Is Microsoft Fabric Business Intelligence? – Ep. 464

Is Microsoft Fabric a BI tool? A data engineering platform? Both? Neither? Mike and Tommy wrestle with Fabric’s identity and what it means for Power BI professionals navigating an increasingly broad platform.

News & Announcements

• Copy Job: Connection Parameterization, Expanded CDC, and Connectors — Copy Job continues to mature with parameterized connections and expanded change data capture support.

• New Dataflow Gen2 Data Destinations and Experience Improvements — More destination options and UX improvements for Dataflow Gen2.

• Next Generation of Data Transformations with Dataflow Gen2 — FabCon Europe Announcements — Big Dataflow Gen2 announcements from FabCon Europe, signaling continued investment in the transformation layer.

Main Discussion: Fabric’s Identity

The BI Professional’s Dilemma

Power BI professionals are watching Fabric expand into data engineering, data science, real-time intelligence, and AI. The question: do I need to learn all of this?

Mike and Tommy’s answer: not all of it, but more than just Power BI.

What Fabric Actually Is

Fabric is a unified data platform that includes BI as one workload among many:

• Data Engineering — Lakehouses, notebooks, Spark
• Data Factory — Pipelines, dataflows, copy jobs
• Data Warehouse — SQL-based warehousing
• Power BI — Semantic models, reports, dashboards
• Real-Time Intelligence — Eventhouses, KQL, streaming
• Data Science — ML models, experiments
• Data Agents — AI-powered data access

Where Power BI Fits

Power BI remains the primary consumption layer—and that’s not changing. But:

• Semantic models are becoming the bridge between BI and AI
• Data preparation is increasingly happening upstream (lakehouses, warehouses)
• The “full stack BI developer” now touches more Fabric workloads

The Expanding Skill Set

For BI professionals, Mike and Tommy recommend:

• Core: Power BI, semantic models, DAX (unchanged)
• Expanding: Basic lakehouse concepts, SQL in Fabric, understanding data pipelines
• Awareness: Real-time intelligence, data agents, data science concepts
• Not required: Deep Spark/notebook engineering (that’s a different role)

Looking Forward

Fabric is both a BI platform and much more. The BI professional’s role is expanding, not disappearing. The best strategy is to deepen your core skills while broadening your awareness of adjacent Fabric workloads.

Episode Transcript

Full verbatim transcript — click any timestamp to jump to that moment:

0:15 Thank you. Good morning and welcome back to the Explicit Measures podcast with Tommy and Mike. I was lost at their morning Mike. I’m there. Don’t worry, I’m there. Yeah, I got you. Oh, I almost made a mistake. That could have been bad. Like, where am I? What am I doing? Yeah.

0:50 Is this training? Are we at a conference? I don’t know what. It’s been a It’s been a long week. It’s been a long week. , a lot of things have been happening in the market, but before we get into the news and the topics for today, let’s just give out our topic today, Tommy, this is an interesting question. And usually, , someone teased us one time about, , any kind anytimes you have a question. It’s a rhetorical question. It’s like, no, the answer is always no. So, this is a rhetorical question, I guess, and we’re just going to unpack this topic. I don’t know. We’ll see this. Maybe this will be different. , is Microsoft

1:22 Fabric actually business intelligence? That’s what we’re going to go talk about now today. So, I think this is this is a good topic. I think Tommy, you’re making a really good point around this one or or the idea or the concept is there’s so many things. We’re not just doing BI anymore. We’re not just doing reports. So many other things inside fabric that’s happening. Has has the scope of fabric expanded what business intelligence is doing? Be interesting. I like this is going to be where do we draw the line in the sand or is there any lines or is it

1:55 Just more sand? I don’t know. We’ll see. We don’t run with that analogy. Water coming up from the beach. Who knows? I don’t care. We just pile on more sand. It’s fine. It’s all good. Just Just make more sand. So, we’ll see how this goes. Anyways, let’s jump into some news items here. Tommy, take us away. You Tommy have been very busy. You did a user group yesterday. Yeah, baby. In Chicago downtown. That was on October 1st. How’d it go? Dude, it was awesome. So, our it was a crash course on Microsoft Fabric and honestly, it was amazing because we get

2:28 The mailbags. We hear from you guys about very consistently a theme we hear from everyone is we would love to do fabric. We like the fabric topics, but our organization is slow to adopt. And the first thing that I I asked everyone was have you experienced in fabric? They’re like, “We all want to, but , everyone, they’re either our organization’s either testing it or, , they’re still weary of fully implementing it. They don’t know what to do with it.” Interesting. And but so everyone who came was interested and wanted to see fabric. And it was cool because we basically did we

3:00 Called it the end to end solution under an hour. Lakehouse did transformations in a notebook, showed for a lot of people actually at the user group did SQL or a lot of SQL work. So I showed It was amazing to see everyone’s just face like their jaw just dropped like oh it’s a SQL database but it’s like what no it is one and we went to SMS and then we did a query there. So we saw what you could do. Oh, then we can create views. And we went through the SQL analytics endpoint. And then we showed direct link and showed how quickly that was. And what was cool

3:32 Where everyone really I think was most impressed with fabric was I can edit a direct link model in desktop and also in the browser. Yes. And I had both open. So I created a measure in each and you saw how quickly they were reflected. Yes. And the the real takeaway was even if you just use fabric for the one lake to direct lake integration, you’re getting your money’s worth or none thing. And that was for everyone, at least from what I saw where everyone was like, “Oh, oh, I get it now.”

4:04 Yeah, this is Yep. So, and then we did the prep for AI. We did a few things there. Everyone also loves the get the answers so to speak where I I pick a visual and say hey suggest some questions here so that it helps co-pilot and so we did we did the prep for AI in less than 10 minutes. I went to the co-pilot standalone and simply said hey for this report how’s my sales people doing? It was conversational and showed it wasn’t the QA perfectly how we prepped it. It gave me I I did test drive this

4:37 Beforehand to make sure it would work. But sure it was amazing that I didn’t have to ask a lot of context once you prep things. We’ve talked about training AI tons and one thing we’re also going to do here. So check if you’re part of the meetup please check your email today because we’re actually going to send out a poll and it’s going to be around all the features in fabric because we covered what 20% of what fabric can do and we want to get from you guys what do you want us to talk about? We will be doing the user group every month in person. we weren’t able to do it in like the live stream, but

5:10 That’s something we’re working on. It’s a lot of work, man. It’s a lot of work to do all that. So, and that was I’ll say this from No, no, no, no. No, no. You It was a lot of work on you, but it was also I was supposed to help support out that one. I just got too busy with stuff this week. I’m traveling next week. It just got to be too much. I was like, I can’t get everything done all in the same time. I need the bandwidth. I got the Well, that also didn’t help either. So, I was looking I was planning on the wrong day and like 3 days before Tommy’s like, “Wait, it’s the wrong day.” I’m like, “What do you mean it’s the wrong day? Switch it.” I’m like, “Oh, no. I’m I’m booked out with meeting things.” So, Tommy made a Yeah, Tommy made a mistake. But no, honestly, thanks for

5:43 Asking. It was a great time. It’s great to be in front of people and also just that different environment. If you’ve never gone to a user group, you really got to join one because it’s not just a session. At least our user groups are not just someone talks and then we’re done. It’s like we really try to encourage and we get to know each other and understand what their, , situation is. So, yep, dude. It was awesome. And I’m so excited to start that up again. Great. Awesome. I’m looking forward to getting back in with you as well. We’ll have to do a little bit of volleying back and forth when Chicago will do one every month. , we’d like to also kick up one again back in Milwaukee as well.

