Raw Data By P3 Adaptive
The Dutchman of Data Dunked on Me, w/ Kasper de Jonge
Analytics Advisor and Principal Program Manager-Power BI, MicrosoftListen Now:
Kasper de Jonge’s areas of expertise are many-Business Intelligence, Power BI, DAX, PowerPivot, General BI, Datawarehouse, SSIS, Reporting Services, Analysis Services, SQL Server, and you can include a mean basketball crossover to that list!
As Analytics Advisor and Principal Program Manager of Power BI, he knows a great deal about the problems that BI customers face daily. He shares some of these experiences and his expertise, and we have some good laughs along the way as well!
Check out Kasper’s awesome blog Kasper on BI
References In This Episode:
- 2:45 – A basketball challenge in New Orleans, Kasper’s MS history with Power Pivot and Power BI, and the culture change at Microsoft
- 15:45 – The culture shock in moving to the United States
- 28:50 – The death of Power BI V1, and some James Phillips stories
- 40:00 – Office VS the Freedom to Innovate, Excel/Power BI Integration, and how software development has evolved
- 51:15 – The Amazing Power BI Cat Team, and the processes that they use to solve problems
- 1:05:50 – Some bad Microsoft product names, DAX VS MDX, and the many different people that are in IT
Rob Collie (00:00:00): Hello, friends. This week's guest is Kasper, Kasper DJ. I call him Kasper DJ here because I'm completely unable to pronounce his last name. But anyone that's been following this community for a while, if I just say Kasper, you know who I'm talking about, don't you?
Rob Collie (00:00:15): Now, I met Kasper over the internet in 2010. This was shortly after I had relocated from Seattle to Cleveland for family reasons. And during 2010 is when I started to say something I thought was pretty funny at the time which is I have more friends in foreign countries than I do in the city I live. It was a very strange departure to suddenly be part of an international community on the outside of Microsoft all sort of coalescing around this thing called Power Pivot.
Rob Collie (00:00:43): It was right around then in early 2010 about the same time that I was getting to know Kasper that I had seen something in Power Pivot that told me that the world was going to change. And most people that I talked to about that I told that Power Pivot which became Power BI was going to revolutionize the BI industry and change the way it was staffed, changed the way that the projects were, changed the whole business model.
Rob Collie (00:01:05): The overwhelming majority of people that I said that to would immediately reject it and almost like attacked me. They told me I was wrong. Those days are over so long ago that it's hard to even really kind of remember what that was like. But Kasper was really striking back then. He had been part of the traditional BI ecosystem. And he was good at it. And he was seeing the same things in Power Pivot in those early days that I was.
Rob Collie (00:01:27): And that was like instant international nerd bonding. He's gone on to do some amazing things. He ended up working on Power Pivot/Power BI at Microsoft for a greater number of years than I did when the dust settled. Today, he's still at Microsoft. He's now part of the Power BI CAT team. And he's a deservedly well-known and well-respected member of the Power BI and Power Platform Community at large.
Rob Collie (00:01:50): Unlike me, Kasper had a ringside seat during some very, very transformative years at Microsoft. When the culture of the software teams was being completely remade around continuous release rather than the two-year, three-year waterfall cycles, he was there for the transition to Satya and the cultural changes that brought and the strategy pivot from Power Pivot to Power BI.
Rob Collie (00:02:15): We talked about all those things and more. We had a really good time. I hope you would enjoy it as well. So, let's get into it.
Announcer (00:02:23): Ladies and gentlemen, may I have your attention, please?
Announcer (00:02:27): This is the Raw Data by P3 Podcast, with your host, Rob Collie and your co-host, Thomas LaRock. Find out what the experts at P3 can do for your business. Go to powerpivotpro.com. Raw Data by P3 is data with the human element.
Rob Collie (00:02:45): Welcome to the show, Kasper De Jonge. How do you do?
Kasper De Jonge (00:02:48): That's pretty good.
