One Person to Do Everything – Ep. 292

In this episode, the team tackles a common hiring fantasy: one person who can do it all - ingest data, build pipelines, model, write DAX, design reports, and still run governance.

Using a Nightingale DVS article as the spark (plus a detour into the Tom Hanks vs. Tom Cruise ‘GOAT’ debate), they lay out a practical way to set expectations, split responsibilities, and scale a data practice without burning people out.

News & Announcements

• FabCon Community Conference (Microsoft Fabric Conference) — The crew highlights the March 26-28 conference (plus workshops) and shares discount code Carlo100.
• Why One Person Can’t Do Everything in Data — The article that frames the discussion on roles, specialization, and sustainable delivery.
• Submit an idea or topic for the podcast — Send a question you want the team to debate.
• PowerBI.tips Podcast — Browse the back-catalog and listen on your platform of choice.
• Power BI Theme Generator (Tips+) — Build theme JSON to standardize colors, fonts, and styling across reports.

Main Discussion

The biggest risk of a one-person data team isn’t talent - it’s fragility. When every decision, fix, and request routes through a single person, throughput collapses, definitions drift, and the whole program pauses whenever they’re unavailable.

The crew breaks down the hidden work that piles up across the data lifecycle and why the right answer is usually not “hire a unicorn”, but “make responsibilities explicit, then staff the gaps over time”.

• Map the lifecycle end-to-end: ingestion, transformation, modeling, measures, visuals, adoption, and governance all need ownership.
• Separate delivery from operations: protect build time and plan for the “run” work (refresh failures, access requests, monitoring, support).
• Reduce context switching: limit ad-hoc requests, batch similar work, and agree on what gets worked next (and why).
• Start with definitions and grain: align on KPI meaning, business rules, and data grain before you lock in the semantic model.
• Make tradeoffs explicit: call out what is manual today, what becomes automated later, and what “good enough” means for v1.
• Document intent, not just diagrams: a lightweight conceptual model and decision log saves future rework.
• Staff to the constraint: add targeted help (part-time engineer, admin, designer, BA) instead of piling more hats onto one role.

Looking Forward

Pick one area to improve this quarter (ingestion reliability, semantic model consistency, or report design standards) and staff it explicitly - even if that just means carving out protected time instead of adding yet another hat.

Episode Transcript

0:27 good morning and welcome back to the explicit mes podcast with Tommy Seth and Mike Mike good good morning good morning good morning jumping in with our main topic today so our main topic for today will be why one person can’t do everything in the data space well more just in data right so if you have data why one person shouldn’t do all the things I think this is going to be a relevant article because I think a lot of times companies expect so much out of that one person oh if I’ll just hire this one person they can do everything in fabric all the data

0:58 can do everything in fabric all the data engineering all the modeling all their poort building all the things all the things and this is a an article written by Christopher lenthal lenthal Nightingale it’s good we’ve done the website a few times with their articles the the Journal of the data visualization Society I guess yeah it’s pretty good so it’s a good article we’re going to go through that one but before we jump in there we’ll do a couple old quick news items or openers here as well so one of the first news items I’ll throw out here just really quick is if

1:29 throw out here just really quick is if you’re not going to the Microsoft fabric conference that is now a thing so there’s a fabric conference that is coming up in Las Vegas March 26th to 28th is the the actual event you can also get workshops on the 24th 25th and 29th and we’d recommend you go it’s actually good I think you can go to Azure or Fabric comp. com or Azure datac conference. com I think either one of those will work and then if you’d like $100 off your ticket you can get a $100 off by using Carlo 100 for $100 off

2:01 $100 off by using Carlo 100 for $100 off your tickets I would love $100 off my ticket I would love $100 to our show to know that we the Benjamins are flowing towards you guys Mike you’re speaking there right I’m speaking in two sessions oh nice so I’ll have two sessions I’ll be speaking in so those would be two that you can bring your pillow to and take a nice nap during my sessions I’m actually getting closer to going to the conference now I’m like what that’ll be fun yeah it’s going to be good I think I I in in a lot of many Microsoft things

2:32 in in a lot of many Microsoft things it’s a lot of salesy things it’s a lot of like stuff that’s going on but we’re going to do some deep Dives deeper Dives on some best practices or things that work well in how to manage your spark instance is one of my sessions the other one we’ll be talking about Azure data Factory and pipelines or or sorry and as your data flows Gen 2 some of the new things that are coming out there as well so just more investments in learning or teaching or educating people around what is inside fabric which should be great look forward to it Tommy you had a another opener topic here just want to go it’s not really dat

3:02 here just want to go it’s not really dat related but it’s interesting it’s a bit it’s a bit off the wall so this is our podcast before the Super Bowl and Patrick Mahomes is being in it and he’s there’s that comparison of Taylor Swift is in the Super Bowl what you meant say isn’t it something something like that I’ve heard so but there’s a comparison of him being the goat compared to Tom Brady goat meaning greatest of all time okay now outside of football actually I was listening to something and it raises the question

3:32 something and it raises the question who’s the greater movie goat Tom Hanks or Tom Cruz and at first I thought I knew this right off the bat but as I thought about it more I’m at a loss this is your opener this is my opener yes you’re you’re gonna make us publicly choose between Tom Hanks or who’s the greater go what if what if I think somebody else is a greater actor yeah I think Seth was thinking Nicholas Cage I believe well let’s say in the generation of 88 88 to 200 get away from

4:05 generation of 88 88 to 200 get away from that accent well yes the accent scre but what else did he do you can’t see he’s better goat let’s say for the sake of today for the context Tom Cruzer Tom next time we are filtering your your okay okay of the two I like both of them they’re both very great actors of the type of movies they create I think I like the Tom Cruz movies better Mission Impossible I think are really they’re a staple I like those I think those are good I’m going to go in the opposite

4:36 good I’m going to go in the opposite direction only because when I was growing up big was one of my favorite movies oh big yeah I never really got into that one in different types of roles Tom Hanks man like I I think just the the breadth of characters that he’s played didn’t he do that one about the airplanes and never leaving the airport or something terminal teral not the best accent good movie but not the best accent good movie good movie Terminal he’s been in a lot of War stuff

