Everything Is Still BI – Ep. 329

In this episode of Explicit Measures, we discuss Everything Is Still BI and what it means for Power BI practitioners.

News & Announcements

• Power BI June 2024 Feature Summary | Microsoft Power BI Blog | Microsoft Power BI — Welcome to the June 2024 update. Here are a few, select highlights of the many we have for Power BI. You can now download large semantic models, Power BI enhanced report format (PBIR), and Subfolder support in Power…
• Everything is still BI - by Benn Stancil - benn.substack — What to expect when you’re not expecting to build a BI tool.
• PowerBI.tips Podcast — Subscribe and listen to the Explicit Measures podcast episodes and related content.
• Power BI Theme Generator — Create and download Power BI report themes using the PowerBI.tips theme generator.

Main Discussion

Using Ben Stancil’s “Everything is still BI” as a spark, the crew debates what Fabric changes—and what it doesn’t. The takeaway: new packaging and new features won’t save you from the fundamentals of BI delivery: clear requirements, strong models, and intentional user experience.

Key points:

• Fabric can change the platform, but the hard parts remain: defining metrics, modeling well, and shipping something users trust.
• They dig into Visual Calculations—useful, but currently limited by discoverability and reuse (patterns matter).
• The conversation calls out friction in the evolving desktop experience (Power BI Home, more clicks) and the need for a smoother “one place to edit” story.
• The new enhanced report format (PBIR/PBIP) is a big DevOps win (true diffs at page/visual level) but has real service limitations today.
• Net: embrace the tooling improvements, but don’t outsource BI success to shiny features.

Looking Forward

If you want to take action after this episode, audit one of your key reports for BI fundamentals first (definitions, model clarity, UX), then decide which new features are actually safe to adopt. The platform will keep changing—your standards shouldn’t.

Episode Transcript

0:33 good morning and welcome back to the explicit measures podcast with Tommy Seth and Seth and Mike good morning gentlemen and happy Tuesday good morning Tuesday again thank you and let me well congratulate you guys or not congratulated but Happy Father’s Day oh yeah belated belated little dad’s day little dad’s day yeah yeah yeah did you guys enjoy your daddy days I did I got a razor you relaxed I got some belts I got the most sad things ever I

1:05 belts I got the most sad things ever I got a tie got a tie yeah is that a nice way of your family just saying hey Mike no it was it was on my Amazon list I put it on there oh okay got it I don’t wear belts I wear you got it I don’t wear belts I wear what what they call those athletic know what what they call those athletic jogger pants what yeah that’s literally like my uniform they’re like other people talk about like oh I go to work and I wear slacks and stuff and I’m like nope I go to work I wear wear a hoodie and I wear Jers go go to work you just mean like go turn on your computer

1:35 just mean like go turn on your computer right like walk downstairs and turn on computers yes correct so I have a weird I have a weird job apparently yeah let’s talk about today’s topic today’s topic is our our main topic is everything is still bi an article by Ben is Ben suback stel no it’s St substack is the that’s the blog right Ben sub Ben sub it’s his website but the name is I don’t see the name at the top here his name is Ben stanel there we

2:06 top here his name is Ben stanel there we go thanks and it’s a heater it’ll be interesting to see this one so there’s going to be some good talking points on this one Tommy’s yeah I think Tommy you found this one actually right yeah okay good article to go through I will put that in the chat here when we get going but before we do that let’s do a Qui little run around the news here anything anything newsworthy that has come out recently hey we have the June 2024 powerbi feature summary update not too many things down the

2:39 update not too many things down the down the docket in terms of major features I would say probably the biggest ones are you can now download large semantic models there’s some visual calculations updates but other than that not a ton of major desktop features yeah what what do you guys think about the buzz around visual calculations not hearing much anymore no there a good point I’m I’m here like I’m seeing some articles I gauge a lot of how a product’s doing or

3:11 gauge a lot of how a product’s doing or one of the features are doing based on how many articles I’m seeing in my RSS feeds and that’s usually a pretty good gauge for me how many people are writing about certain features and right now I’m I saw a ton come out with visual calculations y not seeing a lot of oh did you could do this case and here’s another scenario and here’s another scenario not seeing a ton no I agree with you on that one I was I I I I think I think it’s an interesting feature I definitely think it’s useful

3:41 feature I definitely think it’s useful for when people need to to leverage it inside visuals the fact that it’s kind inside visuals the fact that it’s linked to a single Visual and it’s of linked to a single Visual and it’s not a library of visual calculations you can bring to a model like I really do think there needs to be some mechanism just like you’d have a theme file like hey here’s a whole bunch of calculations you can use you got to make it easy for people to get these things written and built and there’s only so many examples they have built in that to extend the feature and I don’t think they did that very well and honestly I

4:11 they did that very well and honestly I don’t think there’s a lot of legs on this feature if they don’t really kind this feature if they don’t really open it up to the community and make of open it up to the community and make it easier for people to go get patterns of visual calculations and bring them into into visuals well it’s not just that it’s if you’re going to put this in your report or put that in a visual you got to be able to speak to it both to your consumers and you’re like oh no that Delta here is just there we can’t add it anywhere else or you’re going to have to recreate a per visual which could lead it to inconsistencies

4:41 which could lead it to inconsistencies that’s a lot of risk I think a a professional is going to be putting on their report agree yeah I get it I no I get your points but at the same time like who’s it aimed at first time business users is it Dax like the L it’s not I I think I’m push back right off the bat here maybe it’s the extra coffee I’ve had today because it’s still the Dax language and if I would argue all the functions there are

5:14 would argue all the functions there are not more complicated than calculate but they’re on the same level in terms of what you have to do in order to create the those measures it’s not a simple sum or divide that’s my point though my point is like this is where the patterns of like this is where I really wish it would do better job at explaining or or more assistance required when you’re building those visuals in the visual calculations yeah it’s fine you can you can do that but you need I need it to be like all those examples