Rob Collie (00:02:49): No. You're just being nice.
Kasper De Jonge (00:02:51): I am. I am.
Thomas LaRock (00:02:52): Yeah. He is being nice. That was awful.
Kasper De Jonge (00:02:54): It's unpronounceable if you're not Dutch.
Rob Collie (00:02:58): The American patois isn't really able to absorb something so sophisticated. So-
Kasper De Jonge (00:03:04): I wouldn't say that. But in the Second World War, we had passwords between each other that only Dutch people can say. We would ask someone to say [foreign language 00:03:13]. It's like it's a city, and no one could pronounce it. So, if you were German, you couldn't pronounce it. So, that's how you would know that you were not Dutch.
Thomas LaRock (00:03:21): Wow.
Rob Collie (00:03:23): It's like that scene in Glorious Bastards where he's asking Brad Pitt to say the thing in Italian over and over again [crosstalk 00:03:29] buon giorno.
Kasper De Jonge (00:03:36): Great movie.
Rob Collie (00:03:39): Yeah. There was also in World War II, the United States used the Navajo Indians with their code talkers, just absolutely unbreakable code that happen to be a language.
Kasper De Jonge (00:03:50): I actually know some Navajo.
Rob Collie (00:03:52): Do you?
Kasper De Jonge (00:03:52): Kasper [Yinicia 00:03:54]. It means Kasper is my name in Navajo. I don't know why I know it. But-
Rob Collie (00:03:58): Well, it's because of the Navajo population in Holland clearly.
Kasper De Jonge (00:04:02): Probably all about the same.
Rob Collie (00:04:03): Yeah. So, believe it or not, when I met you in person for the first time was the same trip where I met Tom-
Kasper De Jonge (00:04:11): Really?
Rob Collie (00:04:12): ... in person. It was that same New Orleans conference.
Thomas LaRock (00:04:17): Did you accost Kasper like you did to me?
Rob Collie (00:04:21): No. See, I already internet knew Kasper before we met. We'd even arranged a basketball duel to happen.
Kasper De Jonge (00:04:29): I didn't want to bring it up, Rob. But-
Rob Collie (00:04:31): Go ahead. Go ahead and bring it up.
Kasper De Jonge (00:04:33): No, no, no.
Rob Collie (00:04:34): We challenged each other to a game of one-on-one basketball over the internet to be played in New Orleans when we got together. We found a gym. We paid for like a day pass to a gym. And we went in there.
Rob Collie (00:04:45): And the way I remember it is that I am on a lifetime one game winning streak against Kasper. But he annihilated me. He annihilated me in the first few games. It wasn't close. He even did that move where he triples the basketball between your legs. Oh my god. So humiliating. But I quitted myself well at the end. For some reason, I managed to find my second win and find my shot.
Rob Collie (00:05:12): I regained some measure of respect. I was able to walk off the court feeling like I was still okay.
Kasper De Jonge (00:05:18): And you told me about hurricanes.
Rob Collie (00:05:19): Oh, my god. Our livers aged like a decade on that trip. Every morning, I wake up going, "Oh, what do we do? Why do we do that? We're not going to do that today." And then, the next day. This was 10 years ago. Holy cow, it was 10 and a half years ago. Really interesting.
Rob Collie (00:05:36): I remember in the course of talking to you Kasper in New Orleans over hurricanes, et cetera, watching sort of this idea come over your face over the course of that week, like, "Oh, my god I could move to Seattle. I could do the job that you used to do, Rob."
Kasper De Jonge (00:05:52): Oh, yeah. I mean I never thought about it. But you introduced me to a lot of people at TK and all the people from the [inaudible 00:05:58] services team and Power BI and, yeah. It was definitely a good time. And I'd never thought about like I was a consultant doing multi-dimensional models and reporting services models for years and building data warehouses and all of these things. But maybe I'm going a bit ahead of it.