6:15 So, if you’re in the Milwaukee area, we’d also like to get one going here a bit and find some businesses. The downside is in the Chicago area. Well, good upside for Chicago, downside for Milwaukee. upside in Chicago is you still have the Microsoft office. There’s a there’s still one there. They shut down ours in the Milwaukee area and that’s typically where we would meet. We’d we’d meet with someone there. We’d get a some a room inside that place be able to communicate had internet. It had decent facilities. We get there and hang out. So since we don’t have one of those anymore, we got to figure out what the next pattern looks like. and if we

6:47 Can find businesses that can help support at least the locations as well. So we’d like to ramp that back up again and keep going. And in both cases, Tommy and I both want are committed to trying to get it may not be the best recording, but we’ll try to do the best we can to record these live events. we’ve been acquiring gear and technology to help us actually broadcast and and send these things out as well. So, we’ll see what happens. hopefully we can stitch everything together here and bring a little bit of a mobile studio with us whenever we do these groups. And we’re geeky, too. We like the technology. So,

7:17 100% 100%. I I love buying the technology. So, let’s move on to a couple other things here. So, we have some other news items. Tommy, you’ve got a couple other news topics here that we want to touch on. First one here, simplifying data ingestion with copy job parameterization, expanding CSC connectors. Wow, that’s a mouthful. This is a quite the topic here. , this is what I So, let me just say this as an observation, Tommy. I don’t know if you feel the same way here. I feel like when Fabric came out, it was like a lot of like fanfare. It was very exciting.

7:49 Everyone’s like, “Oh, great. let’s go try this thing out. And it very quickly ran into like just a lot of like technical debt is what I’ll call it. And for the next, , 6 months, nine months, there was a lot of things just like getting reworked, , just being rebuilt. And then all of a sudden, I feel like over the last 3 months or so, maybe in the last 2 months, it’s very recent. It feels like the floodgates have reopened and all this technical debt and like Microsoft restructured a little bit. there’s, , who’s on what team. ,

8:21 What’s important right now? , let’s get the priorities straight. Is everything going to be built with CI/CD incorporated? Yes or no? Like there’s all these like really large rocks to lift. And I feel like we’ve gotten over this initial threshold of like a lot of those initial large rocks are being removed. And so I feel like I don’t know what I don’t know what it is but this one around data flows gen two that was one of the areas where I was like ah there’s just so much missing there’s so many missing features and all of a sudden at Fabcon they announced

8:53 Like new compute engine better parallel processing better speed in the browser all these things that are like okay yes yes this is what we were asking for but that must have been just a really large rock to move and now we’re starting to see a lot more features come out quickly which are more creature comforts. Mhm. Refinements, , making the tool better and easier to use. Did I just I’m just an observation. Do you feel the same way? Let me ask that question first. Not only I feel the same way, I think we’ve had a part to do with that honestly, Mike. So,

9:25 Well, and I don’t think anyone from Microsoft really listens to us that intently, but No, not at all. Not at all. Not trust me, not at all. We’re happy that you’re here, whoever you are, Microsoft. But no, I think that your mom your mom’s still favorites of your mother. Hi mom. she still comments on the podcast. Hey mom. She’s probably listening to this one right now. what the sad thing? What would she say right now if she if she was listening to the episode at this point? Because you’re so funny thing. You’re so funny. And it’s like that doesn’t mean nothing for me thing. It’s like so and I love what you did with the backdrop. Did you Did Sarah

9:59 Do that? That’s all all that’s only Yeah. So that’s what she would say. So, I just get teased necessly from my family. They’re always like, , dad keeps talking about data and they just walk around the house just going data lake. Data lake. All he says is data lake. They just tease me. They are. They do. They mock me. That’s it. They mock the hand that feeds them. I don’t know what’s going on here. Well, I probably have an inflated head because like if they watch like Dude Perfect, that to them is TV, right? Yeah. Well, next to that suggestion is the podcast. Yeah, exactly. Oh, Dude Perfect explicit measures. not having any idea we’re not worth $750

10:34 Million. When you scroll through your feeds and you see our video, do you sit there and actually watch it the whole thing? The shorts the shorts I do. So, well, I do it more to say what can we do better? I’m always thinking on like, okay, that joke didn’t land. Let’s not go that way again thing. And just the topics, what topics work? But no, , listen, I’m not going to lie, I enjoy it. I enjoy the conversations and it helps me. And if it helps me, great thing. Helps you even better. So, but no, I I think going back to your point, man,

11:06 Like I I think what this whole copy job and the other news topics are all around conversations you and I have had. And if we’re them, I know others are having this. It feels like we’re getting unblocked. Yeah, 100% agree. And it’s good to know it’s not just great features that they’re updating. It’s like they’re hearing they’re hearing the community as a whole understand these are the deal breakers. These are the blocks. We’re not going to use it unless XY and Z is done. So whoever is leading that direction, kudos, absolute kudos to that with the

11:39 Copy job. The Gen Flow updates are incredible. Yes, agree. All right, let’s let’s go unpack this article then. So simplifying data ingestion with copy job. All right, talk me talk me through what you think here is interesting, Tommy. So it’s the biggest thing. It’s the copy job feature with the nice user interface. We’ve talked about that. But some of the major highlights here is CI/CD support, CDC capture data cap or change data capture support. We have new data source and destinations. So SAP Hannah folders Cassandra Oracle with Amazon RDS, Amazon Access. Okay. U but

12:16 There are some companies that still live on that one. Like okay, not not recommended, right? But even SF SFFT, FTP, MySQL, Oracle Cloud Storage. So these are and this is important. The reason I’m for me the biggest thing I’m focusing on is the source and destination because user interface is already great. But you if you’re going to get more and more people adapted, part of what PowerBI made it so rapidly adopted was out of the box on day one. I think I had over 130 data sources that were out of the box that

12:48 Were already packaged in. So, it made it incredibly easy. We’re not there with chain, , copy data. It’s I know it’s different, but you got to have these things supported or ways to support them. So, to me, that’s the huge win here. I love it. And, , these there’s a whole bunch of new mirroring items that are being added as well. So, that this is just again, all of this makes a ton of sense. I I really like where they’re going with this one. copy job is is the ability of like look there’s there’s a common pattern. I’m

13:21 Going to bring data over. Some of the data will change, some of the data will not make it efficient. Just just make it easy. And I think this is good. I like this feature. I think this is really going to be useful. also you’ll notice here there’s a lot of another note here around CACD report as well. So variable libraries is also something that’s incorporated here. A variable library I think is extremely powerful for organizations who are really trying to start building process around what they’re building inside fabric right I need to be able to have a dev a test a production environment and there are some things variables