5:45 need it to be like all those examples they give you I need to be a lot more examples or I need to have a whole large list of examples because that’s how I learn I learn from other people showing me patterns that I can follow and I think that’s that’s when I I want to extend it so I’m just I’m just miss miffed on the idea that like okay fine you fix some things around it you you fix some things around it okay but I really want it to have know okay but I really want it to have more or ability to add your own functions and have a a library them and again I think this is the also the big miss that was with the calc the quick

6:15 miss that was with the calc the quick calculations or quick measures quick measures also should have been driven by the community let the community Define those things that would have been way more impactful than just here’s a list of 10 15 that’s all you get you can’t add anymore yeah I’m curious what do you think of powerbi home and desktop now that it’s technically in our June or not technically but in the June update it’s generally available have you found that a better experience for you having powerbi home in desktop have you seen any benefits to it

6:47 desktop have you seen any benefits to it you’re talking about the ability to like open files from surface that new page you have to click through to create a report yeah that exra page I’m just making sure just making sure the same thing there yeah yeah not loving it I don’t use it really it’s I always click like new report or open an existing one that I’ve already got going on don’t really love that new the I do think that yeah the save screen just saving something as a file to my machine

7:18 saving something as a file to my machine is painful now they added more clicks there and it’s just I just want to save the file as a file on my desktop and it’s like it’s like three clicks now you have to hit click save and then you have to do do this other thing and then you do a save as and then you have to go click another button to find the folders and it doesn’t it’s it’s try it’s too aggressive to try and shove everything into like recent and one drive well it seems like they’re trying to have that Synergy between hey we understand that if you’re going to be saving something in desktop there’s some people

7:49 in desktop there’s some people who are doing git and there’s this integration if you can also edit your model in the service so understanding if you’re saving it locally you may be already out of sync with everything the service so I think that’s what they’re trying to do with a lot of that integration I’ve I’ve clicked on links on the open page is like we’re going to bring you to the service I’m like no that’s I’m in desktop because I want to edit something in desktop I think that’s what they’re trying to do because again there are three areas right now I can modify or report three functions I can do it I can do it in

8:20 functions I can do it I can do it in desktop normally I can do it in the model in the web or I can do it in git and each of those are different experiences and again couldn’t literally lead to some out of sync so I see where they’re going with home and I see what they’re going to have all those areas but right now if I don’t have one place to sync then it makes it harder no matter matter what I’d agree with that one yeah I just not I’m not a fan of that part give me less clicks not more

8:52 that part give me less clicks not more and I feel like a lot of these things recently have just been more clicks to get to what I want to do anyways moving on for that one one thing I do want to point out another note here it’s not actually in the this is surprising to me this this maybe it’s down here at the PBI level maybe it’s in there just very briefly the developers the new parbi enhanced format the ppir format has now been announced announced with an improvement and a major Improvement actually at the report side the report side no longer is

9:23 report side the report side no longer is being built as a complex string saved object it’s all Json and everything is Broken Out by page so you can go all the way down to a page and a visual level to identify all the objects inside the report which this is the Holy Grail of like man this was this has been waited for forever everyone’s always wanted this feature pretty much and people since people were ripping apart PBX files in the very beginning it’s fully here now you have the ability to go all the way down to the visual type and

9:54 all the way down to the visual type and it’s great I think it’s really good it gives you diffs all the things you’d ever want with making your IND idual reports so very pleased about that that feature that came out I will admit I was playing around with it it’s very very good you definitely need to read the blog article and the full documentation on the PBR format that was announced there are some very strong limitations when you’re using it so there is some in the PBI format they’ve enhanced it but there’s a lot of things

10:24 enhanced it but there’s a lot of things it can’t do inside the service so if you are going to use that the service today it’s not currently the service is not currently up to up to par with this new PBR format like you can’t include it in a powerbi app you can’t downlo PBI what was that you you can’t use it in a powerbi app if you take the PB format and push it up like so if you go from desktop and publish it out I think it packages the file and then puts it up in the service so it’s still a pbix but if

10:54 the service so it’s still a pbix but if you’re using the PB format from like git how you’re using the files yeah you can’t do that you can’t download you can’t download the file as a PBX you can’t export to PBI the PowerPoint or PowerPoint or PDF it can’t be included in a subscription can’t be deployed with deployment pipelines there’s a lot of things that are restricting the PBI format right well I think they’re G they’re going to close those gaps you can’t have a PBI format it’s in preview so they’re just saying hey it’s out early the format’s

11:24 hey it’s out early the format’s basically done there’s a lot of things that still don’t work yet so then is it done not be able to sorry but this is our whole thing we talked about like releasing features rapidly but yes with some of those limitations like the if I if I’m going through this we used to deal with there I remember there’s one at desktop it’s like hey you can enable this new preview however you can’t publish your report once you enable this feature I forgot what preview that was back in the desktop days yes and yeah that’s

11:57 desktop days yes and yeah that’s that’s that’s good to know that’s to know that that’s something to be aware of one other thing here that’s also interesting here for just for other Prosperity there seems to be some hard limits inside the service as well as far as when you bake a powerbi file now I don’t know if anyone’s actually ever done this but it says the PBR PB has some normal size limitations for example your PB can only have 1, 000 Pages Max

12:27 your PB can only have 1, 000 Pages Max per report okay could you imagine a 10, 00 page report I’m G have to edit some of mine I’m to knock off a couple Pages here it’s 300 Max visuals per page a 5 megabyte Max for each bookmark file so we can only have five megabytes of bookmarks my God that’s a lot that’s a lot of bookmarks if you’re just doing like Json to write the bookmarks out there one one megabyte Max for each file I’m not sure exactly what that means I guess maybe each file in the

12:58 means I guess maybe each file in the thing thing itself 1, 000 Max resource package files 300 megabytes Max size for all resource package files and a 20 megabyte Max size for all report files so I’m guessing there’s just some hard limitations in the service that you can’t get to you may not be able to see those things because desktop just maybe forces you to say no or it doesn’t do it for you so if you’re adding a lot of things in there there are distinct limitations on the report that you’re actually building so you have to build