5:06 Terminal he’s been in a lot of War stuff he’s also behind see I don’t do war movies I can’t do war movies Geto em the masters of the air which is like the the new version coming out there a lot yeah I’m gonna have to throw my hat into Tom Hanks and that’s where I originally was so this is where the context changed for me after catch me if you can when you actually look at what Tom Hanks did not really some of the greater films compared to Tom Cruz he had 10 movies in a row where he made a100 million doll where the movie

5:37 made a100 million doll where the movie that’s but that’s not that’s like what an actor makes is a completely different like well who’s the goat guess made are are you asking who our who we think the favorite is based on criteria like if if it’s I just I like the best okay then that’s than based on criteria of how much they made how many movies they’ve had how popular they are like I I don’t know right like I watch I am also the guy who likes to watch movies and 99% of

6:10 guy who likes to watch movies and 99% of the time do not remember the actors that were in there like the face right like oh yeah I totally recognize that oh yeah that’s their name but I am like my friends are the complete opposite they can name every character in all the movies they’ve been in and and all this other stuff and I’m just not that guy yeah and I think the perspective is Hanks has that more probably death of type of acting Tom Cruz you go I’m Tom Cruz we’re going to have a fun time stuff is going to get blown up have fun with me let’s go but he’s also he’s also had some different types of roles it’s

6:40 had some different types of roles it’s true but he’s played a lot of the Special Operations or military things as well yeah it’s a harder conversation less drama Tom’s Tom Han’s been more drama than oh yeah yeah Cruz this is all anecdotal dude I don’t I don’t know we’re the wrong folks to be doing film critiquing about anyway Tommy where do you welcome to your welcome to your bi and film critique show a little fun it’s a

7:10 critique show a little fun it’s a Saturday morning for us we’re pre-recording we’ve done 292 mind you without fail we’ve never missed a an episode so so I interrupted you you’re now bringing into gross net worth how much money they’re making so you land on the Tom Cruz train I’m saying from a goat point of view greatest of all time out of those two I may like Tom Hanks better as the actor but I know if I’m getting Tom Cruz in my movie we’re gonna do really

7:40 do really well okay so I I think I think you can use I think this is Let Me Tie this back to data just here a little bit right right right you can perceive things however you want right you can use data to be as like a proxy for popularity this is kind like a proxy for popularity this is like where what do you what of like where what do you what do you measure as what is your what is your definition of success what does it look like right so you can you can go into well how many awards did they win well Tom Hanks won two Oscars while Tom

8:10 well Tom Hanks won two Oscars while Tom Cruz has never won an Oscar like so you can start looking at like comparing that things and so you comparing that things and so assumptions are whoever’s making know assumptions are whoever’s making the nominations for Oscars they’re looking at who’s making good acting choices and who’s doing good acting in the films then you can look at like box office things as well so you can say okay how in every movie that Tom has been in what’s the cumulative total of all the money that’s been returned by them to your point Tommy like if I’m going to put Tom in a movie will we get a good return on that movie and I’ve also seen some very short YouTube video clips of like Tom Cruz is

8:40 YouTube video clips of like Tom Cruz is like crazy he does a lot of his own stunts he loves being that stunt person so like from a I’m Allin but he’s like a it’s like a I wouldn’t say it’s Sci-Fi movies but he’s more of like action-packed movies that’s and that’s a that is a genre that I like more than some of the other genres that are out there but if you look at that as well Tom Cruz is I think it I’m looking here it’s got around s 10. 7 billion is his and and Tom Hanks is seems like it’s 11. 3 they’re very close so we’re like

9:12 11. 3 they’re very close so we’re like 1 billion apart in like who’s got the a billion dollars is close well when you’re doing doing 10 to 11 billion billions right yeah yeah right when you’re talking billions so that similar there as well so like can bring in the the money or the revenue but again I think you’re looking a lot of these These are these numbers are reflections of how people perceive the popularity of a lot of other things it’s not just you can’t put a Tom in a movie and be like oh yeah Tom is the only thing that counts towards the benefit of that movie to an

9:43 towards the benefit of that movie to an emotional response and what I should have had Tommy do is Define what he meant by goat like in specific measurable measurable things literally back to implementation failed I failed thank you Tommy but so all all this all this said though correlates this is I think the same challenge you face inside businesses where you have different people looking at the same question there’s there there may be an answer but the different kpis you pull out and the different things you look at change your perspective slightly

10:15 look at change your perspective slightly on what the answer may look like and then you recognize that all of it was just false completely because you chose two people out of all the men in all of Hollywood to choose between them saying which one’s the goat as opposed to saying who is the goat based doing two Toms was pretty good though it was good it was good I’m just saying put

10:39 it was good I’m just saying put limitations on it I like the critique feel this feels familiar this this feels familiar what what kpis you decide so this this really F ultimately this goes back to okrs right what was our key objective here right and and you our key objective here right and and how what goals did we want to have know how what goals did we want to have is our goal to get Awards is our goal to make money like what what are we doing here so what I’m digging is what I’m digging here is you guys saying I’m not really digging the visuals that you chose here Tommy the colors here could have been a little better I think your colors and your report here Tommy don’t

11:10 colors and your report here Tommy don’t really match the theme of our company we we’re not branding correctly here if you could censor the titles a little bit more and the colors seem not quite right there too much too much color here who cares how much what the actual report looks like and what the data is actually showing but let’s just let’s critique the report look and feel only yeah lipstick on a pig good intro good opener I liked it it was a thought-provoking moment Tommy but from now on all your other intros will be

11:43 screened what screen from getting this one out worth it worth it this out the door meanwhile there’s okay for all of you who have actually hung around for the movie podcasting conversation let’s transition into our real topic today why one person can’t do everything in data take us off give us a couple like key nuggets here that you’re thinking through and on this article well I’m not going to think I’m not going to throw in my thoughts yet okay you should outline what the article is describing that’s fine give us a little bit of an outline what’s going on yeah so ultimately the the

12:16 going on yeah so ultimately the the point of the article is how business has a tendency to look at an individual who is creating value with data and assume that because they’re successful at doing certain things that this infinitely scales into all other areas of data and how that quickly becomes a problem not only for the individual but for the organization as well not understanding why things aren’t getting done when in the past things were so he lays out kind