Kasper De Jonge (00:06:13): When we started talking about Power Pivot, I don't know something happened in my brain that just I got obsessed by it. I was blogging and started blogging and started doing presentations about it too. That's one of the first [inaudible 00:06:27] common from the Netherlands, SQL PaaS when we had SQL Saturdays there. He said, "You have to come present."
Kasper De Jonge (00:06:32): And, yeah, I just was obsessed by it. And I loved our conversations and talked about DAX. And that trip was amazing.
Rob Collie (00:06:39): To give you your due here, I really want to make sure that I say this. For someone who had come from what I would call the traditional BI world, you are building multi-dimensional cubes. You were writing MDX, the feared MDX that I could never write. I really just couldn't learn it.
Rob Collie (00:06:57): For someone who is immersed in that world and essentially validated by that world, you were good at that stuff. For you to see Power Pivot and to embrace it as something cool and with something with potential and value rather than being afraid of it, it wasn't a threat to you. But it was a threat to a lot of people with your background. So, you stood out to me. And, yeah, we really bonded over the internet over Power Pivot, over DAX, over this awesome new thing. And then, it gets a little hazy. And then, I beat you in a basketball game. And that's all I remember.
Kasper De Jonge (00:07:34): Exactly. I do remember people telling me afterwards a couple of years after I started to work at Microsoft is there was a lot of debate whether or not they would actually hire me.
Rob Collie (00:07:44): Really?
Kasper De Jonge (00:07:44): I heard. Yes.
Rob Collie (00:07:45): I'll tell you that years later.
Kasper De Jonge (00:07:46): They tell me yesterday exactly. There was a lot of debate about them hiring me, and, yeah.
Rob Collie (00:07:51): Well, I mean you were filling an awfully big pair of shoes there.
Kasper De Jonge (00:07:55): Exactly. Yes.
Rob Collie (00:07:56): I mean you actually filled the head count that I had vacated. I mean I think you knew. From a distance, it'd be hard for someone to know whether you would feel like sort of like one of that team until you meet some members of that team. And meeting, I mean we can even just make a joke about it like after you met me, you're like, "Oh, my god. I can totally do what that guy used to do." But, yeah, you did. You get to meet a bunch of people on that team. And I remember being very, very, very excited about that.
Rob Collie (00:08:23): When that thought first became sort of acknowledged across the table or whatever day it was at that conference , I was like, "Oh, totally yeah. You should totally do that." That sounds like such a good idea. I'm so glad that came to pass. That's really awesome.
Thomas LaRock (00:08:38): After all that, can you tell me what it is you do there?
Kasper De Jonge (00:08:41): Well, things have changed significantly over the years. But when I just moved to Microsoft in Redmond, I was a program manager on the then analyst services theme/Power Pivot team. And I was working on some of the UI aspects of it like building the product there and shipping it into Excel as an add-in. And then, later, Office 15 shipping it as part of the Office add-in.
Kasper De Jonge (00:09:08): And after that, the first version of the tabular model and also tabular the way we have it that was one of the PMs on that team creating it and shipping it. First version of Power BI that we killed luckily, the one that shipped inside of Office 365, that was not a good thing. So, we kind of killed it.
Kasper De Jonge (00:09:27): After that, we really went full steam with Power BI when James Phillips came along. And that was, I must say, I mean that's really for me before and after time. Before, we were like we have SQL server shipping releases where it takes two years or four years to ship something and the world where all of a sudden, we were like no longer inside of the SQL server organization. But we're now part of a different team with James Phillips who came from a completely different world. He was starting Couchbase. He comes from Silicon Valley. It was a completely different world.
Kasper De Jonge (00:10:01): And he said, "Okay. I don't care about any history. What do we need to do?" And what I want you guys to do is to say to build a new product that says five minutes to wow. That's really what he said. Someone needs to be able to log in and go to Power BI and five minutes to wow. You need to be able to say wow.