13:53 Parameters options that have to change across those interfaces you have to be able to build in dev use these things called variable libraries and to incorporate parameter changes and then as you deploy between different environments the variable library can pick it up and switch out different connection strings, switch out different database names. Like that’s that’s the intent of it. So, I really like where they’re going with this. This makes a ton of sense to 100%, man. So, it’s really, dude, it’s so exciting to see that. And Mike, I’m I want to talk about these data flow

14:26 Articles that came out and I’m going to combine the two here, but there are two articles that came out from Microsoft on what they’ve done with data flows. And to your point, Mike, they are from the fabric Fabcon in Vienna. Yep. One is around data transformations and this called unlocking the next generations of data data transformers in gen two. The other one is titled around data destinations and experience improvements. And these are those articles that come out every so often that really do change I think people’s behavior or they’re going to

14:59 Really make people come back to features. And these are more than just oh cool we’ve added a folder or we’ve added some destinations. What they’ve really done here that I think for you and I and what has been our biggest agitation with dataf flows gen two is the cost is the cost and the speed and it’s like okay why would I use a dataf flow gen two if I want to spend more it’s going to take more time outside of the user interface it didn’t seem worth it to us on that cost benefit scale well they’ve done a ton of work here where there’s smarter pricing with the

15:32 Genflow Gen or Gen 2 CI/CD with a cost effective way based on query evaluation duration. So basically, if you have a query that runs for 20 minutes, you’re going to save 50% of cost. If you have one that runs for 10 minutes, 25% of cost. If you have one that runs an hour or around an hour, you’re going to save 80% of your cost that you normally would. Y holy crap, man. That’s absolutely incredible. So you’re just in 10 minutes, you’re going to save a quarter of what you normally would. They also run significantly faster. This

16:05 This does two things for me here. So, there’s a couple things that I think that are interesting here in this last article. This is the last one I put in here. This is the one that’s called unlock more performance with Gen Two. it’s interesting to me here if I look at this, , there is a premium to be paid for when you have like a clicky simple interface to work on things. I get it. I I understand, right? But to me, when I look at the pricing that they’re adjusting here, this really incentivize two things. This incentivizes one for users that have

16:39 Longunning data flows Gen 2 jobs. It’s going to incentivize Microsoft to get them done as quickly as possible from an incentive side because Microsoft is making the max amount of like talk about the volume of data or volume of money they’re making like in those first 10 minutes that window the pricing is a little bit better but not a whole lot better right it’s it’s basically the same so for Microsoft if they had a job that was typically running 30 minutes again talking about like the area under the curve it actually incentivizes Microsoft to then make that job run as soon as you hit 10-minute mark, they

17:12 Want that job to go fast, right? And that’s typically bigger data, larger experiences. So, it’s up to Microsoft to optimize and tune that whole side of things. And so, there’s probably a lot of things happening like cost-wise when you start up a job and spin up a VM and do some other calculations. So, if you look at like when you look at a copying job experience, there’s probably a lot of planning that happens at the beginning of the job. This is what Spark does. And then once that’s done, you can just like set the plan and let it run, right? And so there’s potentially something there why they’re doing that. But the graph that makes me really excited

17:44 Is the making data flows gen 2 significantly faster. this is like substantial amount of runtime here where they’re talking about there’s they talked about some things in different ways. They have dataf flows gen one dataf flow gen 2 no performance enhancements. Dataf flow gen 2 with the modern c query evaluator huge improvement. And then they have dataf flow gen 2 modern curry evaluator and the parallelized execution. This is like on bigger data though. So this is like talking about but we’re going from a huge amount of CU consumption

18:18 Down to a much smaller number. Yeah. This I like this is where we should have been like it’d be nice to have this day one but if I think about the scale of things right this whatever we’re building in data flow gen 2 should cost less than doing it in gen one. Period. End of story. Like there should there should be no more there should be no more conversation around that. That should be like a given. If I do things in other tools that have more code, I understand those tools might be slightly less expensive to run. But I got to do some testing on this. I got to figure out how this is going to work. How how actually rolls out to me as a business.

18:50 And what I really want to see is I want to see data flows gen one the most expensive. Dataf flows gen two cheaper, more efficient, closer to notebooks, right? And if I run the same thing in a notebook, there’s a lot of variables here that you can like tweak and dial and tune things so it becomes cheaper. , but I’d like to see it like, , 20 to 30% more than a notebook because it is providing a UI that’s experience that’s cleaner. Don’t charge me double because that that’s absurd. Like I’m not going to that’s too far away from like where we should be price point wise, but you can charge me a little bit more than other experiences

19:22 That are more code ccentric. The other things I like here too Mike to your point is they’re all there is also a concept here of have this better skill in this is going to lead to better performance. So some of the things is when you actually do your parallel query partitioning or the modern query the evaluation engine which you can enable turn off but there’s also an other concept around preview only steps which again you can enable which again this I am happy that they’re adding in features that are not just enabled by

19:54 Default that you need to know what you’re doing. Sure. And again because again you can’t idiot proof everything and I know that that’s a big push around this and you can’t you can’t. So the fact that there are a lot of options here that they’re allowing but not necessarily enabling by default also too for me as someone who’s going to either hire someone or work with someone know what you’re doing here which can reduce cost here as well. But there’s other features here. Mike, I just want to make sure that we mention. So there’s expanded output destinations. Lakehouse CSV

20:27 Files, Azure data lake storage, snowflake, which you can now do as a destination, SharePoint and Excel, which are coming soon. One lake catalog integration, there’s schema support as well. There’s better. So there’s AI powered experiences with the get data improvements, which they really focused on is how to get data in. So Mike, there are so many features here more than just the cost that really show me how committed Microsoft is to Gen 2. I would agree with that as well. I think I think there’s definitely like real

21:00 Money being put behind this, real effort being placed to make this an efficient use case. , one thing I just want to comment here on someone in the chat. So another gentleman in the chat who’s got a beautiful name by the way, Michael. Yes, great name. Love it. I thought it was going to be like Consiliano or something. No, it’s not Pulia. Sorry, there’s no one in the chat named Pulia. Sorry, Tommy. , so, so Michael makes a comment here. He goes, “Glad to see there’s improvements coming for dataf flows gen 2.” He works in a smaller organization. Isn’t quite big enough to really see savings from data flows gen 2, getting these performance

21:32 Improvements, but he goes for them, the right choice has been just work in notebooks. That that that saves them CU. And I I want to be very clear here, work with what your team works with best, right? And I would also argue here, especially to Michael’s point, I love the fact that you’re doing that. That’s exactly the right thing to do. We we should be using the tools at our disposal to make the most amount of data as efficiently and as low cost as possible. That that’s what we’re trying to do. We’re trying to do our job as the best we can quickly with the least amount of cost. That’s 100% what you should be doing. What I’ll also note

22:05 Here is I think there’s even like a concept around if you have a smaller team and you don’t have a lot of data engineering people, it’s it’s less of an investment for you to educate on okay, we’re just going to do notebooks because it’s efficient. Like that’s less of an investment for the time that the the level of effort that takes to educate that team. So because there’s less people to train, you can do it faster. It’s less cost to get people up to speed. when you’re in a larger organization or you’re really big and you’ve got a huge audience of people and

22:38 You’re constantly turnurning through like new people. Not to say that you shouldn’t be doing data flows or you shouldn’t be using notebooks. I’m just saying the level of investment at a larger organization to educate. You got to put some real money behind this and actually have a process to get people into notebooks and doing these new things over time. Let’s use a term that Microsoft uses a lot, particularly Christian Wade. In the fullness of time, whatever that means, in the fullness of time, , eventually everyone would ultimately be comfortable and already know how to do notebooks.