12:48 the past things were so he lays out kind the past things were so he lays out an analogy of and the importance of of an analogy of and the importance of the different roles and people right that are skilled in the particular areas of data and the the main areas he talks about are data Architects data Engineers database administrators data analyst analysts information designers and data scientists so while he doesn’t lock all of us into the analogy because he kind

13:18 of us into the analogy because he kind of us into the analogy because he opens the do saying this is of opens the do saying this is just an idea of how you build the cge dorm or uses this analogy of like how that dorm is built yep etc etc and we can talk through those different components But ultimately the the point is outlining that it’s important for organizations to understand that there are different roles there are different skill sets and to be cognizant of those as they’re trying to develop

13:49 as they’re trying to develop Solutions did I sum that up I think it’s a very good sum up I like I like the middle part of the article where you touched on he outlines what he perceives as different groups of skills that potentially you find people not not grouping themselves by but like things that people are interested in and I like to think of there are a couple of these skills that people will be interested in and they’ll gravitate towards those type of skills right I’ve seen a lot of MVPs gravitate to the data architect data architect data

14:22 the data architect data architect data Engineers database administrators and maybe some data sciency activities right that those people have they like the data they like thinking through the numbers and producing the kpis the outputs the measures right that might be a certain skill set data analysts information designers dbas they may be people like that type of work and kind people like that type of work and gravitate towards that space so yeah of gravitate towards that space so yeah I think this is a I think it’s a good article and I really like the buckets that he presents here and I think that’s a a good some good talking points we I have a question right off

14:52 points we I have a question right off that we have a a wide spectrum of users or or people that listen to us where I think this this issue initially I guess impacts people more is more on the business side that’s just my perception how how prolific do you think this problem is like because the core thing he’s pointing out is someone starts creating value with data in a business right not not a not

15:23 data in a business right not not a not an organization that’s data literate that’s like completely has all the bells and whistles of the adoption and what we talk about on Tuesdays at it it’s I would probably say the small medium size businesses or business areas where someone pops up and all of the sudden people realize that data can can do good things yeah I would maybe say some of these some of these skills are more formly trained right so I would say I’m

15:56 formly trained right so I would say I’m just saying like the problem right like the one person how does that like do you think it is does does the problem he’s outlining here live rampantly across organizations I think companies don’t have a good measure to know what their data culture looks like and I think inherently at every company there is a data culture whether you like it or not there’s something that’s there and where where I again when I’m looking at when I look at this from the business right there’s usually this there has traditionally been this fight between

16:26 traditionally been this fight between I’m the business give me data and then they reach out to it or technical experts in the data engineering database administrators and maybe even the data architect level and say okay we have purchased this computer system to track all of our sales well help me get it out I I don’t I don’t have the Technical Training to get the data out of those things but I do know once you give me a table of data I can shape it so I think the data analyst or the business side of things has a way of taking information and try to shape it in a way that helps

16:56 and try to shape it in a way that helps them make decisions so so I think it’s being done regard it doesn’t M it’s like it’s usually happening in like Excel and maybe in Access typically but as that data culture matures and now the technology pieces are giving us more capability to do higher volumes of data and Analysis now we’re seeing the data analysts and the business I think reaching further Upstream into more of those technical roles back into the central bi team or it or whatever you

17:26 central bi team or it or whatever you would want to call it so I I think there’s a there’s a there’s a shift happening where the tooling is giving more capability to users and as data culture needs to evolve now the data culture now needs to educate or prepare or have plans to help these data analysts which are doing great things they’re making decisions on data and figuring out how to run the business businesses wouldn’t be running if they didn’t have any data to support that what’s going on I I see where you’re going with that

17:56 on I I see where you’re going with that Mike because for me very similar the technology is I think to me is prolific like a virus at organization yeah it’s it’s commoditizing data to some degree well this is the difference because I was thinking why is one of the article deserves to be written why is this such a case I’ve experienced this myself and I’ve heard of a ton of others well this is not like back in the day where your skill set was so siloed where this was like years ago when someone asked a

18:26 like years ago when someone asked a colleague of mine to build an SSRS report and didn’t know SQL but you had to know the tooling all the technology is part of a user interface and this is even more prevalent with fabric but let’s just go back seven years as powerbi to do modeling visualization reporting communication all I had was the single UI and my expansive skills is is immense where that was never the case before really in any I think any type of Technology much less just dealing with

18:56 Technology much less just dealing with data so we’ve gone this prolific and I like the word they use there Seth because I think it is so prolific at any organization who’s taking these te this technology because really to a degree one person can do a lot of this yes but they can’t scale it and they can’t do it for long but you days that’s another part of this question and I really like where you went with the scaling and yes cuz that cuz you’re right I can achieve a lot of this yes you can one person can

19:27 a lot of this yes you can one person can do it all do it all but for how long and for what size scale are we talking about there are unicorns and there are people that have a lot of all these skills together however I also would argue you’re the the amount of information that is required to know to get to this level where you could do all these things is going to be very hard to find someone who has all of this information see to me it’s the sustainability of it I think someone could figure it out what figure it out once where I had this particular common

19:57 once where I had this particular common problem that I figure it out and I learned a skill in a certain area yeah but can I sustain this one from my own State of Mind do I have enough hours in the day yeah but I was only dealing with one one situation and again the technology allows us to solve a lot of this at least once in in one particular project but the problem is the Assumption and and from what I see from the article but what I’ve seen happen in myself is well since you’ve done this you can obviously solve other

20:27 done this you can obviously solve other data problems and do all the data things because you’ve done this why would we hire anyone else because this has been solved not understanding the amount of time that took and if it was actually built to actually be long lasting and the yeah I I like I like the conversation I think it’s jumping more ahead of my my question which is fine because I I think what I want to stress that I fully agree in in this article

21:01 that I fully agree in in this article is is I if if something organically is happening where an individual is bringing value to the organization through data a lot of times that’s going to be Excel right it’s GNA or it’s going to be a tool or something that they know

21:17 to be a tool or something that they know how to do that with that instantly can change especially in this new EOS system where you guys are talking where there are Technical Solutions here there are maturities and cap abilities to scale and reach further upstream and like instantly go into the conversations that we usually stick in but I think what he’s pointing out here is these people don’t know those things like our audience we love having part of this community because we talk about these things all the time but in like if I

