Will Fabric Overwhelm the Business – Ep. 253

Fabric can feel like a paradox: it’s marketed as one analytics platform, but the menu suddenly includes data engineering, data science, pipelines, and lakes. That extra surface area is exciting for IT—and intimidating for the business.

In Ep. 253, Mike, Tommy, and Seth answer a great mailbag question: will Fabric slow adoption because it sounds too complex? The answer is “it depends”—and they walk through how to frame Fabric as an on-ramp for Power BI users while keeping the heavy engineering capabilities governed.

News & Announcements

• Notifying Outlook and Teams channel group from a Microsoft Fabric pipeline — Fabric Pipelines can now send built-in success/failure notifications to Outlook and Teams, which makes real production workflows easier to monitor without custom glue code.
• Tips+ Theme Generator — A consistent theme is a low-effort governance win: less design churn, fewer stakeholder surprises, and a more professional “suite” of reports.
• Subscribe to the podcast — New episodes, show links, and the full Explicit Measures archive.
• Suggest a podcast topic — Drop a mailbag question or topic idea for a future episode.

Main Discussion

The core question is adoption psychology: does Fabric make Power BI easier to adopt—or does it make the story so “big” that analysts and leaders decide to opt out and stick with Excel?

This conversation is less about features and more about rollout strategy—how you message Fabric, what you standardize, and what you intentionally keep out of the business user’s day-to-day experience.

Key takeaways you can apply:

• Frame Fabric as Power BI plus options: most users can stay in familiar tools; Fabric expands what’s possible for the teams who need more.
• Lead with outcomes, not nouns: talk about faster delivery, fewer copies, and reuse; avoid drowning people in “pipelines/lakes/engineering” vocabulary up front.
• Define a safe self-service lane: approved workspaces, starter templates, certified datasets, and a clear “when to escalate” path.
• Separate building from operating: experimentation is fine; production needs ownership, monitoring, and cost checks.
• Publish repeatable patterns: 2–3 reference architectures inside your org beats 20 one-off experiments.
• Use guardrails instead of bans: roles, naming conventions, and visibility reduce risk without killing curiosity.
• Onboard like a product: a short “Fabric for Power BI users” intro beats a deep architecture lecture for first-time adoption.

Looking Forward

Roll Fabric out like a product: pick one governed, high-value scenario, document the pattern, and expand from there—so business users gain confidence instead of feeling like they just signed up for a new job.

Episode Transcript

0:01 [Music] foreign and welcome back to the explicit measures podcast with Tommy Seth and

0:32 measures podcast with Tommy Seth and myself Mike that is me hi I’m myself whoa it’s gonna be a little interesting episode today all right anyways let’s kick some things off with some intros here we have some interesting developments across the ecos ecos I have an intro Happy Tuesday Mike man here we go slow slow down slow your roll calm down take a breath all right breath taking a breath of the week all right ease into it I feel like I’ve already run two days through just Monday

1:03 already run two days through just Monday so it’s it’s Tuesday I feel like I’ve already been working pretty hard already oh all right anyways something we found across the interwebs was the fabric log has announced that there is some additional actions you can use from pipelines this is something I thought was missing from Azure data Factory and or synapse was the ability to easily at the end of your activities send an email like hey we finished something hey this thing failed I think that was very helpful so they have

1:33 think that was very helpful so they have ability to send a message to A team’s Channel and they have the ability to send a message via an email via Outlook which I think is great I’m very excited about that one it’s gonna help people out from a notification standpoint post no notify and notify yep I’ll throw the blog post here in the chat window for anyone who wants to try it out or use it that should be there for you as well well and Tommy do you have any other openers any any fun anecdotal stories I

2:06 openers any any fun anecdotal stories I do actually Mike so I realized that I do a lot of training now but almost on a daily basis but I’ve realized my ability to train comes directly directly my mother asked me last week can you help me with Excel I set out dude pro bono yep because she’s such a loyal listener oh yeah oh yeah she’s definitely the number one fan good morning good morning

2:36 one fan good morning good morning Tommy’s mom if you’re listening good morning Joanne but man I I you when you realize when you deal with someone who just gotta be careful here didn’t really need a computer growing up yes correct and it’s a great point so just walk through because she’s my father and her when I remember when the iPhone came out you had to plug into iTunes iTunes probably the worst experience I’ve ever had in my life or up there he’s like

3:07 had in my life or up there he’s like okay what we’re gonna do we’re gonna get the phone you’re gonna plug into the computer why your War now open iTunes but that’s for music yes I don’t want music you’re not gonna get music yeah and they’re like I don’t know why they would open it why do they make an iTunes like that’s just what they did absolutely are you an only child Tommy I no no but you you then were labeled the de facto tech support oh yeah for your parents yeah oh yeah I dodged this bullet my youngest my youngest brother what’s that so like you going through

3:39 what’s that so like you going through this is very reminiscent with some of the things that he’s the trauma there’s trauma there yeah so but that that if I can teach her like to use an iPhone or Excel I can teach anyone anything that’s what I’ve realized because it’s the patience of it’s not just like click on the cell like why is it a cell that doesn’t matter that’s what they call it and that you have to deal with that best mother in the world computers now double best mother of the

4:09 computers now double best mother of the world she now knows how to do excel right she’s now doing video yeah we know yeah yeah well she she asked like do Excel I’m like yeah yeah yeah I know Excel and then I always want to ask you do know what I do to a degree right but we don’t we don’t I’m still like I. T if she I work with computers that’s hilarious okay what I want you to do now is click on file

4:40 on file now I’d like you to make an entire database in my Excel file aware file oh no it’s like where’s Excel it’s like yeah higher left higher higher left go go nope a little bit more left yep oh file yeah okay okay now scroll down or like actions this is something I go through like when I when I do classes as well people who are yeah it’s shocking to me what when I say the ribbon people don’t know what a ribbon is in any of the

5:10 know what a ribbon is in any of the Microsoft application like there’s this like there’s some very common terminology so like I just have to make sure I have very clear like this is how I will describe where to look because I have to give people indicators to like there’s not a visual indicator I can’t tell you exactly where to go it’s an environment we’re like okay go to the ribbon read the sections so it’s the ribbon named home the section named copy and paste and then the button is found in like a group of six like and you have to do all of this without yelling correct so in baseball there’s

