour. That was super super quick. Um so, sorry. At the end of the the episodes, we kind of do like a little speed round. We try and mix multiple choice. I wrote a list down and I I have no idea where it is. It's somewhere on notion, ClickUp, Jira, notes, Slack, or my my messages. It's somewhere in there. But uh Notepad++. Yeah, we'll uh everything's on Notepad++. Yes. Yeah, we'll uh we'll kind of jump into that real quick and then um we'll we'll wrap up here. But um I think I probably know your answer to this, but AWS, Google, or uh Azure? Azure, you know, it's not because I've tried them all and I have a strong preference. It's because like I just spent a lot of time in Azure. Um so, you know, each industry kind of has like drifted towards one or like certain types of companies end up in one. They all got parallel tools. I don't know that I really would have a preference if I had rigorously tried them all, but Azure's my world and I'm I'm sticking there. Nice. Um, so I guess kind of piggybacking off that because I wanted to ask earlier, but we'll start the question just be what's your favorite open source library? Um, but then also kind of do you dabble in open source or have you like do you have you stayed mainly on the Microsoft stack? You know, uh, in the Microsoft stack, I would say related to that though, would be Apache Spark; is probably where I spent most of the time on open source project where I've used the open source project, usually using some managed version of it maybe that's in Synapse and data bricks, but um, Spark has become like the de facto big data engineering runtime. and I've had a lot of fun like uh with that tool in different contexts very flexible remarkably scalable um so Apache Spark would probably be the main open source tool that's open source Yeah. Yeah. 100%. In addition to Notepad++ probably. Yeah.

Um who are some some uh either social or technical follows that you uh that you would recommend in either the technical or the kind of what would you call that software development ops?

Uh let's see. There's probably LinkedIn's my world. I don't really spend much time anywhere else. Um, so data people on LinkedIn that I enjoy. You guys mentioned Ethan Aaron at portable. Um, really enjoy his framework and how he thinks. Um, he's a good follow. Um, there's another gentleman named Chad Sanderson at another startup data startup called um, Gable. Chad is a very very capable thinker, really uh, thoughtful guy around uh, data data world. So, I'd throw out those two names probably as two people I enjoy interacting with online. Um, let's do

favorite. You got a couple good uh books or podcasts that you would recommend to folks in this space? Oo

books. Um, there is a book that is um by Douglas Hubert called How to Measure Anything. Um, kind of a kind of a classic. It's only metal probably maybe 10 or 15 years ago. Um that's an excellent book when you think about yes we can measure revenue and like really metric things but there might be a lot of other things we need to measure on our team in our company that are intangible. Um and so he provides kind of like frameworks and thinking about how do you how do you measure things that aren't numerical? So yeah I throw that book out there as a as a as a good option. Yeah. There's a another another one I'll add would be a book called Crucial Conversations. Um, and it's kind of a I don't know counseling interpersonal like relationship book. Um, but like there's a lot of examples in it that apply really importantly to the business world. Uh, what do you do when talking when the stakes are high and there's maybe competing interests or um that come into play when you think about priorities and teams and careers. Um, it's a really tactical book about having hard conversations that show up like all the time in our daily world uh or in our daily lives in careers as well as like family and interpersonal. So, I found that book really uh holistically valuable to me. No, that sounds like I'm going to add that to my list. That sounds really good. Very valuable across I like I like things like the one you were talking about before where you're measuring I was going to throw out measure what matters. Bobby turned me on to that. That's another really good one. Okay. Um, and then but yeah, like you know there's always a situation where there's going to be competing interests whether it's personal or business or whatever, right? Like that's that's some good stuff there. You got one more? Yeah, I go and it's not even related to data or anything, but what's your favorite place to go on vacation? Oh,

the last two years I've gotten really into mountain biking and so I would say I love the mountains. I've loved the mountains forever. Um, and I haven't got to go to the mountains since I've started mountain biking. So, now I'm like really itching to like head out to Colorado or um, yeah, somewhere where I can like hit some real or Whistler. Um, something where I can hit like some real mountain biking trails the first time. So, I I love I love the mountains in the summer. Um, skiing in the winter would be cool, too, but right now like summer mountains would probably be it. So, Rockies generally probably would be the number one place I would look at. Yeah, 100%. One of our prior guests, David Thu, is also a big mountain biker. And uh I went to my undergrad is at the University of Arkansas and at some point in my life, I hope to ultimately retire in Northwest Arkansas, but he and that's like mountain biking haven, right? Northwest So that's why I was bringing it up is cuz I would he highly recommended uh I didn't mountain bike, but we would hike and camp and stuff in college all around there. That's why that part of the country is just so awesome. But it's also super slept on. like a lot of people really don't know about it. Um so I would highly recommend checking out Fyville in Northwest Arkansas. Um