21:47 things all the time but in like if I look at the Spectrum of people in organizations that are creating value how many of them are engaged in communities and constant learning in Tech skills it’s ET ET not as many not as not as many I would agree so so that’s where I think I I think you guys agree with me and I think we jumped ahead to like hey what you have to do is this next next next next next because the Technologies the and the the tooling especially in today’s day and

22:18 tooling especially in today’s day and age is in a state that you could learn it it’s not this obscure like landscape that you have to completely ramp and like but you still have to learn and you still still I think automatically get pushed in where this article goes which is these people know how to bring some value out of data but what you’re asking them to do is be a data engineer like in the in the fabric landscape great you can create a

22:50 fabric landscape great you can create a report I heard Microsoft is introducing fabric now I want you to build this whole ecosystem that’s exactly what this article I think is pointing out as is the problem where you may have somebody that is that is capable you might have an idea that they should be able to handle all of this but that doesn’t make them a data engineer that doesn’t make them a big data expert and understand spark and notebooks and machine learning and all these things and I think it just reinforces that like where I’m leaning is I agree I I think

23:22 where I’m leaning is I agree I I think one of the challenges of business in general is to assume that somebody like and even in the realm that knows powerbi right because we make fun of that in quotes a lot of the times is I know powerbi well do you like the whole spectrum of Technologies underneath there are you an expert at it or are you the person that can like connect to an Excel file and create a report wildly different things right and that’s where we talk about skills matrixes etc etc so I I do think what I I like 100% agree I

23:56 I I do think what I I like 100% agree I think assuming lot of individuals even from a business perspective and or powerbi user perspective puts people in problem in in these situations where I think this is a very relevant article because of the fact that it’s bringing to light there is a lot more under the covers and how do we communicate that to the business the only thing I I think if I’m hearing correctly that I would slightly disagree with is the idea that

24:27 slightly disagree with is the idea that well this normally happens when you have those unicorns those Superstores who love to learn I to me even that person who’s doing something simple in powerbi I think they assume they can do a lot more and all that makes them is just a master of card building where there’s a lot of house houses of cards at the organization because that Excel thing like well I must know data engineering because I put this structure together and yeah we can intake data and create this automation so we can really extend that when they’ve done done maybe one or

24:58 that when they’ve done done maybe one or two projects I don’t think this has to be that person who loves data like we do where that’s the prerequisite I think this happens with honestly because of how easy and how TR transformative powerbi can be just taking Excel files and making them into reports not I was not even a call yesterday showcasing all these Excel files and we were showing how we would build a model from it connecting to what would be as a standard table and the amount of Wows and you’re whoa I never

25:30 amount of Wows and you’re whoa I never know that we had that oh my gosh and all those Wows that still happens even with Excel Excel files and what I’m trying to say here yeah so what I’m trying to say is it doesn’t take much for the Assumption and the load to go onto one person where it’s not because they’re doing amazing things in powerbi all I’m saying it really takes one or two things for all of a sudden this influx of assumptions of what that what that person is capable or assumed to be capable of

26:02 or assumed to be capable of doing so maybe it’s not disagreeing but I’m I’m taking a different look at this as it does not take much it’s not a this gradual process where that transformative effect of the the load on a single person or team can occur so so that is an interesting point that comparatively like the the opinion of the article are you disagreeing with

26:32 of the article are you disagreeing with to some degree where like the article is pretty straight out like you need this person this person this person this person or this role this role this role and and these are not the same things and you’re saying well in today’s world the tool sets would allow somebody to understand these more and I think that’s where you’re at as opposed to disagreeing with me that me that my my opinion from just the individuals that do a thing MH are not the like that

27:04 that do a thing MH are not the like that group of people it’s a small subset that is going is going to improve themselves to learn more just because they spot on you’re spot on okay so yeah so but I do think you bring an interesting point where are do the tool sets of today and be it fabric be it powerbi as a standalone right right break down some of what what is being pushed at as an

27:35 is being pushed at as an individual that does this role in versus like an individual that does the role right can can those skills be married or merged or learned pretty easily because of the tool sets that we have today so let I I want to jump in here you guys have been talking some really good things and I want to I want to throw down some analogies here that I think will this is more analogies my my brain runs analogies or no yes if Tom had a train that was running South

28:06 Tom had a train that was running South at 45 miles an hour and and Tom Hanks was running train at North at 50 miles hour which way does a smoke blow had a jet yeah exactly Tom Cruz is in his private chat just kidding so I have a couple I was this is funny because this literally came up in conversation in my family recently and we were talking about you different people in our family were talking about how they interacted and wrote things down on paper like writing papers my son has just recently done a paper through a competition and actually

28:37 through a competition and actually wasn’t even a paper it was actually a website so he he actually made like a little from a little website builder wrote words and paragraphs and then added images and then cited all his references all these things so he has this whole little mini website that he does and they apparently they were competing on this he did really well we’re very proud of him but that perspective compared to let’s call it more seasoned individuals in our family was talking about how would you write a report and the the the writing a report a long time ago before

29:08 writing a report a long time ago before computers it was on a typewriter and you had to write the report using a typewriter typing everything out and if you got the wrong comma or space or period in the wrong spot or if you were doing references remember remember I don’t know you you guys remember this one but doing references you had to go to like a library find the book go get the pages and then you had to go look up like the APA formatted standard that you’re going to use and you had to go copy various pieces out of I remember I got to college I remember I got to college and I was like you can go to a

29:38 college and I was like you can go to a website you can literally just Google the book and it just says here’s the book tell me what page number it was and it just spits out the whole reference so I was like oh my gosh this is incredible so to go from the spectrum of like we’re typewriting things out you’re trying to reference thing the amount of time taken and and my family member was telling me he would write a whole page and if he messed up halfway through you threw out the page and you started over you started typing again could you imagine in word if you got through a page and you got halfway through and word you’re like whoops I put the wrong keystroke

30:09 like whoops I put the wrong keystroke you gotta delete the whole page and start over again white out why do you think white out was white out white out was yeah yes out you’re like over the typewriter I know yeah never let it dry blow let it dry go back be impatient do it too soon now now the but think but think of all you mean this is to me that analogy fits very well with we’re talking about here at data now because this is the same thing right the the effort it took to get data