5:42 correct so in baseball there’s a there’s a it’s called a donut and you put it on your bat the donut basically it makes it heavier to swing the bat so when you go up to the plate you take the donut off it’s a lot lighter to swing I think anyone who trains to teach their parents how to use power bi and if they could get through an hour of that they’re building up patience with people it’s like it’s like going to the gym yeah now the difference here is a really good analogy there’s a slight difference here between secreting your parents to do things versus working with customers or training a class because and sometimes

6:14 training a class because and sometimes I’m just like making money I’m making money I’m making money I’m making money I don’t care how long it takes I’m making money oh no when when you train your parents you go I would I wouldn’t even get paid for things they fed me when I was younger you’re thinking happy thoughts yeah I was able to grow up like let’s see if we got a live feed just in case so far so good so we’ll see I’m sure I’ll be getting a call yes ungrateful

6:45 I’ll be getting a call yes ungrateful little it’s good though it’s good and it’s good that you’re you’re teaching people telling you the ways of excel that is that is the gateway to All Things data well at least she’s aware so at the very end and I’ll leave it at this she’s like are you gonna tell Mike to Carlo about this yeah I’m absolutely going to tell them about this in the world and not only that Mom we’re going to talk about it podcast for all five people everywhere in front of you every everybody that’s awesome train your parents become

7:16 that’s awesome train your parents become a better teacher yes I agree with that that’s wonderful well today let’s talk about our main topic today I think we haven’t this is a mailbag today so I believe we’ll need our mailbag voice to kind we’ll need our mailbag voice to come in and Pitch it so this is a of come in and Pitch it so this is a really interesting topic and I think this is a very relevant topic especially in lieu of people are trying to Grapple with what is fabric how is this going to change our landscape and I’ve been very vocal about the position of fabric and where Microsoft I think sees fabric

7:47 and where Microsoft I think sees fabric sitting and where I see fabric sitting I think fabric has got a is a great Advantage for the business user I don’t think it’s maybe necessarily the right tool yet for data engineering and data science yet we’ll see how they mature the product but I think it has potential to bring a lot of really interesting workflows into Power bi yeah the oh go ahead just real quick before we get into the actual question yeah I found some awesome mailbag questions from people you guys have been submitting awesome leave your name we want to give you guys credit for who you are and these these

8:17 credit for who you are and these these awesome questions so keep submitting them but please leave your name and who you are where you work well we love what you guys are doing so far yeah we appreciate it thank you yeah very much for the questions and we do we do regularly review the questions that come through through the mailbags so if you want to submit your own mailbag if you want to go find that you can go do so at Power bi tips podcast and that will take you right to the page where there’s a forum you can fill that out and you can enter your question there so we do regular review that as well Seth take it away read our mysterious

8:48 Seth take it away read our mysterious if you are a current live listener and you’re the owner of this mailbag question feel free to shout out in the chat we thought it was a good one though so here it is when Microsoft added power bi desktop to office 365. Dynamics 365 something 365. it was seen as something that will really increase adoption as more business analysts are exposed to it but I wonder if fabric which isn’t I as an I. T professional I’m

9:18 which isn’t I as an I. T professional I’m really excited about will have the opposite effect will business analysts and leaders find it too intimidating once they start to hear about data engineering data science pipelines data Lakes Etc and willit departments pull back on it out of fear that those new tools will be used inappropriately you’ve talked about it from the I. T Department side some but I’m really curious what do you think about overwhelming the business side I can see a lot of no thanks I’ll stick to excel

9:48 a lot of no thanks I’ll stick to excel thinking for those newly introduced to power bi slash fabric Legacy business users maybe not so much I think this is a great question I think you should have added in there no thanks I’ll stick this I’ll stick to a cell said Tommy’s mom the thought was there I just didn’t want to get a checked on the actual name question that’s a good question so I think I think this is a superb question

10:14 think I think this is a superb question and I think this definitely comes like I don’t I’m not sure the the it organization quite understands what fabric is yet so I think there’s there’s some things that are going to slide under the radar here fairly easily here and I I also think well we’ll get into it more here but I’m not sure how the power bi premium and fabric and the fabric f skews are going to intermingle because my understanding is if you get to a p

10:45 understanding is if you get to a p something skew you already have some fabric capabilities there so initial reactions gentlemen what do you think like where where do you want to take the conversation here well I would back up to what you said where IIT doesn’t understand fabric yet Mike I think a lot of I2 organizations does not understand the integration of power bi at an organization yet or the full impact okay where when you think about and from like think of managed self-service on the control a lot of organizations still have or want to still have my and also

11:16 have or want to still have my and also the the I think the backlash of Microsoft constantly releasing these features like the new SharePoint now available in power bi and you have to manually disable that if it’s on top of that that’s going to be something someone asked them they’re gonna go man all these features that we don’t know are available we’re just going to shut things off and not understand the full the full feature set the full scale with power bi can do a lot of organizations a lot of it could organize departments would rather start with tighter control so this is and that’s just power bi

11:48 so this is and that’s just power bi that’s not even introducing any of the elements of fabric yet so we’re I think we’re starting from there in in my mind I don’t think there’s some challenges with that I think still people a lot a lot of people still feel power bi is just just a visualization tool yeah and I think that’s a I think that’s a major Miss on a lot of people I think that’s oh it’s it’s just like Tableau or it’s just like click it’s just a visualization tool provided wide tables and it’ll do things and it’s actually much more than that it can get

12:18 actually much more than that it can get much more complex it can be much more integrated into your business processes and I’m I’m definitely seeing the need for for individuals that can span very technical spaces modeling data engineering as well as communicating with the business and asking them what do you need to see what is actionable in this report and that’s that’s becoming more of a skill so we’re really we’re blurring the line I think a lot more between what the business user can do and what was traditionally done by it by it your thoughts Seth yeah I guess

12:50 your thoughts Seth yeah I guess your statement made it almost sound like the business would would be and is leading the charge and saying like yeah yeah turn on fabric give me fabric which which I I would think the exact opposite right because the it’s an entire Tech stack right and it’s still ultimately controlled by I. T in terms of enabling user access in the entire preview where they’re not people would even see it etc etc so I don’t know it like I get more