23:10 They’re already mature enough. Like it just becomes part of the world because it exists. It’s easy to use. Some people start learning it and it just spreads like a weed throughout the organization, right? So, I do think there’s something there. Oh jeez, now Alex P is showing up. He’s going to he’s going to correct me now. I can see. He knows that we’re going to talk about this telling him that we’re talking about Power Query. Does he have an alert and AI must have an alert? He’s got an AI agent listening to the Please listen to the explicit measures podcast. If they say data flows, let me know. I’m going to show up. So, , so I I I just want to make the counter point here of of just like look, if you’re if you’re in a smaller organization, choose what

23:42 Works best for you regardless. I think it makes sense. your investment to learn all these different tools or learning which tools are most efficient for you is is less of a barrier to entry which I would argue it’s like larger organizations that need more skills or scaling up more to move people over to those notebook experiences. So I thought that was a really neat comment. Loved your comment, Michael. Really really good idea there. I I love the idea of so many tools and that’s actually going to lead very well into our main topic for today. Yeah. Which is actually have we moved away from business intelligence and has

24:13 Fabric pushed PowerBI into something new that it wasn’t before. So Tommy maybe with that do you have any more comments on dataf flows gen 2? Do you want to go into the main topic now? I think we need to get to the main topic here. Yep. I think we need to we need to have there’s a conversation that there’s an elephant in the room and we need to talk about that elephant. Alex is exactly right. He said beep boop beep boop. alert alert. Data flows has been mentioned. I need to show up. So thanks Alex for joining. Awesome. All right, Tommy, give us the topic today. Is Microsoft Fabric even business intelligence anymore or has it

24:46 Mutated into something different? So yeah, and honestly, this is one of those that came from yours truly’s brain. , and I haven’t seen a lot of conversations around this, but to me, when I think of the normal definition of business intelligence, power BI, power business intelligence, and what the focus was there, and you look at the features that we’re talking about today, I think there needs to be a conversation. We really do need to have this conversation around does fabric count as business intelligence? It has

25:19 Components of it, sure. But I think about this from an individual’s point of view and a larger organization or the industry as a whole. If I’m working solely in all the features of fabric, am I calling myself a business intelligence specialist? More importantly, if an organization’s hiring or building a team to do fabric, are they looking for people who are in business intelligence? This is the critical moment, I think, for us on the podcast. I think for people listening and as you move on how you build your resume and how people are

25:52 Looking for resources in Microsoft’s data platform that I think we need to have a hard conversation here. We need to have a so to speak come come to the water moment. H this is interesting tell me. So, I have a couple what what I like to do is I like to try and find like similar analogies that that seem to make sense to me and other maybe more defined industries potentially that would maybe help me allude to like what’s happening here inside the the fabric world. I I would I would argue like

26:29 I try and think about like the manufacturing sector. Not an easy one. It’s not an easy one. I’m trying to think about like manufacturing, right? So let let’s let’s think of our data like a product like like you’re building a car or you’re building something complex, right? It takes a lot of different pieces from a lot of different places, right? I feel like this could be an analogy to like an assembly line to some degree, right? And we talk about data pipelines and like deployment pipelines, all these kinds of things. But if we think about data as a a product that we’re we’re outputting, the output is well-defined models, understood data

27:03 Sets, and reports that people can interact with and and get that into other tools to look around and poke in the data. Right? So that’s that is the product that we’re trying to produce. Now, if I go back to like the automotive analogy and like building a car and all these different things, when we build the assembly line of these cars, these products that we’re pushing out, we have to go get the raw materials. We have to go get , shape those. We have to refine them. We have to enhance them, build specific wiring harnesses, make the doors, put the wheels out. All these

27:35 Different there’s so many industries that are attached to like building a car. But there’s all these different, , specialties. these thing these people are you could build a tire a number of different ways you can build them in different countries like you can get them from wherever so I feel like that analogy somewhat fits here as well and would you would you take away the terminology of the the assembly line would you call it anything else other than like manufacturing so manufacturing is like the broader term but inside that term you have like breaking down of like segments

28:06 Or areas or specializations like we have like the supply chain Hey, I’m going to go get these particular materials and bring them to the assembly line at the right time. That’s part of your supply chain. So, there’s a whole team of people dedicated to doing that part of the process. And then there’s people on the line that are like bringing the doors together and assembling things and we’re programming robots and we’re doing all these really interesting things. It’s still part of the broader term called manufacturing, but now we’re talking about specializing in certain areas to make the whole process more efficient. Okay, that’s that’s where my

28:38 Head goes when I think about this analogy of like the data product. Okay, so let’s step back here a little bit and let’s maybe bring it back home to what PowerBI and and what PowerBI is doing. I think we’ve expanded I I would like to take the term I think we have greatly expanded what business intelligence is and I think I would probably step back and say we probably aren’t really building business intelligence anymore. We’re probably building more like a data platform. So I think and again the reason I say this is because when I

29:10 Looked at what we were at before PowerBI like before fabric and it was just PowerBI it was a lot of like if I needed really complex data engineering if I really needed like specialized things I was going to Azure for all that right it was it was in Azure it was in synapse it was something else I had to go build or buy in front of the PowerBI ecosystem. Fine so be it. It is what it is. But now I think what happened was for me the big moment was Microsoft embraced this delta format brought in a bunch of these open-source sparklike technologies continue to refine them

29:44 Making them more efficient and when those came into fabric it gave liberty to pull in all these other products real time analytics with custo databases data warehouses and so back to Donald’s got to give us a good hard time on this one right so Kimell talks about the data warehouse BI terms. There’s a data warehouse attached with the business intelligence. It’s it’s all got to fit together. Can you do business intelligence without all the other things? Probably. Will it be easier and faster?

30:15 Probably not. I would argue. So, one of one of the things I love about fabric is it just integrates so easily between all the different tools. And I’ll I’ll pause there. I said a lot of things. No. What’s your reaction, Tommy? There’s a few. I’ll I’ll briefly mention the one about the manufacturing one. The only thing I would disagree is you would be right in your anal at least with the definition of manufacturing in the components here. However, I think we would have to change the definition of what business intelligence is. Like I I agree with the literal definition of manufacturing in those components.