13:29 actually building so you have to build still the build efficiently I just thought that was interesting because you’ve never you’ve never heard that before you’ve never you’ve never heard a PBX file can’t have more than a thousand pages I don’t know who’s doing that but you can’t do it so if you’re on page 999 just be careful good luck good luck anyways I don’t even know how You’ count that many pages in in the report I that would be impossible yeah anyway someone someone should try it to see if it actually does fail anyways that’s all I wanted to

13:59 fail anyways that’s all I wanted to note here about the new features for desktop not a lot of features it looks like build and ignite or sorry build and fabrion exhausted all new features now we’re probably just in bug defix fixing mode at this point any other news for topic items all righty those are the big ones let’s move over to our main topic today then everything is still bi

14:29 then everything is still bi Tommy take it away give us a a quick little summary of this article and we’ll jump in from there so I think this is more or less one of our more hot take just overall on what is the industry of bi right now and what Ben I believe goes through is he’s going through the evolution of business intelligence tools and how new companies and products are attempting to innovate in this space say hey we have something completely new but often they end up doing the traditional features

15:01 doing the traditional features and it always goes back to what users are looking for and no matter what the market trends are trying to stay different there’s still customer expectations of what a bi tool can do and eventually how we always lean and go back towards those main features where we say hey we have a new query ability and when Tableau came out with things how what powerbi has been doing and you see mode quilk I think there’s one called sens too that tried to innovate

15:31 called sens too that tried to innovate but end eventually ended up looking very similar to all the rest of the tools out there yeah so and I think that’s a really big lesson to learn for us in general and as we kick off this conversation there’s a lot here about we’re innovating constantly in this space and I think more or less just the tools but talk about like our teams and what we’re trying to do in the bi space like Hey we’re not just building reports anymore we’re building

16:01 reports anymore we’re building insights and we’re be able to do data science and we’re going to build all that for you eventually we always go back to the fundamentals and I think there’s some very big reasons for that but I’ll pause there to see what are your guys thoughts on the article one thing I I’m finding here is looking through the article and reading through it he has a ton of references in this article which I found very interesting most of the articles that we that we review are someone giving an opinion and writing

16:31 giving an opinion and writing down a couple notes here every other sentence in here is a link to another article you could spend an hour just going on rabbit trails and all these little links and references he’s putting out in the beginning here because he’s using a lot of of existing shatter talk I guess what and to verify his points and I would agree I think a lot of what he’s saying here is is it’s not what is the real differentiator between different

17:01 differentiator between different business intelligence tools what’s the difference going on here and I think to his point a lot of it’s the same it’s not a new it’s not a lot of new innovation happening here the only thing I can really say is you’re competing on price at this point who can do it better faster cheaper and I think that’s why you’re seeing the proliferation of powerbi it’s it’s a better mouse trap for a lower price I don’t I don’t know if if I have the right word but it it reads almost

17:32 the right word but it it reads almost satirical right where oh yeah he’s got a there’s a there’s an undertone to this article right like oh you have a you have a everything’s bi you have this great new idea and he walks through how you’re going to take the the next thing that’s so unique from business intelligence I can’t talk about I can’t it seems like he’s talking about the company mode that he works for or it almost sounds like he’s part of the the development of it and it it’s it’s

18:05 development of it and it it’s it’s like the whole thing reads like you’re going to do all these great things and then you get all these invest like a startup story and then you just land right back where at all bi tools land right which are like there are a lot of needs there are a lot of customers or you are a lot of customers or prospects that start to force your know prospects that start to force your vision off into different directions right and I I think there there are some interesting cautionary things for startups in that realm like you’re

18:36 startups in that realm like you’re forced with well if you guys have ever worked for a startup all of the sudden it becomes very well this customer wants this well we don’t that’s not what our tool does well we’re we’re gonna build it for them then right and then you become beholden to the customers in the development of your product heavily right as opposed to what I think he outlines well which is a much harder path which is you have a vision you’re going to stick to it and for all intents and purposes customers and

19:07 intents and purposes customers and others are going to conform to that new vision I I think cautionary tales within here where I think he’s even providing feedback where it’s like yeah and the vast majority of those companies fail like it’s it’s nice to say Amazon and it’s nice to say SpaceX and all these things but it’s like those are two companies out of how many companies that have started right so like yes can it happen yes is it hard obviously so to your point let me quote probably I think

19:38 your point let me quote probably I think G to be our my favorite paragraph here and I think some of you guys all appreciate some people might ask what Garder magic quadrant you’re in ridiculous you’ll say the old categories were in something that we can’t fit into the categories of yester year you’re going to set your own Gardener research to prove your point right and we don’t fit in the normal Gardener quadrants we’re so Innovative we’re we’re different than all that yeah and I think that’s a really there’s a we we talked about Silicon Valley 2 we did not get to season two but once we go

20:11 did not get to season two but once we go to our next MVP Summit and talk about season two the L least the next season is talking about well what do you like they’re talking about the cloud stuff and they’re like what do you want a box we can build a box like why can we get that box delivered and everything just becomes about a completely new product that the customers want and Seth it’s a good point about talking about the startups I wonder if we could take this conversation or or I don’t want to divert too much but talking about the service side because obviously we’re not the one Building Products right I’m not

20:42 the one Building Products right I’m not building I’m not competing with Microsoft on a business intelligence school or I it’s probably not the greatest greatest idea but me as a service is constantly

20:52 idea but me as a service is constantly trying to deliver something I’m trying to go through on what are my services what can I do from an effort point of view view that’s much different than and I think that or least changes because I say oh well we can do X Y and Z it’s like well it eventually just becomes a report I don’t do you guys see the the equivalence there not sure I’m not sure if I see exactly I gotta be honest my my mind was going down more of the Silicon Valley route yeah teacher

21:23 Silicon Valley route yeah teacher question Tommy yeah so obviously we’re not products we’re not building business intelligence tools to compete with Microsoft however we’re in the game or in the industry where it’s our services and I think we’re always trying to deliver better services and it’s not always the same if I was doing the same thing I was doing in 2015 I’m going to build this report I don’t think that’d be very successful and I’m trying to look at other ways I can deliver especially with the new features and tools available however I find myself or