13:22 so I don’t know it like I get more questions of why do I have to go through this dialogue to sign up for fabric when all I want to do is look at my power bi report that’s true I can see that yeah that’s what I’m getting more from the business than oh my gosh fabric like enable it all for me so I can go do data science work so I I don’t know like how have you been engaging with your customers do you have any of them talking about like enabling fabric yet so which which part of the company okay

13:53 so which which part of the company okay this is great questions so I think it’s going to be a mixed bag here I think it really depends on your organization I think in some organizations where it’s more challenging around getting access to data and producing things quickly with value like for example the data warehouse stuff or the it group that’s slow I think the business is actually clamoring a little bit more for it to relinquish some more control around some of the data elements hey we just need copies of data you’ve got these really large tables there are hundreds of millions of rows in size

14:24 are hundreds of millions of rows in size we now have tools that can help us and so the business is looking to explore can fabric speed up our ability to go get information and shape it into what we want like what we need for the reporting I think there’s that side of things from from that perspective there’s also the the idea that some organizations are taking a very cautious approach they want to know about it but they want they don’t really understand why would I want to use it where does it fit and I think this whole Paradigm Shift around a lake house

14:55 Shift around a lake house and this Medallion architecture is fairly new and I feel like a lot of companies that I work with are very SQL based in nature so they’re used to staging tables they’re used to servers sewing up they’re used to like turning things on for a long period of time like this is going to cost me a lot of money not unless you design it wrong so there’s there’s definitely to me there’s a lot of Education from what is a lake house how do you better use it use it I’m going through some testing things right now with the fabric lake house right now I was I was doing a streaming data test for like two weeks with some

15:26 data test for like two weeks with some streaming data and very quickly my Lake filled up with around 25 gigs 26 gigs in my fabric test trial and so I was trying to vacuum and optimize the lake house tables these are these are things that you would do if you were doing data engineering things and as I was doing these I’m getting weird results from the fabric environment it’s different than what I would have seen when I was working inside data bricks so there’s there’s definitely some refinements that still need to come I don’t think the vacuum process was

15:56 don’t think the vacuum process was working correctly it wasn’t removing files as expected in my my lake house is getting increasingly in size where where one of the major pickups I find when people work with lake houses is they don’t understand there is a side to the lake house that you need to manage the storage yes storage is very cheap but if you’re not on top of it and removing old files or vacuuming things out these tables can get really large and you can make lots of files that aren’t really helpful and actually adding no value after a period of time so you’re going to be spending extra money on data you’re not

16:26 spending extra money on data you’re not even using Mike I I go back and I’ve been going back to something you said really early on when fabric was introduced where we have the I. T department or the data engineering teams they’re not asking why fabric they’re more asking well why switch the point is nothing new is in fabric all these other features and tools already exist just in those areas yes and rather than people asking well why should I go to fabric it’s more well why should I switch I already have

16:56 why should I switch I already have synapse and I have the control and I have the teams and I have I want things in a sense separated right now because I have that greater control so I T and at least the data engineering teams the data the Big Data teams are going we already have a great setup you have to prove to me this is a better experience and a better workflow than what we currently have and then on the other side the business analyst is going what in the world is all of this this new world what’s a pipeline what’s a lake house I now I’m overwhelmed so that’s

17:28 house I now I’m overwhelmed so that’s the Two Worlds right now and I’m probably right now because this is all very overwhelming to people because you’re getting all these new tools I just did a whole kind tools I just did a whole training session around fabric of training session around fabric recently for the class that I run a power bi deployment and governance and in that class we’re going through like here’s all the things you can do in front here’s the here’s the architecture of power bi oh by the way here’s an equivalent architecture of now Fabric and there’s this whole it doesn’t look like a lot but you now get

18:00 doesn’t look like a lot but you now get not just power query M as a compute engine to go low data you now have spark Cousteau SQL Server SQL serverless and you have all these other engines that are all living together there’s like four or five now compute engines that can now manipulate the data round so as a so as a data engineer slash data scientist coming into the business realm I think the business is going to get a lot more capability but I think you’re you’re right Tommy a lot of this is well I just want my table of data I’m used to doing

18:31 want my table of data I’m used to doing that and so we’re so the the culture shift is now even further away from where the business is used to working and so there’s even a larger gap between what that’s what’s that’s looking like for the business oh well what’s interesting about this conversation is it and ever since fabric came out is [Music] [Music] proponents of fabric or people who use it would be the ones that are promoting it right like business people by and large have no idea that fabric even exists outside of the fact that

19:01 even exists outside of the fact that they can’t get to power bi or I’m sure now it changed I I guess what’s interesting to me about like the the two areas in terms of the the setup here in it is I’m I’m pretty certain that like all of us instantly go hey from an I. T perspective how can I roll my existing solution onto fabric right and that’s not what Microsoft says like there there is no expectation right I think last term they’d love that

19:32 right I think last term they’d love that but like it’s about new implementations does our resolution did we simplify our solution enough and reduce the friction and the the trouble because if you think about it if you have any of these sort about it if you have any of these Big Data Systems or Cloud systems of Big Data Systems or Cloud systems which is typically what we’re talking about a lot there’s there’s a lot of setup and configuration there’s a ton of permission and there’s there’s a ton of things on the back end that completely fall into the IT Realm and continue to

20:02 fall into the IT Realm and continue to whereas we were talking about oh well how do you how do you graduate like elevate your power bi reports how do you certify them how do you get the data sets to a point you want to well I want to Source from all these big Enterprise back-end systems does fabric and what it’s promises to provide simplify all that if you don’t have it it’s a resounding yes right like they’re

20:24 resounding yes right like they’re simplifying that whole thing to the point that we’re actually having a conversation around business users integrating with these Solutions which never would have been possible before that agreed which is great so paper so there I and I agree and that’s what we’re going to dive into today but like related to the topic of conversation that we’re talking about it’s like I think that’s a distinction I want to make from new versus existing and roll on yes and then to Mike to Mike’s Point business the value behind fabric would

20:55 business the value behind fabric would be like there’s there would be a level of data access and transparency to the processes that they wouldn’t they don’t have visibility in 200 yes right like not and this is where I think just because you put things side by side are they going to understand them I yeah I think I think that’s part of the friction that maybe spiked my interest with this question because we do talk about it from the I. T