30:41 right the the effort it took to get data processed and completed and changed and manipulated it was it’s now only possible with computers and now that we have computers we can do all these calculations much faster there’s things that are built in ways that we can use them much quicker and every every new technological advancement here is pushing us further away from like the the the core of what was being done and again I think about computers another analogy that goes along with this as well is early computers were nothing but code and cards the early computers

31:13 code and cards the early computers you had to put a card in it’s a punch card to program the computer we don’t do that anymore the the computers are much smaller we have we’re not using Punch Cards to program computers anymore and now we have Bas programming languages and we have mult M layers of programming languages on top of it that make it easier for users to do things now you can build power apps basically an application without even touching most of the code you can you can pretty much have it either automatically populated and we’re getting to this place where it we’re requiring less and

31:45 place where it we’re requiring less and less deep technical knowledge of like where the code and the data and the things live and because of that we’re

31:52 things live and because of that we’re making it more accessible to other people my final analogy here and again I I’ll keep get in my soap box here the lake houses that that we know today have never been able to be built like that until now before lake houses were all on Hadoop I would store a bunch of files I had to manage what was being created I had to save the data down all these there was no acid transactions inside storing data inside lak houses it was just very manual so companies were using it they were building it but you had no way of of of grooming that data and

32:22 way of of of grooming that data and making it easier well data bricks comes along they produce this spark thing it goes much faster technological advancement has occurred and now we have Delta tables that can handle asset transactions that someone has solved these harder problems for us and so a a small group of really smart people have built a new technology thing that has now opened the accessibility of that thing to now hundreds if not millions of people around the world to use these Technologies I do love the website analogy well yeah but all these

32:53 website analogy well yeah but all these things so everything I’m seeing here is know people keep putting smart brains on solving these what is the most difficult things about data and it’s going to keep getting easier to do it’s going to keep getting easier to run we’re going to have things like we see these different roles here right yeah maybe 5 10 years ago we had more data Architects data Engineers dbas now what DBA do you need for for powerbi there there’s maybe some

33:24 for for powerbi there there’s maybe some organization but I feel like a lot of the roles those roles have changed now because I’m not actually worrying about a server anymore it’s those things have shifted to other roles so I think I think yes and no yes no I agree but let me just say so I feel like what we’re going into is these roles are turning more into a Melting Pot to me right the roles were very hard hard set because there were specific skills needed to do these certain things I think the technology is is reducing the barrier to entry to some of these roles and I feel

33:54 entry to some of these roles and I feel like we’re seeing a Melting Pot of this swirling more and more together I I don’t agree on I agree on some aspects I don’t agree on others I think there is a Melting Pot related to how people are engaging and working with data that’s okay Tom I think where that that gets where that is a completely different thing is yeah depending on the structures that you’re still using a DBA is required for a SQL Server that is a managed instance right like so where

34:24 managed instance right like so where data lives how it’s managed it’s ET I think take specialized roles if you want those systems to work what I do agree with with you especially from the technology perspective is things are moving in a direction whereby we don’t have to invest in the storage in the speed of what something like that thing that’s serving up the data I think is more and more pushing into Realms where you have Services you have yes you

34:56 where you have Services you have yes you where you have Services you have yes services that allow you to only know services that allow you to only care about the data and I think I over time th those aren’t completely independent yet although it’s gotten way way easier with a lot of these Services I do think to your point over time it completely goes away like a Ai and looking at like how data is being used in a system you can’t tell me you can’t automatically reindex repartition redo redo all things yes dynamically or at on

35:26 redo all things yes dynamically or at on scheduled basis you don’t need people to do do that because people are are going to do it worse than the systems that understand like how to how to serve that up better I agree I agree from that standpoint there are Technologies now here and in the future that are just going to make it easier so we can focus more on the data itself instead of the infrastructures that are used to to build the data oh on I got some problems

35:56 build the data oh on I got some problems so so I got I got some problem just to be clear I just I want to talk about Seth real quick comment here real okay Seth said AI is going to do things yes I know because I used AI to summarize this article and so I I I’m using it more and more for all these like I used AI to help me out with Tom kson and Tom Cruz like I don’t need to be expert on a lot of things anymore you can supplement your knowledge much faster now than you ever had before I got some important information too as the status

36:27 information too as the status AI guy at the very bottom of my chat GPT it says chat GPT can make mistakes consider checking important information so that’s one soap box but I I really want to focus on the Melting Pot idea and I think if we said there’s a prolific virus going around on this strain on individual users or teams the idea of the Melting Pot is the cause or that’s like the origin of this because with all the elements right I my first job out of college I worked at a marketing website agency where we charge

36:59 marketing website agency where we charge an arm and a leg to build websites that were cool but janky and now I have square space to manage my own website for my own company that’s fine but the difference is and this is the big kind difference is and this is the big the outlier it works to a scale not of the outlier it works to a scale not to scale it works for me though I don’t need to know HTML I don’t need to know all the intricacies of a website because I’m building one for me for one individual and that’s a lot of that’s my original thought here where yeah I can build a lake house in Fabric and I can

37:30 build a lake house in Fabric and I can do a notebook and I figured out some of those things but until I know how this works in a pipeline and I don’t mean a data pipeline in a actual organizational pipeline when all things are coming at you there are certain skills that you need to know and the idea that well because you can do a visual you are a data visual expert that is not the case because no I’m not saying that I’m not saying that I think you’re going a little bit too far on what I’m read I think you’re reading a bit too far what I’m saying is okay I think I think I understand your point yeah my point here is my point is less

38:02 yeah my point here is my point is less around that it’s more around the commodity aspect of data engineering of data and moving around it’s becoming that’s that’s purely my point I do recognize there needs to be skilled training skilled thoughts around how do I communicate through data the visuals and I would not assume hey you’ve been building power B reports for a year okay you’re now a visualization expert that’s not necessarily the case because like I’ve been I’ve been around people who’ve been oh I’ve been doing powerb for two

38:32 been oh I’ve been doing powerb for two years great tell me about filter context no clue yeah like we I know that there’s gaps there but I’m I’m more talking about the idea that if you were going to do visualization development previously even even that by itself it was either only in Excel or you’re doing d3. js and writing everything visuals from scratch like and that’s why Tableau shows up and Tableau goes look you we have a way of shaping some SQL queries and letting you drop data fields down to a table or a visual and boom