21:55 tools available however I find myself or people still asking for some of the same Services of the same features well yeah it’s great that you could do XY and z and I I like that you’re moving this direction but can I just get a report can I just get a table again I don’t know you’re going with the question honestly see the question here in in an ironic twist of fate I think Seth gets it oh

22:27 Seth gets it oh wow wow so yes we we provide services within the constructs of the bi tools that we’re using I I using I I think are you’re just talking about the different personas of the solutions that we build within the context of those tools right so whether it’s this tool he’s talking about which I would like to talk about further or what we’re providing I think yes there are you just need to report things yes there are we need a little bit more

22:57 are we need a little bit more of a complex solution yes we have the we need some automation around ETL whatever I I think how you find Value Within These These or the tool you build to solve these problems problems has to take the like all of those things in mind right like are you building things in this service that can extend the value to the business all the time right where yeah great you need this report or you’re asking for this

23:27 this report or you’re asking for this data dump why is that oh okay well we’re going to preliminarily start to collect and consume the data you’re going to do some ad hoc analysis and then we’ve talked about this it filters up if there’s a path if there’s somebody that’s showing them the road to success right in terms of how okay great I did my ad hoc analysis what’s next okay great I built my report what’s next okay there’s a lot of value here how do we get this across the organization right like the steps to create and

23:58 like the steps to create and maintaining and then ultimately enhancing your Enterprise level Solutions right and I I think what’s interesting to me in this article is we look at fabric right and all the offerings Microsoft is running out it the design here is is interesting because it’s bringing all that together but in a in a very backend focused way mhm so I I don’t know if that answered your

24:29 so I I don’t know if that answered your question no you I literally could not said it better myself so what’s very interesting to me and I’ve said it multiple times in the previous pod like previous podcast is I’m a Microsoft Fanboy obviously I know Microsoft products extremely well and fabric was the thing that I’ve like has Consolidated everything I’ve never seen anything else like it what is interesting about stumbling across this article is he is talking about mode and I’ve never heard this tool but it it’s

25:00 I’ve never heard this tool but it it’s it’s mode analytics which apparently reads very similar to an ecosystem that you bring all of these capabilities across the business intelligence stack together and obviously it was compelling enough because thought spot bought them last last year and now they’re part of that where there’s I now know of a second tool where ad hoc analysis self server reporting interactive dashboards

25:30 reporting interactive dashboards Advanced analytics custom data apps like a a tool set is out there and apparently according to them they’re in half of the top 500 companies in the world so I think there’s a very unique perspective in story around whether it’s building a product that’s supposed to be revolutionary and completely different and whether or not the current version of mode is that or he makes it seem like it turned into something that is just another bi

26:00 is just another bi is is a lot of where business intelligence is going trying to rectify the Gap that has always existed in the different areas of the business doing stuff and I think the answer is yes I stuff and I think the answer is yes especially with where generative Ai mean especially with where generative Ai and all the other things are going is to combine and bring these teams closer having said that it’s very interesting how I think just

26:31 interesting how I think just reading quickly through mode thought spot Etc like how there how you attack that problem is it from the business side or is it very Tech heavy and I feel like fabric is very backend heavy just getting well harbi I think was not very backend heavy and then all of a sudden fabric shows up and fabric was only back in there there’s no business side to anything in fabric honestly right name name me one

27:02 honestly right name name me one thing in there that’s front none of it is it’s it’s more backend heavy tools that are being simplified for backend users and maybe some business users are going to be adventurous to go forward into that I’m just I’m doing a streaming project right now yeah and streaming data like that was all about build build was streaming everything it’s all so easy to stream everything we’re building some solutions and it’s it’s streaming things but it’s not as smooth as you would think so it it’s

27:32 smooth as you would think so it it’s working but it’s not quite there yeah and and in some respects like a lot of that I think is is probably enhancing the ad hoc analysis aspect of things right because most of those folks no SQL or python yes right and now you’ve opened the doors to data right and then consolidating data so maybe it’s just my perception right of of how the the documentation and how we’ve launched into the different aspects of fabric but I think what’s interesting that in this

28:03 I think what’s interesting that in this article that drives forward it in this approach whether it’s mode and thought spot or fabric is are they staying true to their Vision or they trying to serve so many customers because now you’ve opened this ecosystem where we can play together and so this is a huge Point Seth because you I had an light me moment as as you guys were conversing half my job or half my time doing training with just powerbi has

28:34 doing training with just powerbi has been seen report to going oh that you’re doing it wrong where they had like a January February March table and each of their things like oh this is terrible but it worked for them from a user point of view a lot of powerbi you could get away with without having the technical expertise and did survive and thrive at organizations not necessarily having the best semantic model because you’re to your point a lot of it was user facing a lot of it anyone can do both from a heavy data developer to someone who’s

29:05 heavy data developer to someone who’s just getting started now we’re getting to this you started now we’re getting to this fabric doesn’t fit into a normal know fabric doesn’t fit into a normal business intelligence category because it’s data engineering it’s all these are the features well Tommy has a moment goodness gracious yeah I was all right we’ll keep going talk about yourself I think I think what’s what’s funny to me and and well not funny there’s a there’s a paragraph in the

29:36 there’s a there’s a paragraph in the article that that reads like this in other words no matter when a data product was created or what its first features were it will inev inevitably collapse into the same Singularity database connectors a code-based query editor a code-free query Builder lots of charts dashboards selfs serve bingo so I like so really like that point and then also I think about halfway down the article it talks about something like

30:08 article it talks about something like they were doing a pitch to some billionaire about their tool and they’re like hey we’re doing a pitch and you like hey we’re doing a pitch and d da here’s our here’s our super know d da here’s our here’s our super fancy thing that we’re building and you fancy thing that we’re building and the the billionaire basically know the the billionaire basically said you’re just basically a bi tool and all the bi tools now are using fast in memory compute engines that’s what they’re doing it’s fast in memory compute everyone’s doing that you’re not doing that so you’re going to need to like you’re just a bi tool see you and basically blew them off and then he says