21:25 because we do talk about it from the I. T side a lot yes and when I think about overwhelming business users we are bringing not just a reporting tool right we’re bringing a tech stack and new technologies to brand new business people who need to adopt them so what are the challenges that we face on that side of thing and that that’s how I took this question because that’s a whole user adoption problem within an organization yeah I think it’s

21:55 within an organization yeah I think it’s important too and not to talk us up a bit but you have to understand our context we’re the the unicorns of the space because we know the business we hold the it we live this we love this and we live this but that’s not every organization where they’re not going to say right off the bat oh this all these Integrations of new tooling or or migration of the systems sounds great there’s a lot of reservation where who are you expecting to really take this on if it’s the business analyst to take this on well

22:25 business analyst to take this on well you have a few barriers of Entry either the skill level of pipelines notebooks data lakes and understanding the best stand or the best practices the it controls whether or not they have the access and then now they’re devoting they’re gonna have to shift their current time the most valuable resource they have time to not working on power bi data sets but now moving to the lake house but again going back yeah yeah it’s the time thing there it’s the time

22:55 it’s the time thing there it’s the time thing yeah like an any new any new technology right like the you throw onto a business user exponentially you can assume that things are going to slow down while they’re adapting it right and yeah Power bi was one thing but now we’re talking lake houses and Spark and like seven years it’s a lot it’s a lot there’s some pretty hot lofty expectations in there but I think this is why if I if I

23:27 but I think this is why if I if I had to put the the Microsoft hat or lens onto things right as well here I really do think Microsoft is trying to say look we want to bring your data engineers and this is where I think things get the water gets very muddy right they really are shy saying we would like you to hire data Engineers or bring in data Engineers to work inside fabric they want the workloads to work side by side now here’s where my governance hat kind now here’s where my governance hat comes on or how I try and think about of comes on or how I try and think about how do we govern this right up until this point we’ve been really focusing on

23:57 this point we’ve been really focusing on like with power bi it’s been around a lot around the data set the workspace and the report and maybe an app right how do how do I bring content to power bi we’ve been very vocal about making models and thin reports that’s the way to do it right so when you do that modeling or or that design I think people are now starting to get their heads around oh this is I see I’m getting I’m seeing the picture of what this is doing and I think what fabric has brought is fabric has brought another whole layer of one it’s Tech Stacks but you’re now bringing

24:28 it’s Tech Stacks but you’re now bringing a whole bunch of other information like tables so now we have tables of data databases as if they’re like lake houses so a lake house we have now these lake houses are collections of tables and or files and so we’re now adding like to your point Seth you said earlier right there’s now more infrastructure there’s now more access to the process the data the tables behind the data that are now could be brought in front of those business users so this is where my mind

24:59 business users so this is where my mind goes is do I really want all my lake houses and all my data engineering to live right alongside the data set workspace and the report workspace and so I think I’m having a harder time figuring out how do we build this do we have more shortcuts to assets and artifacts that the it organization already builds do we let them build do we let it or the engineering side of things build the data engineering elements and we just land links or relationships or tables of data into

25:29 relationships or tables of data into these lake houses that the business can use so I it’s to me it’s providing a lot more opportunity or more diverse teams to work together in the same environment now I would also agree with your comments earlier and I was very much with you a lot of it or did Engineers or data scientists are like we already have tools they already work for us today why would I want to migrate over to fabric for what really what’s the advantage right I already have data Factory I already have other systems that

26:01 already have other systems that produce value from data well I would go why would I want to go to fabric that has less features than what I already have right so to me there’s like a there’s a feature Gap right now for a lot of the data engineering and data science space that’s not enticing those people to show up and do those types of efforts that’s it that’s an interesting point Mike because I’m thinking about this in the question like overwhelming the business side I don’t know if so much the technology itself like learning what a lake house is we’re learning the areas

26:33 a lake house is we’re learning the areas I think it’s because the gray area of there’s no road map there’s no best practices yet like that’s true we we didn’t we’re still in preview right it’s not even real yet we never came to a conclusion on how much data is in a lake house on our previous episode and there’s no success metrics how do we know if we do this we’ll be more successful than if we didn’t there’s none of that right now that we can really gauge that and or the business could so I don’t think it’s like learning what a lake house is it’s well how should we use a lake house in our

27:04 how should we use a lake house in our department or team we’ve had a hard time trying to answer that question definitively yes and I like your your point there Tommy right we I I want to keep leaning on that point that you made there the idea is traditionally we were managing data sets and reports that’s our main facility now well every data set is made up of a collection of tables so all the fabric is providing to us is another whole layer deeper of right here’s all the tables you want to manage or you may want to see so I think that’s really

27:34 want to see so I think that’s really good in the fact that there’s more you good in the fact that there’s more exposure to you see them easier know exposure to you see them easier it’s not as hidden but on the other hand I think you’re right Tommy I think the proliferation of potential artifacts things you can get access to things you might not have access to if someone doesn’t know what they’re doing with how to build lake houses workspaces and and there’s not a there’s not a plan and this is one of the main points I was talking about in my class was you need to at least think through what is your strategy around data whether you whether you really Embrace Fabric or not

28:04 whether you really Embrace Fabric or not it’s going to be here what do you want people to see how much of these tables we’re going to produce is accessible to the business it’s almost neutralizing micro power bi’s best selling point is that quick quick adoption or the five minutes to wow where I can build a data set may not be great but I can begin to visualize right away that was a rapid adoption and now we’re introducing all these things that almost disqualify that

28:34 is fabric for the business oh I don’t know if I can answer that right now that’s a great question Seth I’m gonna say the way in its current state I’m gonna say no honestly I think when you listen all the documentation you’re put your watch Microsoft right about it it’s talking all about the data engineer the data engineer the data engineer everything everything they write is fabric is around data engineering all the things they added into fabric aside from the power like like other

29:06 aside from the power like like other everything except direct like all of it is towards data engineers then then why would I allow a business user to have access to all those Enterprise systems that is a great question yeah and so this is this is what I was trying to allude to earlier right maybe there is a fabric workspace that you have people on your team that are prepping data right that’s the data engineering hey we’re gonna do notebooks hey we’re gonna do spark whatever that’s that’s a part of fabric