39:03 down to a table or a visual and boom stuff starts popping out again this is another thing the whole visualization space has also simplified Itself by giving you new tools that let more people access that same information so so let me rephrase it a bit and I think this will be a little more proper just because I’m capable to do a certain solution or capability doesn’t mean that should be my requirement or my expectation and I think that’s what occurs just because I’m capable to build a pipeline see that point I can see that

39:33 a pipeline see that point I can see that right or capable to build some data engineering and fabric just because I can do it and maybe I’m okay at it doesn’t mean automatically that should now be my expectation at the organization or in my primary function disagree with you one yeah I agree okay but I don’t think that that’s not that has nothing to do with my point though my point is less around that it’s more about the the ease of you the ease of learning the ease of accessibility to these things and so you accessibility to these things and so no longer do I need to write you

40:03 know no longer do I need to write you know no longer do I need to write for example I don’t have to write know for example I don’t have to write the Dax formula to make the visual run the Dax just happens behind the scenes and parbi just makes it work I don’t have to write the M code to make the M go grab the data from the data source I just write the UI I click on the button so to it to me it’s more of the I still think you need to understand the concepts there’s still some education training that needs to be occurring there but the ease of clicking on buttons that’s the general Trend the general trend is to make all these very technical aspects more easy to absorb

40:34 technical aspects more easy to absorb and that’s actually one of my bigger challenges with Microsoft fabric right now is there’s a lot of things that fabric is doing well but that’s not addressing what I would consider are the main Core Business problems main Core Business problems is how do I handle slowly changing Dimensions how do I handle an incremental loading table in a fact table like the how do I what in my model if you if I throw a bunch of tables at a computer software how do I understand where should the dimensions and where should the facts

41:04 dimensions and where should the facts live these are things that people are being applied to do right now I think Microsoft should take notice and say okay we will identify these areas of common like it’s a common problem everyone’s trying to solve over and over again let’s build a UI let’s build a system in place that handles 90% of that problem for you so you have to think about it less if you other tools are doing this Microsoft just hasn’t done it yet in fabric it will get there I’m I’m confident but right now they’re behind other companies

41:35 right now they’re behind other companies and other tools so the last real quick I’ll say on this and then I’ll jump to Seth is I think you and I agree but what the problem the Crux of what you just said is yes you still need to know the fundamentals but to an organizations like oh well since the button can do that for you and it can cause that well we can put that on you too because we don’t we don’t know all those intricacies we just know that you’re capable and it’s possible now so you can figure it out that’s what I see occurring I think I think what’s really

42:07 occurring I think I think what’s really poignant as you guys are talking is if that this article brings up is and accelerates to some degree is because we have tool sets that are opening these doors to individuals I think I think there there are pitfalls here for businesses that actually make this worse right so what Christopher outlines here is like he’s saying listen these are

42:30 is like he’s saying listen these are these are different skill sets regardless if it’s one person or many these roles are are not interchangeable and some of them are very specialized like like theyre that takes a lot you have to know what you’ve done it before before you can actually architect the good decisions D I say you need training for DBA yeah or experience right or experience yes the the analogy he uses is a college room in remote Town wants to grow doesn’t but but can’t because the structural limits so the first thing

43:01 the structural limits so the first thing that they want to do is build a new student housing so he he walks through and he says you need a contract a construction firm to develop a new build a new dorm those are data Architects what are we going to build contract moves movers to populate the dorm data Engineers hire a superintendent manage of the dorm database administrators hire resident assistants data analysts higher communication staff skilled in telling student stories information designers contract with researchers to study the impact of these students data scientists

43:32 impact of these students data scientists okay and and while like part of me doesn’t like that analogy because it’s like okay these are some of these can be shared we have these new tool sets that allow people to do more I do think the cautionary thing that’s coming out of here is especially with things like Fabric or even powerbi and we’re talking anecdotally about like people and what they can and can’t do within here is is essentially Nestle down to like this

44:03 this idea with people who are pushing data and doing things just because they can make it work doesn’t mean that it’s built built right right and and while there’s value there because the business gets insights quicker I I think what’s important is that organization needs to understand that what’s being built is neither efficient nor performant unless that

44:33 efficient nor performant unless that person has the skill sets or is learning them behind the scenes because all of those are very important still these tools aren’t like creating the best Solutions all the time or guiding lay people who don’t understand the Dynamics of them what they’re doing is allowing them to do something and make it work so I guess the caution in here is yes more and more people have access to

45:04 yes more and more people have access to these tools they’re getting simpler but you guys are in business because businesses still struggle with understanding how do we how do we get this throughout the organization how do we skill people up what are the things that they need to learn etc etc and there’s a cost there if you don’t invest or can’t invest in some of these skill sets right and and that’s I think the the blessing and the curse with business tools that we have that allow quick wins

45:37 tools that we have that allow quick wins and allow business units to get data faster than they ever have before there’s a cost there until you start to support that with people who have the skill sets to think long-term to build things that are going to be efficient so that as you’re spending x amount on on licenses and whatnot that it remains that way for as long as possible otherwise you find yourself in this like everybody can do it the skill set like the tools are there we just expect

46:08 the tools are there we just expect everybody to do it and what happens your costs explode and we Mike you this very well because there’s a guy who works very closely with you from a previous org and what happened boom like just like massive explosion in cost because why because people without the skills were doing things and they were making it work but it wasn’t efficient I think that’s the cautionary Tale in here is like yeah the small College remote town could grow could

46:40 remote town could grow could they do that by slapping in a bunch of temporary housing and then assuming that that’s going to work and everybody’s going to be happy with that for a while but eventually and that’s where I I dovetail then into okay how do we get this to work for us could could you slap together temporary housing quickly while you build the dorm right could you rapidly expand which is where these tool sets allow us to play right now yes but I think the

47:10 to play right now yes but I think the importance here in my mind is any of those quick wins that allow us to get insights allow us to get data Etc are are only going to serve you when you’re also working on the long term strategy and that does require these skill sets it does require these people otherwise you’re you’re going to create quite the conundrum where a business unit or many business units right are going to start leveraging all these

47:40 going to start leveraging all these tools and playing in areas that they are not skilled in but do work you’re just going to pay for it in dollars for you going to pay for it in dollars for for for services as opposed to know for for services as opposed to dollars for somebody who can extend your dollars I’m going to set there’s so many good points in this one that you talked about literally I had to start writing down like bullet points of like what how we going to respond to this is okay I’ll just say this a. m. conversations are good but man a. m. or 11 a. m.