30:39 basically blew them off and then he says literally the next statement and then six years later we rebuilt the whole thing using an inmemory database and now we rebranded as modern VI like right there there are I think there are certain trends that are just existing and this is this is one of the reasons why I think I really like Fabric in general is they didn’t really rebuild any like of this is rocket science honestly like if I look at the product of fabric now it’s not new stuff it’s pipelines that were ADF it’s it’s spark that has been around in other

31:09 spark that has been around in other tools forever notebooks are not new to fabric it’s something else that existed

31:13 fabric it’s something else that existed other places in synaps and other places data bricks other other companies are doing this they didn’t invent the format of Delta that’s been around for a while they’re not just using Delta in their tools they didn’t invent SQL they’re not just putting SQL serverless directly inside Fabric and letting you buy it like to me like the better mouse trap of fabric is it’s a now a package deal I don’t have to go turn on Individual Services you can go turn them on as needed and they they now pull from the

31:43 needed and they they now pull from the to me that the difference here is they all pull from a pile of compute units there’s a there’s a pile of common compute units that I’m purchasing each month and in order for them to get to this fabric thing they needed to shift our minds away from a premium capacity as it was in Virtual cors but it need to be this compute unit thing because there’s a lot more services other than just computer you need to bundle together to get this thing to work correctly well I I think the argument on the other side though is the people like you may get you may get the products to

32:15 you may get you may get the products to continually to evolve and rapidly evolve right now what we’ve done with fabric is basically taking everything on steroids so to speak to put it all in one space but that doesn’t mean the people have and that’s what lags behind usually and again to the point of what people are looking for the people are going to dictate and sense the use or can be a big part of what’s actually going to be used is the expectation right now with fabric that everyone’s going to move over and this has been our

32:47 going to move over and this has been our our Everlasting conversation or question is is everyone moving to fabric or are we growing people in house in fabric to start using those products you may have have all those features and that’s great to have all the data engineering features and all the data science but you still have the same people on power B right now who are the the data modelers and you may have some data engineering but you or you have a lot of the business users are we expecting them to pick up and now expedite their own

33:18 to pick up and now expedite their own skill or are we expecting to bring everyone in and I think that’s the ever the the the question that we’ll have until we get to more solidified place place for me I’m looking at this and it’s great to have all the features there but what are we doing about the people who come have continued to work in powerbi that are the business users that are are just the powerbi people are I think we still need to have that environment for them on it’s still that bi tool of I

33:49 them on it’s still that bi tool of I need to build reports and I need to build semantic models just because the products are there does not mean that everyone should be using them yeah I yes and no right I I I think I think using getting use out of them versus things you build maybe those are the same thing I don’t know I think I think one of the the challenges in that though and I’m I’m trying to quickly

34:19 though and I’m I’m trying to quickly scroll through some some things because one of the one of the challenges with all of these data teams working together is is ensuring that you’re supporting the the efforts of those teams Andor giving paths for an organization to set standards around right so even if we say hey this is the net new platform business users here’s how you use it like you can create these objects you can do your ad hoc analysis

34:49 objects you can do your ad hoc analysis you can do these things this is the path for graduation right which we still don’t have we’ve been talking about how do you even do that in the powerbi ecosystem I think one of the challenges I have in here and it it’s still it’s not front and center to me like H how how do you successfully manage data objects and data within this ecosystem you you have to have visibility right like you have to have connectivity you you as the data Engineers need to understand how objects

35:20 Engineers need to understand how objects are being used and where you as a business user are looking for things where what can I go look for where is my data catalog right like where are all the objects I have available to me that I should be using instead of my own spreadsheets and I think in large part like to me that’s perview right and we we have like an admin view of the perview Hub which shows us some things that are interrelated but I guess the question that’s like a hot take is does that need

35:51 that’s like a hot take is does that need to be more front and center for fabric to be successful across all these groups in that like hey it’s very easy to Traverse this ecosystem of here’s where stuff is here’s what you should be using because if you think about even like metrics right like these are the metrics in the organization here’s what you want to use well that could be simplified if there’s a way in which I can Traverse and get people to align and use the right artifacts across this ecosystem it

36:22 right artifacts across this ecosystem it would it would definitely streamline what people are trying to to find and access C and what organizations are trying to build from the other side right like if I’m going through all the effort to build a data set that is certified I want you I want publicly that easily discoverable for my organization so that they can use that right otherwise we’re going to find ourselves in the same problems that we have right now congratulations you’re building something in your workspace and

36:54 building something in your workspace and nobody else knows it’s there so 10 other people are doing the same thing albeit not as good as you guys right and I think I think for any system to work really well in this ecosystem where we still have all of these different groups with different needs and working in different parts of the tool you got to syn around the data and it’s I don’t I don’t think I’ve seen a ton of stuff that tells us how to do that like well yet right we’re just stitching it all together but at the

37:26 stitching it all together but at the same time like having and maybe this is where co-pilot comes in I don’t know I need to find this data set okay here it’s certified it’s this workspace you need to go ask for permissions or something like that but I don’t know man give me the spiderweb well let’s focus on that spreadsheet part because that makes that makes me chuckle because the user story now or what it’s been in the past is we go and it’s like oh you have the spreadsheet well we can we can automate this we can create a report it’s a very different conversation like oh you have the spreadsheet we can Medallion approach this and that’s a

37:57 Medallion approach this and that’s a very different conversation for users where if because spreadsheets are not going away just because Fabrics here doesn’t mean business users are going to stop using spreadsheets and we’re going to still have to go in to help improve their their processes I think where I’m going with that is does that mean that we have to stop the approach that we’ve done now with hey you have these spreadsheets we’re going to build a semantic model office we’re going to really save you some time where it’s like well we have fabric so we’re going

38:28 like well we have fabric so we’re going to have to do this Medallion approach we’re going to have to put this in a Lakehouse I don’t know about that I don’t know where I lean on that question but I think that’s really a Prominent Point right now for we have all these features and tool sets you but you’re still going to be dealing with spreadsheets you’re still going to be dealing with some very raw unstructured data that’s not necessarily meant maybe not necessarily meant for a med Medallion but it’s still coming for the business