29:36 whatever that’s that’s a part of fabric that we don’t get access to the output of that though is something I would want to give the business users right so if I have a medallion architecture I bring in data I refine it I mature it right that’s something I would want to give to my data engineering team right and then the output of that I want to easily get that into Power bi and I think I think to me the value that I see from fabric for the business user is directly how amazing would it be just to be able to point a data set at tables that are automatically being populated

30:06 that are automatically being populated or hydrated by some team that’s to me I’m really excited around the direct lake side of things and feel like that’s been a really big advantage on top of this but you’re right Seth I don’t I think you have to really slowly progress through that growth because in the same fashion like you don’t want to just randomly give a power bios or access to a workspace do how to publish do what you’re building do how to build a data set are you going to build data sets that are super abusive and going to take all my capacity away like there’s training that’s involved to

30:37 there’s training that’s involved to getting even a business user into Power bi bi we would have to expect the same effort education training around letting other users get access to fabric inside power bi do you understand but are they right like the the challenges and some of the mixed messaging it could just be like as things roll out that I that I encounter is we there Microsoft is taking terms that we’ve been using for business users and business intelligence related

31:08 and business intelligence related activities for like power bi and Reporting where it’s an interface that is is is is for business people like you don’t have to code you can click through things and do a lot of transformations in power query you can connect to multiple different data sources and it makes sense and the the only reason your time investing in calculations or things to enhance that in the visual layers because you need an output but it still makes sense in that world what Microsoft is doing is it’s taking

31:38 what Microsoft is doing is it’s taking like find your champions but now they’re Champion like Champions aren’t business users in the marketing material Champions are are engineers and Architects and data scientists and it’s like ever we just exploded into the organization everybody’s a champion of of like if you’re going to create your Champion network of fabric it’s across all these spectrums and what’s interesting to me and like even in this conversation like and I know it should be on the business side is

32:08 is I I I don’t think we should we are like part of this conversation is bring the business analysts in to have them plug into the number two thing for the the training which is oh start teaching them lake houses start teaching them Apache spark to start teaching them Delta Lake like they don’t care so where’s the spectrum of like of course they’re in power bi the pop the reporting side do they Branch into a few other

32:40 do they Branch into a few other areas within a workspace that has access to a full Enterprise pipeline or to your point should you be creating a separate pipeline that then like so now I’m separate like why why not bring that to the table right away like and I haven’t seen anything where it’s like hey a great way to implement this or like break this out is all of your parts of Champions that deal with Enterprise whatever we would put them in here because they’re going to be very concerned who

33:11 they’re going to be very concerned who has access to things just like they do in this world and we’re curating data sets that will make the most sense for the business and then if you want to analyze then if you want to enhance then you like that you do that in a separate space or there’s some way to separate within the space the permission levels of people and maybe that’s going to get there there but I like right now there’s this disconnect of like man I don’t think what I’m saying is bring on a business user analyst champion and have any sort

33:43 user analyst champion and have any sort user analyst champion and have any expectation that they’re going to of expectation that they’re going to ramp up into all of these these areas of a tech stack because that’s because you bring them together doesn’t mean that they’re still not like very good functions large things that you need to understand in the current state animal caveat because what six months from now like how good is co-pilot going to be right like who knows are we unifying all this because it’s just going to be that simple for a business user to interact with all these things and not know but

34:13 with all these things and not know but that’s all also scary because like if somebody’s just somebody’s code and copilot and then dumping it in and don’t pick me like run and this one’s bonuses based on that yeah well I I end up but I I honestly push back I don’t think it’s the skill or lack thereof right now that’s the problem or even the overwhelming I I really go back to sorry for what hold on so what you’re saying in terms of where where does the business analyst exist in this or where does the it or where does the data

34:43 does the it or where does the data engineer exist in this you don’t think there’s a skill gap between trying to teach them those I know there is a skilled business I completely agree there’s already dead Engineers out there that are doing this right so you can bring them in but like the business user there’s a gap in their skill knowledge I think Tommy you’re saying those there’s other there’s already people out there who understand this like that they are Engineers will they be able to use fabric bring them in is you’re basically bringing in new people as opposed to educating the existing business user platform yes and

35:13 existing business user platform yes and no let me back up I’m saying the fundamental problem I think we’re having here and this question where are we going to overwhelm or is it going to control I think it goes to the utility of what the question that has not been answered yet what does the fabric leak house do better than any existing solution out there currently because if that can be definitively entered and proven you will see business analysts begin to adopt this and learn the skill you will see it organizations invest the resource and time for their ba and the

35:45 resource and time for their ba and the power B bi users to get the skills into this you will see the it beginning to integrate their existing Solutions into this sure but until that question where this whole core Story Goes of what does the fabric lake house do better than any existing solution now I don’t think you’re going to see either side you’re going to overwhelm because they don’t know how to use it not because they don’t have this they may not have the skill but they don’t know where it exists in their current workflow and it’s going to go wait so you’re telling me we’re going to duplicate our efforts

36:15 me we’re going to duplicate our efforts or I’m going to move all my efforts to the fabric lake house why and that no one’s going to invest the time or resources to either get past that fear or the overwhelm because there’s really no against true better utility for the fabric lake house well what what what’s the pitch right I well what what what’s the pitch right let me I want to answer a question mean let me I want to answer a question directly though Tommy like I want to jump into your interview so I don’t think fabric does there’s there’s a couple things a couple items that I say

36:47 couple things a couple items that I say fabric does better than your current systems today right so synapse is kind systems today right so synapse is like my point of reference and of like my point of reference and databricks is my other point of reference when I talk about those things right so one thing that fabric does better than databricks does is this whole concept of connecting power bi data sets to the lake the Delta tables that databricks provides so databricks does do a table but when you’re if you’re using pure data bricks and power bi you have to connect power bi through a databrick SQL endpoint yeah but you’re still refreshing things and they’re still in import modes

37:17 and they’re still in import modes there’s there’s some things there that are a bit janky now that being said databricks can now write tables down to fabric the lake house fabric can then then power bi data sets can then connect to those tables directly and now you’re using directly so one feature that I think that power bi fabric does better is this direct Lake and the ability to connect all these different Services together right so without a lot of extra configuration it’s painful to get synapse blob storage data bricks all those things