30:48 However, I actually agree with you with the data platform. I have I’m actually laughing because that’s what I’ve actually been using when I’ve been talking to clients and projects. I’ve been straying away. of saying no it’s the data platform that we’re working bigger than it’s bigger than just BI right and so that’s the the first note that I wanted to have the other note is we’re mentioning here a lot around business intelligence the warehousing and the data integration we’re expanding it but that to me is then taking away what 40% and I think the core function

31:24 Of someone in business business intelligence is when you think of the definition like my job ultimately if I’m specializing in business intelligence Mike is to measure what matters provide those measurements to an organization that they can act on that data now you can argue there’s a ton of components going in into that right we have to have the clean data we have to have that data integrated yada yada yada but the warehousing the ETL those are all parts of it but it’s really ultimately led

31:56 Just for really the reporting and it’s to provide that consumption and all these other aspects that we’re doing in fabric can lead to that. But again, why have we been saying since fabric’s come out that we’re combining six careers into one is because we’re not I don’t think we’re not saying that as a hot take. You have not junior versions of a lakehouse or diet versions of a warehouse no in fabric. So, what I wanted to do, Mike, is I’m going to

32:28 Invoke our old host co-host on the podcast who like to bring out a dictionary from time to time whenever there’s a definition. And but I’m going to take this a step further and I actually ran an agent to scour the REM on a definition around business intelligence. All right, sounds good. And perhaps this is actually a good place to start. So, business intelligence, what is it? It’s a comprehensive technologydriven framework that encompasses the strategies, methodologies, processes, and technologies used by organizations to collect, integrate, analyze, and present

33:02 Business data for informed business decisionmaking. At its core, BI transforms raw data from multiple internal and external sources into actionable insights that drive strategic, tactical, and operational business decisions. with that definition which I think is a to me a great definition here and again we we are more than welcome to expand or contract that definition but when I think of fabric there are parts of that that are fabric absolutely but

33:36 Mike I let’s go back to just look at our previous episodes what we’ve been talking around fabric I don’t think a lot of them are touching that I would agree and I think I agree with your topic and we’re not touching on those topics And I’m actually reading the chat here, so I’m kind there’s a lot of things going back and forth here. So, , Brad has appeared in our chat. So, Brad, welcome. Oh my gosh. Hey, Brad. He was on the on the podcast just a little while ago. Data warehousing. Yeah, the data warehousing. So, so I love that all these tools exist. And I and I think Tommy, what was to me when I look at

34:08 Like where we were and where we’re at now, what’s changed? And again, I’m going to go back to this idea of like before we had PowerBI, even go let’s go way back. Let’s go back before PowerBI existed. like we would do things in other tools. There were certain programs you had to buy as an organization. The price point might have been very high. You could do some levels of transformation. You had to wait on other teams to build all the stuff out, right? So, I I look at it going what we done for PowerBI, which is make the report building side of things, the the business intelligence, and actually Donald uses another term here that I

34:40 Really like, which is decision support, right? We’re doing decision support items. The reporting is helping makes decision supporting supporting those decisions that is become a commodity. We can make it so much easier. The tool you need to use is free. You can build and explore and create. So the same thing that we’re doing for the report which was make reporting more of a commodity. Again back to Jav’s paradox, right? If you if you make it easier to consume and lower cost to do it, everyone will just

35:13 Use it. it will just it’ll just become a given that everyone needs to use and they will be have they’ll need to use it to to move forward. That’s the same thing that it feels like occurring with me inside fabric. It’s the same idea. We’ve taken the data warehouse, we’ve taken this data warehouse concept and it’s becoming commoditized. It’s easier to do it. All the tools work together. The barrier to entry before was I had to get my data factory to work with the lakehouse to how do you get it into SQL and all the tools didn’t quite talk together very easily or if you did

35:47 You had to write a lot of code and just know what you were doing. Now it’s a lot simpler and so I think really to me this the sassification software as a service on top of all business intelligence really works for me and I think that’s that’s where we’re moving to. I would expand the term. We’re still doing business intelligence, but we’re expanding it also to data platform. I I’m going to disagree with you here because I’m going to disagree specifically on the point around it’s easier to do the reporting. It’s a commodity. Mike, did Microsoft Fabric make reporting easier? Did it make

36:21 Building visuals and the right visuals easier? Did Microsoft Fabric make those easier? No. That’s that’s been PowerBI’s job. Exactly. But it’s not just PowerBI’s job, but it’s still the user’s job, the creator’s job to be skilled at that. But what did you do before though? Like I would argue like what was it? It was it was Excel and more effort to shape and lock in business data. You’re saying before PowerBI, I’m talking the whole progression here. Like I’m looking at like the whole story and saying if I see what the story did

36:53 For the first 10 years with PowerBI and now I’m looking at fabric for the next 10 years, I’m seeing the same pattern. Like I’m seeing it’s the same rinse, wash and repeat. Yeah. Yeah. Yeah. So Microsoft that what PowerBI did when you when you think about that the jump from Excel to PowerBI, you took people who were much more in the business and then you actually really created careers with PowerBI. There were not as many if you were either working in Tableau or you worked in Excel and a bunch of other things, right? Those were the distinction. You didn’t hire an Excel data analyst

37:25 Specialist. That was not someone what people hired for because you worked in a ton of other things. PowerBI created careers. It created career paths. For us, Mike, , if PowerBI didn’t come out, we wouldn’t be doing an Excel user group. Mhm. We wouldn’t it. That’s just the case. PowerBI created a platform around a single career point of view, the business intelligence role, and it basically took Tableau out of the water. It it basically destroyed the stock in so many words. But there’s a reason I think Salesforce I think there’s other factors. Here’s the thing. They started

37:58 Getting desperate. They were getting desperate at that point. But but there’s a reason for that because again Microsoft created literal careers, teams and departments that did not exist before PowerBI. That’s just the fact of it. And the the thing is with fabric that jump, it’s not making me better at my skills around reporting and creating the measurements and doing DAX and semant semantic models. It didn’t make it easier. I still need those skills.

38:30 And those are still skills if I’m hiring a business intelligence specialist that if they know fabric, great. And they do data integration, great. But if they don’t know DAX and they’re no, I almost used a word there. If they are subpar on data visualization theory and providing the right context and talking to the business then it doesn’t matter if they’re great at fabric and that’s where that the I think the the divide is here now there there is an interdependent relationship between data engineering data science and business

39:01 Intelligence that obviously but when if you’re hiring Mike someone I need a business intelligence specialist and 80% of their resume is around fabric and they do lakeouses and notebooks and they only have maybe six months of PowerBI. That’s not what you’re looking for. , for me, I’m like, okay, well, that’s not a business intelligence specialist. The only thing I’ll say I’ll jump I’ll run back to you here or hit the ball back to you is part of me also thinks, are we even looking for that

39:34 Now? And this is where I’m I don’t want to say struggling with, but this is where I’m asking a lot of questions. for example, , as a consultant trying to you got a lot of messaging. I work with recruiters to do, , part-time gigs and I’ve had to change my own tune. I’m changing the instrument that I’m using on how I’m marketing myself from I’m the PowerBI guy. I’ve been doing PowerBI. That’s a lot of things on the resume where it’s like I got to talk about fabric. And I think it’s also changing everyone’s perception around what you’re looking for. are

40:07 Organizations going to hire just a business intelligence specialist too? So, there’s a few things there, but what I’m going to kick it back to you here, Mike, is if you’re hiring from a business intelligence point of view, are you looking for the same skills that you were three years ago? Or is is really just the game has changed again where business intelligence is a part of it, but now there’s a a greater game here. I like what you’re I I like this question, Tommy. This is really good.