38:58 business and I think it’s like what’s our take on that that interesting there’s a couple thoughts rolling on my head i’ I’ve been sending you guys some messages in the chat about things that I’m seeing here as well one of them is a very fiery comment from Marco Russo basically talking about this whole in memory tabular model it’s the best thing since sliced bread kind the best thing since sliced bread deal and it’s been around for of deal and it’s been around for a long time to be very clear like Tabler

39:29 a long time to be very clear like Tabler models is not something new this is a very long running technology and it’s been just getting refined and improved and faster and it’s getting awesome I’ll put the chat I’ll put the the comment from Mark he must have read something somewhere about someone touting some other bi tool that had this incredible performance or whatever the thing was but I I I think I agree with them I think you I I think I agree with them I think the the the tabular model semantic know the the the tabular model semantic model everyone is copying that pattern

39:59 model everyone is copying that pattern right now it is load things into memory tabular model things and then hyper Fast Access of data through that Tabler model anyone else is doing things differently than that it’s just slower and my experience has told me like if I’m connecting to a SQL Server if I’m connecting to other like if it’s not part of that Tab model it’s just slow compared to like the speed of the TBL model it just can’t do some of the impressive things that the tabl model can do at scale now there’s probably always edge cases for every little thing going on here but I I just found

40:31 going on here but I I just found this comment interesting and and very relevant to our topic as well so I’ll put the the Marco Russo Twitter comment here I don’t know where that came from U but I thought that was very relevant in this conversation of like look nothing’s new it’s the same thing that we’ve always seen it’s a Taber model I think I think we need a button Mike that has like a like a rewind like an old school tape like yeah cuz you’re comment was like 10 minutes ago in the convers very relevant

41:01 minutes ago in the convers very relevant for and then we can be like then we can be like back to I know I know see what I do with the editing of the PO I I realize this comment is is very I know I’m catching up here I was waiting to see if you were gonna like weave that into are we weing ecosystem mess well okay I’m I’m getting there because I I’m what I do is I write I’m writing notes and teing up my thoughts here so like I I’m catching you guys talked so much I have

41:31 catching you guys talked so much I have to catch up with my thoughts have a thought from the intro so how about PB

41:38 intro so how about PB right so okay that was one thought I had I thought I’m thinking about this whole article and I’m thinking like this is very relevant with marker Russo okay I’m G to catch up a bit more all right skip ahead in like five minutes now all right so now you’re talking about where do we organize all this content and the thing that comes to mind is I just saw a blog post from data bricks talking about they’re now open sourcing the unity catalog so I don’t really know what that means from a practical standpoint I already use data bricks and I already like Unity catalog so I’m very pleased

42:09 like Unity catalog so I’m very pleased to see them doing this and I think the idea here is if I if I look at there and and you can go to look up more information around Unity catalog it’s Unity Unity catalog. is the website where they’re promoting this new open- source Unity catalog and it’s talking about these different catalog Partners it’s talking about the different systems that you can play with there’s Delta ecosystem the iceberg e ecosystem and Hoodie system so those are basically the three formats

42:40 are basically the three formats you would use to store these tables and then it’s talking about like hey this is a unified system for openness and cataloging and here’s all your data it’s all I don’t know how yet to leverage it all yet but to your point Seth this is not like to me here it’s not the Revolutionary of like this is a newbi tool this is a it’s it’s a more it’s a tool that’s taking the same stuff we’ve always had and making it a couple less clicks or more of a unifying

43:12 less clicks or more of a unifying experience of bringing every so to me this the part is becoming it’s more consistent around all the information coming in the same place and to your point Seth it’s becoming more discoverable and I think that’s the the key here is what we’ve been doing traditionally in bi has been maybe more siloed in nature or has been focused solely on Purely technical teams I think what we’re trying to do is we’re trying to take the existing stuff that we’ve been doing for years and make that

43:42 we’ve been doing for years and make that easier for more people to access and potentially not necessarily move the data into different places but more more or less along the lines of like what Unity catalog is doing it’s providing more quote unquote shortcuts to where data already lives let’s leave it where it is here’s a better mouse trap to see all the artifacts we’re talking about I I could also make the the complete opposite argument and lean right back into same Singularity right like data bricks is where they just release in preview like it’s basically

44:13 release in preview like it’s basically easy connections yes correct that’s what it is all the connectors that powerbi and ADF and Microsoft has invested lots of money in doing right like it’s a singularity thing everybody wants the easy button so they’re building or they bought a platform to do into that I don’t think they’re doing anything Innovative I think in that regard they’re catching up or trying to right they’re serving that part of the the customer base that’s like yeah but it’s a lot easier over here to do it right and and I it’ll be interesting to

44:44 right and and I it’ll be interesting to see how they extend as a platform because right now they’re hard it’s data science data engineering right yep yep e ETL at some at which which is actually funny some ad hoc like ad hoc analysis but then they just spent a bunch in integrating with powerbi in their visualization Matrix so it’s like making that ad hot experience better right exactly so so I think there’s a lot of play across

45:14 there’s a lot of play across all of these companies to be competitive in a market and I think some are obviously further ahead than others because of the the capabilities like power ey whereare you’re you’re deep in the business right like your business people are getting value in a way that’s easily consumable to them and I think this even relates somewhat to the visual calculations point which is how how do you bring people business people along for the ride well it give them easy paths simple

45:44 paths simple todos like the tutorial to follow that they can apply to their data and get value and then they’re hooked well the argument I’m going to make is is that misleading then just because I can easily sign up for something or easily create a shortcut or easily create a new connection I have Unity catalog I have all these shortcuts in my Lakehouse that doesn’t mean it’s easy to look at if anything I’ve just made it much more complicated if I don’t come in with the right plan you’re making all of

46:15 with the right plan you’re making all of these things incredibly easy to sign up for and to create for a business user where they don’t have have the technical skills right but all of a sudden you’ve now signed up for Unity catalog you’ve now created all these short shortcuts because you can it takes no development skills but if you are a business user with just like we have with powerbi I’m creating tables for every month now times that by five in the fabric ecosystem or with all these different tooling all of a sudden my data is living everywhere and I didn’t come in