Kasper De Jonge (00:10:18): That was a fundamental shift for us. That was completely different to anything else that we've done before even though we already were pretty amazing all the things that we were doing with Power Pivot in my mind at least then, but this was taking to the next level. And also, Satya came on board. It was also a big difference like Steve Ballmer before Satya. And he brings James Phillips there.
Kasper De Jonge (00:10:40): And it was a tough time for a lot of people on our team as we talked about before. A lot of the old school guard, it was tough because he had to do things very differently. And he came. I remember as see, can you imagine James Phillip is a vice president. And he comes in. He storms into the room and says, "I need DAX. Help. I'm writing DAX." What? He's running it. And everyone points to me. [crosstalk 00:11:03] shit.
Kasper De Jonge (00:11:02): And he says, "Come. Come." Okay. I had to go to his Office. And he was doing something. And he was like, "How do I write this DAX formula?" And then, it's magical answers. It's like, "What you're doing? You don't even need to write DAX," because he would try to solve problems in an Excel way. But he actually created the table. I said, "You just create some relationships here. You drag it in. It just works."
Thomas LaRock (00:11:27): Wow.
Rob Collie (00:11:28): Oh, that's awesome. Of course, James has a strong Excel background.
Kasper De Jonge (00:11:33): Yeah. He's an exec. And although [inaudible 00:11:38], you've probably seen all his keynotes and all these things. He comes across an exec. But I mean he's very technical. He was one of the founding people of Couchbase. He definitely knows what he's talking about.
Rob Collie (00:11:47): Well, just the fact that even attempting to write DAX is an awesome, awesome, awesome story. That's so cool.
Kasper De Jonge (00:11:55): Well, I must say that's another thing. So, one of the first jobs actually on the Power BI team, me, together with Josh Kaplan, we were doing Power BI Embedded. So, it's an embedding story. And you have APIs. It's not really because I had some developer background too because I used to be a developer way back when, some developer background. So, we were PMs for that.
Kasper De Jonge (00:12:15): One moment, we got an email saying, "Oh, yeah." So, Scott Guthrie tried out your API. And he has some feedback. Shit. Scott Guthrie has some feedback. Oh my.
Rob Collie (00:12:25): Remind for the audience who Scott Guthrie is.
Kasper De Jonge (00:12:28): So, Scott Guthrie is, I don't know exactly what the title is, but he's the vice president reporting to Satya. And he owns everything of Azure and everything Power BI. I don't know how big his team is. They're probably 30,000 people or something on development side.
Rob Collie (00:12:44): Yeah. It's like-
Kasper De Jonge (00:12:45): He's the big guy.
Rob Collie (00:12:46): It's like an Amazon-sized team.
Kasper De Jonge (00:12:49): Yeah. He's that guy. And he's trying these things. And he's looking at the documentation. If your documentation sucks, he'll tell you.
Rob Collie (00:12:55): Here's a similar story just to... We'll definitely come back to what you're talking about. But I want to make sure I share this. So, for a brief period of time, when I worked in Office, I drew the short straw. And my team was assigned sort of maintenance ownership of the Office web components which were ActiveX control versions of Excel essentially.
Rob Collie (00:13:18): It's this completely false step into the web like ActiveX, like, "Oh, god. The whole thing was a debacle." So, these Office web components someone needed to, I don't know, shepherd them. We weren't really actively developing them. So, we were given ownership of the support distribution list, the email list where people would ask questions about how to get their OWC components to work. And we woke up one morning, came to work. And we checked the distribution list like we were supposed to every day.
Rob Collie (00:13:48): And at three o'clock in the morning, someone had asked a question like, "Hey, I'm writing this script to this ActiveX control. And it's not working, blah, blah, blah." And like at 3:05 AM, five minutes later, vice president Steven Sinofsky replies and says, "Hey, this is the code you need to use instead," and gives them the alternate script that works.
Rob Collie (00:14:10): And the person goes, "Oh, thank you. That works great." So, we're all sitting there just sort of chilled to the bone that morning, like, "Did you know Steven was on this distribution list?" Like, "No." He's been watching us this whole time every single thing we've ever said. He's seen it.