48:12 are good but man a. m. or 11 a. m. conversations are way better holy I’m having a blast man this is great okay so I want to go back to your phrase you you you led with just because it’s just because it’s built doesn’t mean it’s built right and I would agree with you right right I have a little bit of hesitation when you say that and my hesitation comes from there is a balance between value and cost on these things sure and so I feel like to to your point again going back to the analogy of the school and toally everything I agree with a lot of

48:42 toally everything I agree with a lot of what you’re saying I’m not sure every company has the skills to your point I’m not sure every company has the skills day one to build it right every time and sometimes it’s going to it’s going to be like it’s going to be a bit of a a road or a progression as you learn and it’ll it’ll mature as you go so one thing is I think in in hindsight sometimes you look at projects like ah I shouldn’t have built that way or I need a couple more revisions so I don’t think you have to ever I don’t think you have to stop build I don’t want you to feel like you have to stop building just because you

49:13 have to stop building just because you don’t feel like it’s 100% right so I do think there’s an effort of like sometimes you just need to get going now my point here is there’s because sometimes a company or leadership is not willing to spend the dollars to get the right solution so sometimes it’s a I’m in an emergency mode I need to get stuff done and I feel like there’s a management problem that occurs where H let’s just build it and then to your point Tommy earlier just because you you you now have a visual person who’s now building all the visuals now that’s their visual

49:43 all the visuals now that’s their visual expertise well the company hasn’t taken time to slow down invest train them up and really done a good job so we get by with what we have but there is this balance between there’s enough value being produced and the cost to produce that solution is reasonable now as you as you continue talking Seth you actually made this point in your conversation and my my question back to you well it may not be built right but does that really matter and I think it does it does matter but

50:13 think it does it does matter but sometimes the suboptimal solution works for a short period of time and over time to your point Seth the cost of a nonoptimal solution will creep up and at some point the cost to maintain or the cost to do that suboptimal solution Falls over and and the cost like why are we spending so much money do this very little value added thing and I think at that point in time you’re now saying you’re stepping back and saying oh I need to need to rethink the what we’re doing the process

50:44 rethink the what we’re doing the process because the value I’m squeezing out of it is not as much as it should be for that amount of spend and I think this is where this is where technology changing changing changes that discussion right if technology never evolved you would just do it the same way every time and nothing would ever change you would just you would just do the work and it would be done and you never there’s technology is never making it less expensive in the future however we know as technology gets better it’s going to continue to reduce costs for

51:15 going to continue to reduce costs for doing more complex more bigger things larger amounts of data all these other stuff so because the technology has been changing we’re now getting this evolution of we do need to sit back and look at our process and think through is there better ways of doing things are we overspending for the value we’re getting from this data this process or whatever that thing is and that’s where I see fabric coming into play here is companies have to really sit back and evaluate and say wow I see fabric as being something that could add a lot

51:45 being something that could add a lot more value now than trying to do it all in Azure and all these other things and all this extra effort so because Microsoft is changing the tech I can now sit back and say what we were doing wasn’t good enough let’s let’s re-evaluate and my only my only Point here is management should just be aware of these things and Technical leaders in these developed Solutions should be thinking this way and providing leadership with at least hey we think here’s some things that we’re doing

52:16 here’s some things that we’re doing these couple areas are probably inefficient I just need you to know about them they’re may be not a problem right now but there potentially could be better ways of fixing those things and bringing the cost down that then adds more value and this is where we go with like Tommy I think a lot of this reminds me of conversations you have with people and clients is I have we have a central bi team that is unwilling to let self-service happen there there is there is an unwillingness for for me letting other people hey here’s a data model go build what you

52:46 here’s a data model go build what you want the delegation of responsibility to that team now that company may not have the data culture to support that day one however it doesn’t mean if you if you spend some time and some effort doing that maybe there is opportunity for allowing the business to have more designed models for them and letting them build their own reports will’ll

53:06 them build their own reports will’ll maintain the models business you can make your own reports in your own workspaces publish it how you want you’re responsible so I think I think there’s a lot of really neat conversation and around this this area and I to go back to these positions I’m going to bring it back home here right these positions I think are very much needed and I think is as long as companies understand they may or may not have all these skills but it’s a it’s a progression you need to know where you’re at and this is why the adoption road map I love because it gives you

53:36 road map I love because it gives you those scores how are you with your processes how are you with your people how are you with the technology what have you implemented from a process standpoint and how does that help you move data around you can actually score yourself and improve your organization and that’s being self-aware of this is hard so I I I want to take my own I don’t rebuttal is not the word but from the framework that says talking about we now bring Tommy to the stand yeah right yeah I testify that

54:06 the stand yeah right yeah I testify that I love what that said there’s a lot of unassociated debt with yeah we can build it and it can look like it’s right but the problems that occurred are I’m calling them unassociated because or you’re not going to think that they’re directly correlated ownership all the knowledge if I bu it one way is usually going to be with that person and it leaves with that person why that does that always occur I think that’s always because of the data roles one person is trying to manage all of this but I’m

54:36 trying to manage all of this but I’m actually surprised we haven’t taken the first sentence from this article which to me is I think my favorite part of the article and the boldest part all this to me is sounding more about the data literacy and the so the sentences data roles are often the first subject I discuss when building data literacy yep why would he start with that well all the things we’re talking about here and if we go back to our definition of data literacy from the Articles we’ve read and the conversations we had the ability

55:06 and the conversations we had the ability to talk read communicate and argue with data data well with all these roles and trying to build this framework one person trying to do this there’s everything still ill defined and again one person can do this or it can be built in a janky way that looks like it works a lot very work looks like it works pretty well but the problem is all these how does it get sustained how is it scalable how is it delivered and how is that transfer of