37:48 data bricks all those things talking together there’s service principles are set up there’s things in the background there’s permissioning issues who’s got the right permissions to the lake and lake and what lake elements are doing so I think that part is a frictional piece that fabric is solving so the the connection or the connectivity of the different Services together I think is a very big win and fabric has done a very good job of saying hey doesn’t matter what tool you’re going to use whether it’s Spark data flows

38:18 data flows or something else we’re going to write the data down in the Delta Farm app that that was a big win I think in in general and even the data sets right power bi data sets are going to read Delta format tables and use that columnar compression type format in order to let the the power bi engine absorb data all of those things I think are where fabric is very nice and I also would maybe even Echo here a little bit as the idea of using this SQL serverless engine that allows you to write SQL against tables

38:49 allows you to write SQL against tables that live in your Lake and explore data now granted that’s much more of an I. T data engineering type exercise I’m going to write SQL I’m not expecting every business use it or write SQL but I’ve walked into organizations where they have they have everyone in the entire company all business users knows SQL but this means we have hundreds if not thousands of access databases ripping out data from servers and doing their own little data engineering so in in in organizations where they have prioritized a workload around everyone

39:19 prioritized a workload around everyone must know SQL we will spend the money to build that into our data culture I think in some ways fabric would add a lot of value to those companies because then they could produce people could produce their own data engineered things inside this fabric ecosystem so I think that list will grow over time there will be more better things that fabric will produce to this but I think that’s that’s that is my short list right now and I’m not sure that short list is compelling enough to say okay stop what you’re

39:51 enough to say okay stop what you’re doing shut off all your old exercises that you’re doing around data engineering let’s move everything over to fabric and get it right there I don’t think there’s enough and one thing that’s more to me is to spend on this sorry go ahead the spend is absolutely something but I I think there’s less of a compelling reason for organizations that have already figured out all of the touch points between these systems systems to incorporate data into an ecosystem right like if we’ve already sorted it out and we already have all these tools and these things in place is there a

40:22 and these things in place is there a reason that you would abruptly move on to fabric and I think the answer is no because you’ve already served yes we if we if we look at just the marketing of fabric which which is valid to all Mike’s points it it unifies the architecture right that’s like already it’s it’s consolidating all these independent services that Microsoft still supports yes into one service one interface is it

40:49 yes into one service one interface is it from one to the next to the next yes is that going to get better yes I think I think more compelling you’re unifying the data storage that’s because correct is if we think about AI if we think about all of these things that allow me to plug something into an estate of where all my data lives right now that’s a it’s a huge challenge how many third-party systems do p does or do organizations use in business units right you would have to

41:19 business units right you would have to have all of those ingestion pipelines created to pull it into a lake house to do all your ETL like with it with one like like great throw it in one link now it’s accessible throw it in in whatever format even if it’s not part of my lake house even it’s not part of my fabric workspace and pipelines and all that stuff it’s still there and then I think ultimately like the biggest thing is just while I have security concerns around certain aspects like unified data transparency there’s a pipeline we all know the pipeline how

41:50 pipeline we all know the pipeline how many conversations do you have where somebody looks at a source Excel file and says I’m trying to compare this against the report and none of it’s making sense so we must look at the report calculation it’s like well yes that’s a component of it but do we know how the data is going from point A to point B and how many Transformations or how it’s joined to other information or cleaned or cleansed or all of the 500 steps that could be happening before you steps that could be happening before from the raw file before it even know from the raw file before it even gets to the report no right so I think

42:21 no right so I think those are those are very compelling and the more like if you think about even some of the the very easy methods within power bi to connect to third parties to hook up to an API to do all these things there’s a very compelling argument really in fabric was like okay if that’s just as easy but now I can pull it into a a more Enterprise ready like workflow instantly and I don’t have to worry about like security there’s

42:52 to worry about like security there’s a ton of use cases yes for new implementations because I would much rather do that than spool off an independent ADF service connect and make sure all of the stories work together and then I have 80 less Gen 2 and okay this is definitely an I. T service request ticket right so so speed to Market is it five minutes to wow but no but for some for folks who have the experience of like

43:22 folks who have the experience of like how do we set this up how do we use it I think it’s much faster and easier to onboard new business units I think it’s much faster to implement and overall I think it’s just going to be a a better and easier experience to get data into a centralized system do we have a ton to talk about as it relates to like how do we organize that how do we manage it how do we administrate it like absolutely there’s no that’s not figured out yet but it needs to be more thought around that but like if I’m an organization that doesn’t have any of

43:52 organization that doesn’t have any of this set up or I’m playing with Cloud but only certain aspects of this I I think it’s definitely worth the POC and look it’s just you need more people who understand a lot of various parts of of these systems as opposed to just a business user who may know some power bi we we may have been approaching Fabric and the the adoption of it maybe in in the wrong angle or possibly wrong in general I really from the beginning

44:22 in general I really from the beginning we’ve talked about this All or Nothing approach well if you need a lake house you need pipelines you’re going to have notebooks there’s going to be data flows all these all Medallion approaches is there a way though if we’re going to truly introduce fabric to a business without overwhelming overwhelming them we have a more controlled released kind we have a more controlled released process where rather than hey here’s of process where rather than hey here’s the entire data engineering stack all at once adopt it starting with segments of fabric starting with the lake house

44:52 fabric starting with the lake house that’s already pre-controlled showing the use case and success of that and then beginning to expand on that just like when you introduce power bi to an organization or to a user you don’t start with hey here’s filter context you start with hey here’s filter context what daxes doesn’t matter here’s know what daxes doesn’t matter here’s filter context that if you have a model you start with here’s some some is like Excel and then we can build up and I think as I’m going through this introducing fabric cannot will never work by All or Nothing yes where we

45:23 work by All or Nothing yes where we start with data engineering we have to start with segments and begin to say what does success success look like we’ll start with a lake house one lake house that someone else controls you have some shortcuts yeah look now we’re going to begin to utilize this in a pilot yes and then we expand yes and I think that’s the approach and think about right now where does the business user already understand portion of the fabric right right because you already understands data flows right so we already have data flows Gen 2 showing up in Fabric and so from that perspective you’re right Tommy I don’t want to drop in and give business users