46:45 living everywhere and I didn’t come in with a proper plan it’s so much easier the barrier to entry so much lower to get my data in all these different systems that doesn’t necessarily mean it’s the right the right path that doesn’t mean we’ve created a best practice the best practice going back to the article is still the same when it comes to bi I think you make a really good point and and the reason I think it’s more relevant than ever before is I think that is potentially

47:18 before is I think that is potentially the thing that will kill these Solutions these products and the reason is not because not because this rampant data proliferation or people using data sets from all over the organization like doing stuff that gives them insights because that’s what happening in business already yeah it’s already being done already doing all this just in Excel the problem is like so the opportunity is hey we can we can do more with this and make our

47:49 we can do more with this and make our ecosystems better and I think that’s where there’s a lot of interest and opportunity Even in our Realms and how we talk about it however the problem is you’re giving them platforms that are going to cost the business exponentially more more money right now they live in their machines they live in their own ecosystems platforms that are driven like it’s all off in the in the side and what you’re offering is tool sets that

48:19 what you’re offering is tool sets that now bring that into paying and that’s fine but all of the sudden when you turn this on and Bam everybody’s throwing this stuff in and Tommy there is no plan right and people are trying to figure out why the capacity maxed out already and we need another one what what is the business supposed to do just be like well this all worked before why is it costing me hundreds and thou hundreds of thousands of dollars a year for business intelligence well it’s because we have

48:49 intelligence well it’s because we have all this this new stuff that was in the business well was it working before why did why did we put it all here and that’s where I think this the sell of why we do all this has to come at some point outside of unifying the stack and unifying business intelligence within an ecosystem because is there a lot of value there yes can we build things even faster than we did before yes so you can scale but you’re also serving this other

49:19 scale but you’re also serving this other side of the business that isn’t costing you infrastructure money right we can be diligent in how we go about building things but you’re opening this ecosystem to everybody right right and just a button yeah cautionary tail man I don’t know it’ll be it’ll be interesting to see because you are you’re pulling all of that data and if it’s relevant cool but at the same time you’re looking at un unskilled data people right I’m just going to load up my data I’m going to do these things because this is what I need

49:49 these things because this is what I need to do and I’m going to slam it out the door okay like it are is this refreshing are you using it all the time are you like what capacity are you using like etc like etc etc there’s a method organiz it brings it into a different realm and I think Tommy that’s why your point is really good because garbage living in the environment is fine right now you pull it into this this ecosystem and it it may not be fine I like your point there Seth this is this sounds a lot like what

50:19 Seth this is this sounds a lot like what I see on there there’s a there’s a higher I’m going to I’m going to make an analogy to what I think I happening in the SQL Server realm as well right so in SQL servers on Prem you buy the hardware the computer’s already paid for you you sized the machine appropriately someone made the business call and say we’re going to buy the hardware it’s done you can then run that thing as hot as you want much as computers you need all day long boom hit it up run go all these things it doesn’t matter you’ve already paid for the machine you’re only paying

50:49 paid for the machine you’re only paying for the electricity at this point to keep that server on and running so there’s less of a day monthly cost to run that piece of Hardware to your point Seth also I’m thinking the same analogy fits for when you use Excel everyone’s got Compu on their computers you’ve already invested the capital expense of everyone needs a computer already we’re going to have to have it so everyone five license which is everything right sure here’s the everything here’s all here’s all the software you’ve already you’ve already paid for the license the E3 or whatever

51:19 paid for the license the E3 or whatever you get office you can’t run a business without PE giving people Word and Excel it just it’s a table stake so in those situations you have all this sunk cost that we’re now moving into this Cloud ecosystem where it’s not sunk costs and the things you do more of actually need to be evaluated against this the spectrum of is it adding value and how much is it costing me and this is what we’ve talked about a lot in the podcast I think I’ve made this point a lot it’s it’s a balancing act of how much value am I getting from the

51:50 how much value am I getting from the information that this thing is providing to me against how much cost it is costing me you may spend a million dollars a year on powerbi if that’s still adding value to your business that may be totally justifiable that may be

52:04 may be totally justifiable that may be right in line with what you need but if it doesn’t continually add value at that level for your business right it’s it’s it’s the pure spend a million on BI make two million in Revenue sales or whatever other things that are work that are happening if that’s the justification yeah all day long I’ll spend another million to make five I’ll spend a million to make 10 right that that easily justifies itself when you have tools at your disposal that can do this now something I will I’ll point out here as well there’s also some

52:35 point out here as well there’s also some Market I’m going to call it Market winds or maybe Market movements that are also Happening Here I think as well I think the proliferation of data itself is also increasing as well so I do think organizations can’t take this whole stance of it’s going to be a set cost forever because the volume of data that people in like the software the people the more things we get computerized in in our business Realms the more data is going to be coming off of machines and

53:05 going to be coming off of machines and automated and cameras and video and all so just the general volume of data is going to increase so something has to give here you you have to either have better more efficient tools to do the work in or you have to have you have to be more metered around what information is actually important to us and finding out we’re not going to store everything at some point in time you’re gonna have to just start throwing away data and be like look it’s not valuable we can’t use it we just it’s not worth it to save for us it’s not worth the

53:37 it to save for us it’s not worth the cost to keep it around so I drive yeah like data itself is not is not the huge cost driver right which is why storage is cheap compute compute is expensive yes and I think I think it just just it we we we typically always take it back into admin Realms right like it if I can’t have an answer for what is happening in an ecosystem right it’s going to be challenging to

54:07 going to be challenging to justify the the the outcomes right already business intelligence has a a challenge to overcome which is keeping track of the value that you’re providing do we know that we’re driving efficien of course we do does the business are you doing a good job explaining it right are we not yeah that’s a great Point are we driving the insights for all this Revenue that we’re making like how do I equate my work towards that right like so all of a sudden it’s like

54:37 so all of a sudden it’s like can we like I think there just needs to be diligence around okay we’re going to create some capacities and these three departments are going to share this for the time being right and then we’ll figure figure out how to do chargebacks and like who owns what why do they need it so essentially what you’re trying to say is we have a unified platform if you guys need business intelligence it’s fabric right here’s Here’s Your Capacity here’s what you do but it it takes diligence like this is where I I don’t think scenarios