40:38 And I think this is really we’re trying to unpack here, , as you add on more skills. So, let’s let’s talk about this the monetization of things. So, let’s just talk about that part because I think this I think this is where this begins to become more of the topic at this point, right? So, you’re maybe touching on what I think are like pieces of this and I’m going to try maybe pull some pieces together to kind like really consolidate this one. So as as if you talk about okay so let let’s talk about report building or decision support right in a in a again let’s go really

41:12 Far back here before we had PowerBI it was I’m going to be given a tool like business objects or crystal reports or something that’s out there. They’re just giving it to me. It’s Oracle something has their their own reporting whatever you’re you’re forced to go to this UI. The model’s already built for you. There’s no data engineering. It’s basically here’s the data and you have to do whatever you have to do to get that data out into something else. Typically that ran into the tools were limiting. I couldn’t build what I wanted to build and therefore I would print it, take it into an Excel document, shape the data there and then build my information on top of that. So if that

41:45 Was the way we were operating back in the day, what this has done is when you commoditize something, it gives more access to whatever that person can do. So the roles of that person that analyst expanded and so now we have the the report engineer or the report builder and so what we do now is okay well there was some data modeling that was being done by a different team data engineering likely but now we bring that forward and say okay well we we’ve made data modeling so easy with DAX and and models and reports you can just use that

42:18 In a PBX file. So again, make a commodity, simplify it, make it easier. The barrier for someone to learn those things are is now easier for you to get through. I don’t have to learn all of SQL to start making effective reports. Okay, the same thing’s happening now again. So as you continue to simplify these roles into the tools and the tools handle more of the leg work for you, it now becomes easier for individuals to do more in their current role. And so what I see happening here is the

42:50 Commoditization, the simplification, building better tools around this, the the barrier to entry to have someone know more and do more is getting less and less. So I see this as being like a consolidation a traditional if you’re thinking about reporting and business analysts as you would back in the day. I think we’re and even now let’s talk about what large language models and AI is doing. That’s going to further change this even more. Like I’m I’m even going to need to know less code to write SQL because now we have like again this is what co-pilot on home is doing. Hey tell me about my

43:23 Sales. It spits out a table and writes the SQL for me based on the relationships or writes the DAX for me inside the model. So as we continue to commoditize more and more of these things more of these technical aspects are going to be there but obiscated away from the user. And so I what I see happening here is there’s like a convergence of technology and people building things here that is going to continually reducing the barrier for people to know more and know more. Do do

43:56 We have a data engineer expert in the same way we used to a couple years ago? Probably not. We’re probably not writing the same language. We’re probably not writing as much code. , can if you if you’re not writing as much code and you’re doing it more with UI based tools, well, what’s the barrier to entry for a business analyst just to learn a little bit more about modeling, learn a little bit more about data engineering, learn a little bit more about data warehousing and lakehouses. So what I think you’re seeing here Tommy is I think that the industry is changing rapidly and what we’re doing is there’s still segmented roles here but

44:30 Before it was a lot more siloed because the barrier for one person to jump between market industries analyst to data engineer to data scientist was so great. Now we’re looking at this and going okay we’re now pushing these silos much closer together. And if you’re really good at being an analyst, you might also be good at a modeler. If you’re really good at modeling, you might also be really good at data engineering because the the the the skills needed are being condensed and and they’re still uniquely different, but they’re the pillars are moving close

45:02 Together. This is how I mentally look at this. So, you make some very profound points here, Mike, and I know that again, as you normally would, but I think there’s something really important to what you’re saying. And as I’m thinking about this shift here, it’s it’s funny because what PowerBI did for people working in data was bring them to the forefront. What I’m referring to is the majority of what I created and did was going to see by be seen by consumers, by the organization.

45:35 I was front-facing, not just in meetings with people, but what I actually created was a report that someone was going to see. But fabric’s almost pulling us back into that slight IT world because a lot of what we’re creating is not going to be seen or manifested by normal consumers. You’re still going to create reports, but lake houses, notebooks, the warehouse, that data integration side, like that’s not what a normal user is going to see. And I think this is very profound for us, Mike,

46:08 Because part of I think when we consider business intelligence in our own careers was we were that liaison like I could not hide from the consu the consumer because everything every visual I created was going to be seen by a wider audience which is now different. And I also think about that in the same fashion in the same token Mike that a lot of the projects I’ve been doing on fabric I think about the lowhanging fruit. What do people want to get out of the project, , in the first few months to know that we’re making

46:39 Progress or the wins? Well, if it was a pure PowerBI project, they’re going to want to see a dashboard. They’re going to want to see a report or something in that aspect. But the projects I’ve worked on fabric, Mike, the the lowhanging fruit is storing the data in Lighthouse, and it hasn’t been the focus as much on a dashboard. The part of that is also an understanding that the data needs to get worked on. But I think this is very different because my lowhanging fruit now at least from my experience is no longer consumerf facing to the wider

47:13 Audience and I think this is really important when we think about our role here. I I agree Tommy but I think this is the the point here though is if you weren’t doing that by bringing it to the lakehouse and doing transformations in notebooks or other things you were doing them in power query it was just built into the tool that you were using before. All we’re doing is, and I think what we’re doing here realistically is we’re building what we should have been building all along, but PowerBI came out with such guns ablazing and just like let’s just build a semantic model and we build this one semantic model and we try and make it the monolithic semantic model for everything. That’s not really

47:45 What we needed. What we needed was a common place to store all the data, an easy way to get very different data sources together, , load them, incrementally load them, , handle. So there there is this it the principles do not change here. Fabric does not change the principles. There is a transactional world of things that are happening. There is an online analytical processing. They call this OATP or OAP or whatever the heck the name is. I don’t really like I don’t it doesn’t matter to me. Like there’s two worlds of like there’s the operational database

48:17 And then there’s the reporting side of things. And those two worlds don’t like to be summarized the same way. And you hear about this in school. I learned about this in my masters in data science. They’re talking about, oh yeah, this system does it this way. It’s like row based and this system likes to do aggregations. It’s different. So the the fact that there are two different ways of looking at the same information, I think is really why this all all this stuff stands up, right? Yeah. And and this is the core of what we’re trying to work on is okay, how can we get the best information from these operational systems into our reporting

48:50 Systems? And this is where Kimble comes into play. star schema efficient storage of data and all of this is now being like shaped into again what I’d like to call is it’s the data platform and a term that I would use a lot was like the modern data platform like let’s use spark let’s use data bricks it’s the modern data platform I think now I’m just going to call it it’s the data platform you need it’s just essential at this point okay so off of that point from that definition I think there’s I have a few questions for you I think that will further clarify what we’re trying to