45:54 want to drop in and give business users a python a notebook all these extra things now the the other alternative here is you may have again this is all culture driven right so I’ve walked into companies where everyone knew SQL everyone would be happy to have a SQL data warehouse inside Fabric and they could write queries against tables they would love that and so that’s their culture that’s that’s something that they invested in and they picked and they hired people specifically with that skill sets there’s likely other businesses that have a lot of python developers already running things at the

46:24 developers already running things at the business level not saying there’s any ones that I’ve encountered but I was actually looking to Stephen on the chat in LinkedIn says my guys are happy like we like using python their business users understand it so that’s a skill that their team has developed or they’ve hired into and it’s nice to know that fabric accommodates those users as well accommodates yeah and so I think where I think I’m I think they’re the majority people if you’re a business user you’ll feel comfortable in data flows okay where do I put the data okay

46:54 okay where do I put the data okay there’s a little there’s a little bit of a growth a mental shift that okay well when we’re done with the data flow we now don’t have to save every table in the data flow we can we can tell the data flow where to put the data oh we’re not going to put that in the lake great you can hand wave a lot and say oh by the way it’s using all this extra technology in the hood you just care about put your tables from the data flow in this lake house here’s your default data set boom done like so there is a very simple story of like a growth pattern here telling that I think

47:25 growth pattern here telling that I think you’re you’re doing a very Job Art very good job articulating is we’re not going to give a business user all the things don’t don’t feel like you need to come in and know how to use fabric you have to have The Medallion architecture like what’s bronze silver gold who cares exactly take this a step level up too right like one one one point before I dive into that is like I don’t think we’ve been on the wrong approach I think what we’ve talked about a lot is implementation and how to build focused Solutions within a new framework yes as well as well as how do we set up the infrastructure for this new thing

47:55 infrastructure for this new thing because anything we’re anything you’re going to do totally like we’ve been in that mode as opposed to talking about how do we implement it or get things going but if you’re if you’re talking on like a more granular but we bring that up what’s striking me this about this like buy segment idea is especially for new new new areas of adoption if you think about a business unit that is using one two three third-party tools a lot of times when they interact with Enterprise bi teams right like we have to oh yeah into

48:26 teams right like we have to oh yeah into them how do we connect how do we get data where do you where do you store that data especially if there’s a lot of high data security around it HR think Finance Etc the segment idea with fabric actually is much more compelling because if I’m a business intelligence team that is in charge of the whole organization now I can have a POS POC that’s specific to a particular business unit and I can build things in an ecosystem for them whether

48:56 things in an ecosystem for them whether or not I give them access to everything like I think remains to be seen how we set up a workspace Etc but at the same time like if I’m walking them through the process of fabric and this is the same interface that they begin to interact with and adopt Etc as far as like all of these things that are part of a data strategy what am I doing I’m actually elevating their data literacy and increasing the value of the business because we’re creating accurate pipelines in an area that I can leverage

49:27 pipelines in an area that I can leverage and reuse when I go on to the next business unit right because every part of that architecture and structure that we’re building for that business unit would be accessible to me if I’m if I need certain net data aspects in a different business unit and that’s where I think where this becomes really powerful is the shared shared tables right between different workspaces or organizational units that are allowing me to build extremely

49:59 are allowing me to build extremely accurate and well-formed data outputs that are owned by a team I don’t want to be the subject matter expert or owner of all Source data right but if I’ve now pulled in for that business unit that all the data that they need and consolidate it and we’ve gone through and applied the business logic and now I have clean data sets for them now I can leverage that in this ecosystem because it’s all part of the same ecosystem as opposed to oh we built that over in SQL for them well now we

50:30 that over in SQL for them well now we need SQL permissions can we go access that table and then what are we going to do we’re gonna we’re gonna move copy that data over to this system right like and this is where the lake house architecture makes a ton of sense because now I’m in Delta tables now it’s all in the same place now all it is is a matter of permissioning and access from one business unit to the other and you can do this in an incremental way and there’s there’s a powerful statement there too because Mike you were talking about the data flows where gen 1 dataflows are are the no Outlets of data

51:02 dataflows are are the no Outlets of data engineering yeah right where there’s only one place to go is it to a power bi data set that’s the only totally agreement now you can actually expand to the teams like hey use the same tooling you can but guess what we have in the lake house a SQL Server guess what you can plug into Power Platform now if they’re the Power Platform power automate you can do right back on your own server I think so now all of a sudden you’re expanding their tooling besides just a single power bi report because again thinking about the pushback well why would I do this and not just a power bi data sets like this

51:33 not just a power bi data sets like this data now exists outside of this power bi that you’re still controlling using the same tooling familiar and yes and I and I think one thing that I’m I’m thinking through my mind here is there’s there’s now still a story around one security that hasn’t really materialized yet so you hasn’t really materialized yet so right so there’s this concept to know right so there’s this concept to your points that’s right we’re gonna have multiple areas of business we can do proof of Concepts in these different departments I don’t have to worry about spinning up any extra architecture when

52:03 spinning up any extra architecture when you delete the workspace all the architecture disappears things get simply it simplifies a lot of things inside all of this so I have to pay homage thank you very much for Joanne jumping into the comments in the chat I really appreciate it you are doing great and and Tommy will get you there Tommy will will teach you about fabric it’ll it will come I was gonna say I got a live update on mine I told her I’m like listen to the pocket she said oh joy I can’t wait

52:38 so what I think is so going back to my comment around one security right so one security could solve a lot more of these problems and and Seth as you were saying those things around hey we can do business-led topical areas right hey your Finance your HR you’re doing specific things that are making data for your team this really does remove some blockers in this data Silo space and now we do have a common way of collecting and and communicating data across the teams in previous worlds we were using data sets which was row level security object level security and now

53:10 security object level security and now we’re expanding to the lake house so now we have tables and rows and tables inside those lake houses so all we’re doing is we’re adding more artifacts to that business user which is good but we’re going to need a thing like one security just says okay here’s all the data artifacts across everything inside your environment here’s how we’re going to easily manage access to all these different artifacts on a workspace level and I don’t I don’t see fabric adding a whole bunch of additional things away from like I don’t