55:36 delivered and how is that transfer of knowledge going across the organization that’s where the problem to me or where that the roles are so important here unless you have the clear roles across who’s going to be building what in what capacity how much resources do we have available to them that that to me is where that to data literacy grows so that’s where I’m going to end my own Testament here but I oh go ahead Seth do you wantan to do something like I’m thinking we’re we’re

56:07 like I’m thinking we’re we’re about I’ll wrap I’ll wrap with my final thoughts here okay yes I think from a structured post or article fantastic yep he starts with this is like a road towards data literacy when I’m having these conversations and before like the final almost the second to the final paragraph is the key to this analogy is explaining the difficulty as well as uniqueness of each role with kindness and sincerity to your audience for example many people can construct a shed but it takes a different type of

56:38 shed but it takes a different type of expertise to construct a dorm and I think this all comes back towards it is because and I guess my final thoughts would be this too not all companies are going to expand at the same Pace a lot of our conversations are really acceleratory right some are will not want to expand or grow more so the level of as long as the level of effort Remains the Same for these individuals right and it’s not expanding into all these skills and they’re fine with a shed they could be fine with a shed and

57:10 shed they could be fine with a shed and that’s fine but if you’re going and moving towards a dorm right this is where articulating these things is extremely important so if you are the person the person right that is unbeknownst to you all of the sudden the center of fantastic things in the business the best thing you can do that I think this article points out is start to communicate the level of efforts and understand that there are varying expertise and skill sets that people are

57:42 expertise and skill sets that people are driving you into and they are going to take more time and you need to articulate that they’re going to take more time there are certain skills that or like methods by which somebody who needs to go like architect the building right otherwise we’re going to we’re going to find problems because you building the shed is one thing and I love that right me building the shed is fine start you want a really big shed because at some point right when at some

58:12 because at some point right when at some point when we have a hundred students in that giant shed it’s going to start falling apart and you’re gonna need somebody you’re going to pay more because it wasn’t built right right so over time I think in expanding Solutions these responsibilities should be different people and it like the big thing is if you’re the person get your head around these roles communicate that there are different levels of effort and as the the company expands or grows the best thing you can do is articulate that

58:44 best thing you can do is articulate that these are Big deals depending on how fast the business wants to accelerate into the data spaces so I I’ll Tom you got any final thoughts on that one that analogy is so good because when you think about the person building a dorm and really data literacy I’ve always thought on the consumer side but if I build a dorm or a shed well have you accounted for people how people are going to behave in the dorm are you g to account for all the possible mistakes that can happen the plumbing or just the

59:15 that can happen the plumbing or just the dumb people on a dorm but compared to building a shed while looks like while may look like the same framework I think too often I focus on data literacy on our business definitions on being able to read a report but man if we don’t have it on the back end on that flow of that information coming in and being able to structure it correctly do we really have data literacy from all ends so that’s really my last thought my last thought here is I like this article a

59:46 thought here is I like this article a lot I think a lot of this as I look at this as a leadership challenge as well we’re looking at this pretty from a data engineering role like the the Tactical the people are doing the work but I think actually this article speaks a lot more towards leaders who are trying to manage an organization and I can’t tell you the number of conversations I’ve had with people where I see individuals in companies at the worker level doing the work of like all these roles and so I have a lot of conversations with people who I I tell them specifically you have

60:18 who I I tell them specifically you have outgrown the data culture at that company more than they are allowed to let you do and your understanding of the powerbi platform and you actually need two things either one you have to go find a company that has a data culture that aligns with your vision of how powerbi is working which means they do respect more of these roles they do respect having Architects and engineers and separation they understand that concept or you or you have to wait for the leadership to change their mind or roll over those are

60:50 change their mind or roll over those are the two things that happen you move on or leadership changes their mind or or the leadership changes and then you get this new fresh wind of like okay let’s actually hire the right people to do this because I can’t tell you the number of conversations I’ve had with people who have outgrown their organization because the tooling allows this one so sometimes it’s not a good culture fit and so I’ll just say it that way I think this is a really good article but I think this article speaks mostly to leadership to make sure that you’re understanding this and as a leadership you’re continuing to learn

61:20 leadership you’re continuing to learn about the tools the tech and you understand that it’s not going to be just one person you’re not and don’t please don’t go out and hire a person that says I can’t tell you the number of rumes I’ve seen that have all these things on the resume we want an architect we want an engineer I need you to do data science I need all these things on a single resume you’re going to burn that person out and that’s probably not what your organization really needs your organization needs probably multiple roles instead of just that one big role and I think this is

61:51 that one big role and I think this is this is something that’s involving evolving right now we’re going to see this changing over time all right very great conversation wow Lively today really good thoughts I think this is probably one of our better episodes I I really enjoyed this conversation so thank you very much Tommy and Seth for bringing up really good points and I have Opus notes here on this one so this is great with that we really appreciate you listening to this podcast getting up early or in the middle of your day depending where you are in the world we thank you very much for listening to this and hopefully you found some value out of us going through

62:21 found some value out of us going through just data roles and data culture inside your organization hopefully this helps you have conversations with your leadership or if you are leadership hopefully you’re thinking about these things go read the article think about how this works in your organization our only ask is if you like this content or information please share it with somebody else we’d love you to share it on LinkedIn or I guess Twitter SLX is now x. com is the formerly n new to new platform I can’t really know what to call it anymore these days but please share it if you can that’ be really helpful we we don’t promote the

62:51 really helpful we we don’t promote the podcast it’s just out there so we we only grow by word of mouth so if you could do that for us that would be a great help Tommy where else can you find the podcast I’m G to Tom Cruise the heck out of this if you feel the need the Need for Speed you can find us on Apple and Spotify or wherever get your hot podcast help me help you make sure to subscribe and leave a rating it helps us out a ton if you can handle the truth or have a question idea or a topic do you want us to talk about that didn’t work dang it future episode however powerbi tip podcast leave your name and

63:21 powerbi tip podcast leave your name and a great question and finally that actually the only quote I got join us live every Tuesday Thursday a. m. Central join the and find us on Twitter LinkedIn YouTube I thought you’re gonna throw some weird scient end there good awesome thank you very much we appreciate your time we thank everyone for listening to the podcast we’ll see you next

Thank You

Thanks for listening to the Explicit Measures Podcast. If you’d like to support the show, share this episode with a teammate, subscribe, and send in a topic you’d like the crew to tackle next.