49:22 Achieve here for an individual’s point of view and I will say whether you’re consultant or whether you’re FTE, what’s the smarter career bet? Do you market yourself as a business intelligence specialist or someone who’s a a data platform specialist like in the Microsoft data platform? Yep. Yep. I me personally again one thing that we’ve done in our company specifically is it’s hard to get data shaped and formed, right? So the you need people that can be able to understand where data comes from those

49:56 Operational systems and you need to be able to figure out how to get it from that thing into visuals consumable products a reports powerb reports something on the other side the consumable side and I I liken the analogy of like you’re building a bridge you build a bridge from both directions on one side you take the business needs the questions the things they’re looking to answer and you start building out what visuals would help you? What things would make it easier for you to do your job? What things keep you up at night? Like com, you ask that question all the time. I I pulled that one right from you. What’s

50:28 The thing that in the middle of the night you wake up thinking about your company or your data? Like that makes a lot of sense. And then the other side of things is okay, what data do I have to support that? So just because you’re asking the question doesn’t necessarily mean the data you’re going to give it is going to support it. And so sometimes you try and bring those two together and the two bridges miss, right? And so then you have to realign expectations. you have to either change your requirements on the report side or you have to go find additional data so that way you actually meet in the middle the data I’m bringing in actually answers the questions I’m trying to deliver. So I think that that to me the and this is a mantra that I think is

51:03 Also going to happen in AI as well with programming. We are going to find the senior and more seasoned developers are going to become more effective. They’re going to be able to write more code build more apps quickly because they can basically project manage the the AI agents to go build the code that they want. Right? So, all those things are like really really important. Thanks, Donald, for the for the the drink on us tonight. I appreciate the the sponsorship there. Appreciate it for throwing something out there. So, but

51:36 That’s where that’s where we’re at, though. I think I think we’re getting to this place where you can’t throw one of these things out without the other. I really agree with Donald’s point he made earlier, like look what we really let’s just talk about what we really want. Wherever the data comes from, however fast it gets there, we want all the data to be in this massive database. whatever whatever you want to put in. I don’t care what you put it in. It needs to be this massive database with all the data we’d ever care about. And then we need to be able to pluck at that data and pull out what we need when we need it. That’s really what we’re doing here. And we’re trying to find patterns in how we bring the data in and patterns on

52:10 What questions are being asked out so we can build efficient, cost-effective ways of doing this. Like if you had an unlimited budget, no money, you just would suck in all the data and let people just build whatever the heck they want on top of the system. But we’re always constrained by this time and money thing that continues to like have us like pair down, scale back, build only what’s essential. That’s the balancing force of like just throw everything into like one massive database. So this this is the reasons I love these conversations. I I’m I’m going to expand on that. I have a few questions for you. But I’ve been going

52:42 Through this conversation today and as I’ve been thinking about this question in general is is data engineering eating business intelligence in terms of definition the perception that people have and that’s where I’ve been thinking before we’ve talked today is the term business intelligence is going to in sense sell the stock if from the definition point of view if you look at resumes but I think the way that we’re talking about this Mike it’s the other way around. Mhm. I don’t know if the industry is going to adapt that, but to me, I’m looking at this and what we’re talking about based

53:14 On the constraints that we put. To me, business intelligence is going to eat the definition of data engineering and data science because I know we’re saying data platform here, but really we’re it sounds more like we’re really just expanding what business intelligence is. The problem is going to be whether or not the industry as a whole and organizations understand that difference, right? Because to your point, I’m considering this now as we talk about this that business intelligence has just expanded. It has grown and evolved. That’s probably the

53:47 Better word. It’s evolved now into these aspects where data engineering is a part of business intelligence. Now, the problem is whether or not that gets adopted. Like that’s my definition. But hey, at the end of the day, Mike, we got to market ourselves the way people are writing the resumes and the job descriptions, right? Like so. And that’s going to be the problem here. So my like whoever is listening, if you have are in charge of definitions and things like that, call it fabric BI. I don’t care. But this has always been my problem with our industry as a whole anyways is we don’t have like, hey, you’re a business

54:20 Intelligence specialist level one, level two. I could find eight different titles with the same job descriptions or eight the same titles that have all different job descriptions. We’re always going to run into this. We’re adding more confusion to it. But for me, Mike, I’m looking at this and what I want to start marketing or I want to start pushing is fabric is an evolution of business intelligence. It’s not necessarily now all these other we’re six careers. This to me is just the new game of BI. So I don’t know do you agree with that?

54:54 Do you think that the term business intelligence is going to expand or in a sense shrink in terms of when you look at job descriptions and what people are looking for? Yeah, this is going to be interesting to see how all of these new tools are affecting what companies are looking for. Tommy, you and I were looking at PowerBI and all of a sudden it was like build reports. Okay, that quick that quickly got outpaced by build reports and model a bunch of data. Now we’re going to start seeing build reports, model data, and do all this fabric stuff as well. And it’s all going to be like the business analyst person, right? , if you if you hire a

55:28 Really sharp person, you can get away with one person who knows how to do it all. , we’re getting to a place where you could have enough projects. It’s been the technology has been out long enough now. You might have had a real-time analytics project. You might have had a data warehousing project. you might like. So the the longer this goes, the more time people have had an ability to go use all the different products of Microsoft, right? So I think to me what I think is going to happen is companies are going to want it all and pay as little as they

56:00 Can, not just how companies are. , is that new? Not new. This is not new. This is this does not change. So what I think is going to happen is the scope of knowledge is going to continue to increase for that business analyst or that that business developer of reports and so again let’s also add to this AI let’s add let’s add in the flavor of that what how does that change this as well so are you good this this I think goes across all industries all different disciplines hey Tommy I’m going to hire you for my

56:32 Company how do you use AI to make your job more effective and faster it doesn’t matter like this is a gamechanging experience for everyone in every industry across every place of the of the of the data space I think because now you can be like well I’ve I’ve been learning prompting and this is how I prompt this is the patterns I use for prompting and how it helps me is it lets me do all these extra experiences again the point here is you need to be at a level where you can prompt the right terms to get the right desired output Mhm.

57:05 Who car? , I people care. Say it. Say it. Say it. I’m going to say it like it is, but I’m going to I’m going to maybe soften the blow a little bit, but who cares if you really know SQL to the nth degree? The AI agent that I pay 50 bucks or 100 bucks a month to knows way more SQL than I’ll ever know. And this is my point. My point has been like, if I want to if I need to take this table of data and shape it from this thing to that thing, I could say here’s the starting definition of the table. Here’s the ending definition table. Do this in M. Boom. AI knows how to do it. Do this in

57:37 SQL. Boom. AI knows how to do it. Do it in Python. Boom. SQL knows how to do it. I can even take great ex great code from one area and say, “Hey, I have this uhnet script, whatever the, , or I have this PowerShell script. Turn this into Python.” And it will know it’ll understand it, flip it around, and output the same thing. So this this the skill over all other skills that is the skill that I think we all must learn because for the the barrier or price to have an unlimited