53:41 additional things away from like I don’t see fabric adding a whole bunch of other elements away from what we do today right so fabric is not it’s not changing the controls and administration and governance very far from what we have today in power bi it’s still workspace related it’s still artifact driven you have access directly to data sets and now lake houses so there’s still the same levels of control we do add more things which I think is good but I don’t think we’re going to be so far away and so I think what what you’ll start

54:12 and so I think what what you’ll start seeing is I think you’re going to start seeing the adoption roadmap I think you’re going to start seeing other language coming from Microsoft explaining how do you manage the governance and finding those data stewards or data owners of the information we put into our fabric environment I think this also opens up another role to a fabric Champion that has to almost like a project manager for fabric to introduce these in the right in the right timing right like a controlled release where it’s like okay we’re gonna introduce the lake house but

54:42 we’re gonna introduce the lake house but we’re not going to talk about all the other elements let’s see what success look like now what guide to whatever that next step is so having that someone who has a skill from the data engineering just like we have that power bi champion who understands the business and the technical side I think we’re needing that’s going to be a need to have a leader in that space that’s going to guide that adoption yeah I would 100 agree with that I think ultimately Fabric’s a much more nuanced conversation and I think you need to really lean into like the obvious

55:13 really lean into like the obvious benefit for business units right and if if there is none because they’re working Solutions that’s going to be a harder thing to sell versus or or a more nuanced way to start to migrate to or integrate versus I think new implementations where it reduces a lot of friction across the board from both the it and business side and brings them closer faster to a better unified experience leveraging tools and technologies that have been

55:43 tools and technologies that have been out there before but much closer to the business right so I I think that’s the benefit of it if you just dump this on the business and assume that a business analyst where they would be able to pick up and see the meaning behind power bi for this disparate data sources and be able to put together reports if you just put fabric and you’re like okay here’s this new ecosystem yeah I don’t think you’d get much adoption but across the board I think it just it’s it increases the

56:13 just it’s it increases the level of effort related to business technology leaders managers within organizations Consultants Etc to really provide like deep deep training like this changes the road map for how you pull pull people forward or skill them up or whatnot but I like to our previous points I think I think it’s a fantastic platform that allows business users a lot more access to systems and things

56:43 lot more access to systems and things that they they may not have had before and that’s phenomenally or I should say that’s that’s an exceptional skill set that they could start to build related to Big Data especially if you’re in this number we’re talking about volumes and and these were inaccessible areas and still today are really hard to deal with just from a access and permissioning level Arena so overall those are my final thoughts I would say if I had to give my final

57:14 I would say if I had to give my final thoughts here around this because I know we’re getting that time here I would say I think a lot of business users are already doing data engineering to some degree and probably more so inside Excel than inside a formal system right so I definitely felt like I was when I was the Excel user and I was doing very complex things across Excel lifting data up making different vlookups and you’re doing a data warehousing experience a lot of people build many data Marts inside Excel I wouldn’t necessarily agree that’s the right way to do it however you’re you’re learning

57:45 to do it however you’re you’re learning the fundamentals of I have data in this weird unformatted way and I’ve got to shape it in a way that adds value to me right that’s that is a skill that they have learned they already know it the the challenge I think is identifying those individuals and saying okay we’re now going to give you better Pro Tools to do the same stuff so we’re going to take your data flows we’re going to make it dataflow’s gen 2. hey we’re going to take your tables of data you’re outputting and make them a lake house so it’s just translating what they were doing and so now where I where I

58:15 were doing and so now where I where I think the education needs to live is here’s a lot of great things that it typically does when they’re doing engineering we’re going to educate you on those things here’s what git is here’s how to build a lake house here’s how to store data here’s a bronze silver gold version it’s it’s adding more capability to people things people already know yeah and they’re already doing now doing now the organization I think has to stand behind all that but this is just an education level Beyond from where they were so that’s I think it’s I think it’s

58:45 were so that’s I think it’s I think it’s great I think we can give a lot of our business users credit I think we’ll get there it’s just going to take time for us to get our head around where this fits what skills are needed in that business team that are going to enable them to do more with their data tell me any final thoughts I think going through each segment of this it’s going through a phase identifying what you want to accomplish and showcasing the value at each stage rather than we’re going to choose fabric here’s all the value yeah it has to be we’re going to learn this incremental incremental value yes okay

59:17 incremental incremental value yes okay saying that to the users why I’m going to invest my time management why am I going to invest my resources on this and then that controlled release I really need to be the way now what that looks like what those components are Mike to your point that’s dependent on each organization exactly right dealing with what the skill level is at the time people agree but that way you’ve wasted a perfectly good hour of your time talking about fabric things fabric related things this is a great question so will the business be overwhelmed maybe maybe not right I think it depends on your organization and the threshold you have

59:48 organization and the threshold you have there I think there are some definitely strong points around fabric I definitely of this scenario I think the business users get the better end of the deal here I don’t think you’re getting with fabric you’re not gonna entice a lot of data engineers and data scientists to be like oh I can’t wait to use fabric I’m going to drop all these old tools and come on over so I don’t think I don’t think we’re there yet but it’s gonna it’s I think the story is getting better as we go with that we really appreciate your listenership we appreciate the time you spent with us we know your time is valuable if you like this conversation if you’d like thinking about how fabric

60:19 if you’d like thinking about how fabric Works in relation to the business user please share it with somebody else share it on social media let someone at work know you found some value from this we’re always looking to get the word out so for listeners thank you very much for participating we’d love it for you to share with somebody else Tommy where else can you find the podcast you can find us an apple Spotify or wherever you get your podcast make sure to subscribe and leave a rating it helps us out a ton have a question idea or topic that you want us to talk about in a future episode head over to powerbi. tips podcast and leave your name join us live

60:51 podcast and leave your name join us live every Tuesday and Thursday 7 30 a. m Central and join the conversation on all of power bi tips social media channels don’t just leave your name ask the question and then leave your name Yeah well yeah you can leave your name yeah we would appreciate the question as well as we earn it so I’ll add that to my templates awesome thank you all so much and we’ll see you next time

Thank You

Thanks for listening to the Explicit Measures Podcast. If you enjoyed this episode, share it with a coworker and drop a question or topic suggestion for a future show.