last question, what's what's one piece of advice you would give to somebody um kind of coming into the the technology space at this time whether it's from school or transitioning from a previous you know career path or anything like that. Uh there's a lot of you probably have a lot of perceptions if you're transitioning into the technical world. If I have a lot of perceptions about what certain jobs are like, like I want to be a software dev or I want to be a data engineer or data analytics person. I have a lot of ideas about what that would look like. Uh, one of the most valuable things you can do is just have informational interviews with people and like what is that job actually like? Um, that was something that I had a a window of time for probably a year or so when I was having calls as often as I could with people across the tech world to figure out like what these jobs look like on the ground. Um there's a lot of either perceptions that you get from online about like the salary or the type of work or what it's like to be these different things. Um but I would talk to people who are in the in the industry in that role specifically. Um what does that look like and what does it look like on a daily basis and then and then try to map that well to yourself like is that is that what I want? Is that what I'm interested in? And and chase it around until you find that thing. Might be IT security, might be data analytics, might be PM or technical PM work. Um, there's a lot of things in technology, a lot of different roles. Some are softer skills, some are harder skills, all sorts of mixture. I would just explore and have like a lot of conversations about it. That's great advice cuz I I'm right there with you and you I'm glad you mentioned it, right? Like people think of software as like a group, right? or like it is like,"Oh, well, you can just be in IT."And it's like,"No, there are so many different specialties within all of this that like the average person doesn't understand and there's so many layers of it and there's very different skills, right? Like if you're really good at math, you should probably look at machine learning, data science because that's where that is useful, right? But if you don't like coding and math and complicated problems, probably not the best route for you. Um, where can people find you uh online? and we'll put it in the in the show notes in the description below. So, let people know where they can uh reach out. Uh I'm most active on LinkedIn on social media. So, um just my name, Sawyer Nyquis, probably the only one out there. Um and then the data.co/value if it's a email boot camp. Those emails come out from me. If they find things interesting in those emails, hit reply and I can I'd love to dialogue and continue conversations around that. Um those are the two places I would say for me. Awesome. Um, everybody, if you like the show, you like our podcast, like what we're doing, please go uh hit the like button, hit the subscribe button, uh give us some some feedback, whether you're listening or watching um on whatever platform you're using. We appreciate everybody. Sawyer, thanks so much, man. We've Yeah, we've had a great time. It's a pleasure, gentlemen. Thank you much. I have a good one.

Open the transcript file

Hosts and Guests

Hosts

  • Bobby Neelon host Feed-stated
  • John Kalfayan host Feed-stated

Credit the people on this episode

Be the first to rate this episode

More episodes

All episodes
  1. EP 90: Jonathan Henderson from Spotfire

    Nov 16, 20261 hr 3 minAudio

    Jonathan Henderson spent years building oil and gas workflows inside Spotfire at Enertel, and now he leads forward deployed engineering at Spotfire itself. He talks about why engineers making seven figure calls still...

    Credit the people on this episode

  2. EP 89: Ryan Yang from GridPilot AI

    Nov 9, 20261 hr 6 minAudio

    Texas woke up to a blue wave, and Ryan Yang, cofounder and CEO of GridPilot, joins John and Bobby to talk about what it means for data centers and the grid. Ryan breaks down the years long interconnection grind, why...

    Credit the people on this episode

  3. EP 88: Kevin Kong from Everstar

    Oct 26, 20261 hr 14 minAudio

    Building a nuclear reactor in America used to take three years. Now the paperwork alone can eat a decade. Kevin Kong, CEO and Founder of Everstar, went from scaling Bird scooters to using AI to untangle nuclear...

    Credit the people on this episode

  4. EP 87: Mahyar (Matt) Mohajer from Blue Spark Energy Inc.

    Oct 19, 202655 minAudio

    Scale builds up downhole and the usual fix means chemicals, pumps, and a pile of equipment on surface. Mahyar (Matt) Mohajer, CTO at Blue Spark Energy Inc, breaks down a wireline tool that fires electric pulses to...

    Credit the people on this episode

  5. EP 86: Samra Nawaz & Vinay Acharya from WellWorth

    Oct 12, 20261 hr 12 minAudio

    Upstream finance teams have been white knuckling giant Excel models for decades, and the husband and wife team behind WellWorth thinks that era is over. CEO Samra Nawaz and co-founder Vinay Acharya talk with John and...

    Credit the people on this episode