55:07 this is where I I don’t think scenarios like powerbi are going to work well because of how fast things could explode where it should like the wild west with fabric is scary to me right like if you just open that up in a large organization like how fast those those those usage charge would just be like boom like it’s there like I’m going to max out all these capacities but whatever I I could just be fearmongering but at the same time like I think there there is diligence in how you need to implement these things and it is I like

55:37 implement these things and it is I like I said much faster to spike capacities when using some of these other tools if people don’t know how to use them or it’s like hey I have access to this thing I’m going to I’m going to query all of the tables I’m going to join them all together all same time full load everything all the time like this is long query I need this wear filter like two hours later the heart of the matter though too we’re talking about the fundamentals or fundamental bi which changes slowly because it’s the fundal

56:09 changes slowly because it’s the fundal to you well fun what fundamental to me the problem exists in these users who have access to these ecosystem and don’t know all of the things that you do that’s very different than just saying we’re I’m saying what are we actually doing is it a means to an end or are we just doing it because it’s there there available exactly like just because like hey look we put all of our data in a lake house in a medallion approach we’re also spending three times more before

56:40 also spending three times more before did everything need to be in The Medallion approach why are we spending three times more is that valuable to the business that my email campaign data is now in a medallion architecture when it wasn’t before and we survived there’s a fundamentals to your point of delivering insights at accurate data that doesn’t necessarily mean that it’s conforming to the new tooling and fundamental bi changes slowly because the principles they it should change but the principles are not going to change and it’s going

57:10 are not going to change and it’s going to say hey what are our means to an end what are what’s the end goal and what’s our like principal principal tooling tooling so I don’t know man I I think for me I’m looking at all of this and it’s like all the tooling and the services are great but at the end of the day whatever I’m going to do if I’m going to have everything in the Lakehouse or I’m going to keep everything in powerbi or have a mix of buf each of those need to have this very planned and mapped approach towards am I delivering the

57:41 approach towards am I delivering the best value in this situation I think I think one cautionary T I want to go back to your point Seth cautionary tales I know we’re at the end here I think the cautionary tale that I see coming up next is this whole message around real time streaming and moving data all the time as it as it comes in as it as it efficiently goes through all these things I definitely think this is a this is a pattern that happens in software development it’s this thing called microservice architectures Data can always be moving around you can always have things talking to each other

58:11 always have things talking to each other I do think it’s a good idea but I think the value needs to be very clearly evaluated against what is realtime data and really exploring the cost of what that real-time data really means in order to to leverage real time things as your data comes through I don’t to me that real-time information every time I’ve done a project around realtime stuff some computer some compute is being turned on all the time and continually running it doesn’t seem to be cost-effective to always do that I would

58:42 cost-effective to always do that I would rather have batching or micro batching experiences to again to your point SE the question should always be asked to the business if I give you data every five minutes what decisions are you making in five minutes that will change the input of that data if you can give me realistic answers like I’m on a manufacturing line and that’s I can turn off the line and save some get not have any bad products that makes sense there’s a reason behind it there’s

59:12 sense there’s a reason behind it there’s a there’s a rationale but to say you’re a marketing team and you need data every five minutes are you really changing your marketing platform or you’re marketing messaging in five minutes in order to tweak that that performance I don’t think your your your cycles of speed of you can’t even make decisions that fast to influence things is is relevant so I think it’s a lot of these to your point set it’s going to be a lot of balancing these needs and making sure that we appropriately Design Systems pipelines of data reports of

59:43 Systems pipelines of data reports of data that match the speed of the decision-making that we’re having in companies so those things need to be paired together all right yeah I I would say final thought final thought yes I’m say from the article from the article article that resonates not only with tools but I think people as well is the biggest lesson though is that it’s actually very easy to think different but it’s very hard different different yeah I highlighted that one and and if I if I if I bring that full circle into like

60:14 if I bring that full circle into like what we do here right is how do you stay different and it’s it’s joining conversations consuming new thoughts and new ideas new ways of analyzing what we do on a daily basis to bring ideas to the table and you stay different in that that alignment when you keep engaging with different thoughts and ideas and hopefully we help with that good tieing Tommy where else any other final thoughts for you I can’t do I can’t do better than that that was perfect I I want to agree I I literally

60:45 perfect I I want to agree I I literally went to I use we use this little raindrop. io app to save and store our articles and in the in the app it allows you highlight where and this is one of the items the article I highlighted verbatim it’s easy to think different but it’s hard to stay different so anyways I hope we’ve done if I think about stuff that we’ve done set that’s been different we’ve been pretty different around how the theme generator Works we’re we’re pretty much carved out of Niche inside the space Around theme generating stuff and I feel like that’s another one of those things that we’ve we’ve thought

61:15 those things that we’ve we’ve thought differently and we’ve for the last six years or so stayed different in that space and we see other people copying us so maybe we’re not different anymore maybe we’re now becoming part of the mainstream all three of us are radically different yes exactly awesome with that let’s give it a wrap here so thank you very much for listening to this article I hope you enjoyed the article today around this one around how everything is just normal bi don’t worry your new things you’re building is just standard bi just go buy Microsoft fabric it’s all

61:46 Microsoft fabric it’s all you need it’s the only standard bi that you’d always just the same old thing so that being said thank you for your time we appreciate you listening around this one I hope this was some good thought provoking provoking notes here around bi in general and what you’re doing in your organization maybe there’s a couple cautionary tales that we gave you that may you may want to look out in your business with that please share this article or this potentially this podcast to other people if you found this useful to You Tommy where else can you find the podcast well you can find

62:16 you find the podcast well you can find us on Apple Spotify or wherever at your podcast make sure to subscribe and leave a rating it helps us out a ton do you have a question an idea or a topic that you want us to talk about a future episode head over to powerbi. com

Thank You

Want to catch us live? Join every Tuesday and Thursday at 7:30 AM Central on YouTube and LinkedIn.

Got a question? Head to powerbi.tips/empodcast and submit your topic ideas.

Listen on Spotify, Apple Podcasts, or wherever you get your podcasts.