58:12 Code knowledge study buddy with you basically at all times like this is like when I was in high school and they’re like you’re never going to have a calculator on you at all times. You got to learn your multiplication. I’m like well I wasn’t learning multiplication. I told my kids that. Yeah. But it doesn’t matter. Like I’ve got two calculators on me all the time. Like I have one on my watch and I have one in my pocket no matter what. I never am around anything that doesn’t have a calculator on me anymore. So there’s the technology pieces are getting so different now. And I think the new world here is AI is here to stay. It’s not

58:46 Going to take our jobs, but it’s going to make your job change. And if you’re not willing to learn how to like use it to leverage that inside what you do daily, the other person on the hiring list will get your job. And not only will they get your job, they’re going to be able to be more effective at the job than you because they’re going to be able to say, “Well, it doesn’t matter if it’s equal. I can tell I can ask the AI to explain it to me and I can understand what’s going on.” the the core pieces of this what skills are going to matter in the next 10 years. It’s going to be

59:20 Communication, how well can you communicate? It’s going to be how well can you build requirements for AI things, agents, whatever. And it’s going to be how well can you debug the AI. I think there’s going to be a whole realm of like crap AI code and things that have been generated. And there’s going to be there’s already resumes out there. They’re like, , the AI cleanup job, right? We built an app. It was all built by AI. It’s a bunch of trash. Okay. Well, now you got to go back through and like like, okay, we understand how the prototype works. We’ve got to reshape it

59:52 Into something that’s actually functional and can be maintained. So, there’s there’s a whole new world that’s just at our fingertips. And this whole data engineering space is going to be drastically changed by AI and how you build these data pipes. Anyways, yeah. And sorry, I went way off the rails on that one. I I loved every second of that. And I’m I’m just gonna briefly say the AI thing because I’ve said a ton on AI and I’m gonna keep saying things, but I’m let me I’m going to sum up what you said to really this if you’re listening. If I’m in an interview with you and I’m

60:25 Interviewing you and I say, “How do you use AI in your day-to-day?” They go, “No, I don’t really use it.” My next words are, “Thank you for your time.” Yep. I’m done. Like honestly, I’m at that point now, Mike, where I’m realizing the efficiency. They say for consultants it should save 20% of time easy and they’re finding that already and I’m seeing that in my day-to-day both from soft skills like helping from documentation to the coding and there’s so many aspects of this and again good job Microsoft because they’re putting a lot of investment into co-pilot and the last few weeks Mike is the first time I’ve

60:58 Been impressed with co-pilot in in fabric where I’m going to take time I see you fabric like the first time instead of going forget it because And you’re not going to if you’re doing AI, you’re either in love with the tooling or you’re not going to use it. There’s no in between. And I think finally we’re seeing that with co-pilot. And so I want to mention that I dude, we’re already near time. what I’m realizing now when we think about this conversation the AI data engineering data science

61:31 I’ve always said and I’m to this point that these are six careers combined into one but what I’m hearing today and I think I’m coming to the conclusion Mike that this is just an endto-end analytics solution there are components of data engineering but if I have co-pilot and honestly the ease of getting something started started because again half the battle with data engineering was the resources and the architecture of it in Azure which was a pain in the whatever it was thing. I want to say it because it deserves that pain in

62:02 The boondock pain in the boondocks thing. So but but to your point Mike like a business can create a lakehouse and do an medallion approach. They’re not a data engineer. They’re doing those components. I think the terminology is going to change to what from what we think that hey you’re a data scientist you’re a data engineer I think that’s going to be eaten up by the in a sense an analytics architect or something like that that is where we see this going part of that is the ease of use part of

62:34 That is AI being integrated into this because again Microsoft did something with fabric and I listen if you’re listening you’re like wow they’re really tooting Microsoft’s horn go listen other episodes where there’s other AI agents that are way better than copilot which I will say claude is by far and the best the the tool that I’m using a lot best coding one but this is the point though Microsoft is still playing catch-up and even even they’re still catching up they’re still producing good stuff now like right and and I’m saying more than just the AI I’m saying the solution if you think that we are just

63:08 Praising Microsoft for all the things they’ve done go listen to our copy job one go listen to everything else where we just like we are vehemently disagree with the direction with some things. But I’m going to give the credit. I’m going to give the flowers where they’re due here. Mike Fabric and I’m realizing this more and I’m coming to the conclusion. I’m seeing the light here where the way they changed careers and disrupted what a job looked like in data when PowerBI came out. Agree. We’re going to see that. And I don’t think we’ve seen it yet. I don’t think we’ve seen the impact of that yet.

63:42 Correct. And it it’s we’re in that I I honestly think we’re in a very the beginning of the disruption phase. So to answer the question, is Microsoft business intelligence? Yes, it is. I But what’s changing is the definition of business intelligence. I like that. I think I’ll go with that. I agree. All right, let’s wrap it here. This has been a good episode. it seems like every episode at some point just falls into like the pit of AI and we just talk about AI by the end of the episode. I don’t know. It’s such it’s such top of mind and it’s

64:15 So pervasive in this space. It’s very exciting. It’s also changing very fast and so I really like I like being able to unpack what where AI fits in these spaces and what do we like again how would you hire? I think a lot of these questions are if you’re a hiring manager, if you’re looking for skills in your team, like this is stuff that’s changing. It’s it’s dynamic right now. So very very interesting. That being said, thank you all very much. If you like this podcast, if you found some value from this, would you please consider becoming a member? We’ve got a membership thing going now and we are releasing all the episodes with no

64:47 Advertisements. If you’d like to come back later, watch them on YouTube. All of our new ads and you’ll start seeing more of our content come out with no ads, that will be all in the members area. So, , if you like the content that we have, if you want to get access to that content without any advertisements, please consider sponsoring down below. We’d really appreciate your support here so we can do more things with the podcast. We’d like to do more with this and continue to grow this as well. That being said, Tommy, where else can you find the podcast? You can find us in Apple, Spotify, wherever you get your podcast. And please subscribe, leave a rating. It helps us out a ton. And share with your

65:20 Friends, share on the YouTubes and the LinkedIns and the Twitters and whatever you want to do because it really means a lot to us to see what people think about what we’re doing. If you actually disagree with us today if you have another question on what we’re talking about or just something in general around Microsoft’s data platform leave us a question and you can go to powerba.tipsodcast. Please leave your name. We want to give you a shout out and a great question. And finally, join us live every Tuesday and Thursday 7:30 a.m. Central on all of PowerBI tips social media channels. I’m

65:53 Going to say you’re going to want to stay tuned with the next three months of podcast October, November, December. We’ve got some incredible guests teed up. It’s going to be extremely exciting. So, if you if you’re not paying attention already, please make sure that you subscribe. Hit the bell button down below so when we’re going to be making more episodes. We’re here every Tuesday, Thursday, 7:30 a.m. Rain or shine, basement or no basement, we will be here. we’re going to do the best I can to keep this going every single week. and keep providing you really good discussion content around things and also bringing experts to this as well. we’re not the end- all bealls

66:25 On all this knowledge. There’s a lot of other people who know great things as well and we’re going to continue bringing experts to this space to continue talking and unpacking fabric data platform and all these things. Thank you all so much. We appreciate your listenership and we’ll see you next time. out.

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.