Rob Collie (00:14:24): And furthermore, he knew the answer to a question that most of us didn't. It was like, "Oh, it was inspiring and terrifying all at once."
Kasper De Jonge (00:14:35): Yeah. I always am impressed with people at that level at Microsoft at least the ones that I've seen is how engaged they are. I remember one since long time ago, I joined Microsoft when Bill Gates was already gone. But I once saw him present in some conference in the Netherlands long time ago. I don't know exactly what it was. And people were asking him, like, "So, I have a setup installer. And I have this on Linux." And I don't really remember exactly the question was. But he said, "Oh, okay. Yeah. You just do this, this and this on your Linux distribution. And then, you do this, this, and this." Is this Bill Gates talking about... I was just sitting there in awe-
Thomas LaRock (00:15:14): Especially the old guard. They were just loaded with stuff like that. So, let's underline something. You packed up your family. You were married, right?
Kasper De Jonge (00:15:22): Yeah.
Rob Collie (00:15:22): You moved internationally to pursue this dream. How did that go down family wise? That's got to be a tough sell.
Kasper De Jonge (00:15:30): Well, it was not that tough to sell actually except my mom, obviously. But other than that, I mean this is just a once-in-a-lifetime opportunity really. I never thought that it would... I was having a nice job. And it was nice. But worked at Microsoft. I actually developed products. And I was super passionate about Power Pivot that really helped.
Kasper De Jonge (00:15:51): I don't think I would have packed up my things to work on exchange or whatever other products. No. I was passionate about it, really. That really helped.
Rob Collie (00:15:59): And also, didn't you have a newborn about that same time?
Kasper De Jonge (00:16:02): She was one when I moved to the US. Yeah. It was a big change. I mean go to a whole different country. But I think the good thing is you've had him on the podcast with John Hancock was my manager then when he hired. Everyone's taking care of me and making sure everything was okay and nice flights and the CATs were taken care of. And there was something like I had a place to stay. It's not like I'm packing up for [inaudible 00:16:27].
Kasper De Jonge (00:16:27): It's a corporate America. Microsoft's taking care of you. So, you get taken to the bank with some lady who helps you. You go to the bank and say, "Okay. So, here's the account you're going to open up," because of course in the US, it's a lot about credit scores.
Kasper De Jonge (00:16:41): Obviously, I did not have any credit score because you come from outside. But they say, "Well, this bank doesn't care about your credit score because they are friends with Microsoft. And they don't really care about your credit score. And they think you're going to be good for it." And they give you tips and tricks and things to do and settle in. So, that's really a big change.
Rob Collie (00:17:00): So, there's something in detail in there that I find jaw-dropping. Is there any equivalent of credit score in the Netherlands/
Kasper De Jonge (00:17:07): No.
Rob Collie (00:17:08): Wow.
Thomas LaRock (00:17:09): How could that be possible?
Kasper De Jonge (00:17:10): No. There's a reverse. So, if you do not pay things on time and you get your really bad shape of paying back things, then, you get noted in some register. And then, when you want to do a new loan, they look into that register. And then, they say, "Well, actually, you're delinquent payer. Let's talk before we do anything else."
Rob Collie (00:17:32): So, it's pass, fail. The credit score in the Netherlands is pass, fail. But in the United States, we're like, "Oh, we really need a thousand-point scale to rate you."
Thomas LaRock (00:17:42): Yeah. A thousand-point scale that only goes up to 849 or something.
Kasper De Jonge (00:17:46): Yeah. It was mind-boggling. One of the things they told me like I was there. And I wanted to buy a car. And he said, "Oh, you're not going to buy a car with cash. Are you crazy? Take a loan because otherwise you'll never get a nice credit score." I had never taken a loan in my life except the mortgage. We don't do that.
Rob Collie (00:18:03): The fact that Tom and I are just stunned, it really speaks to the American culture, doesn't it? It's like there's another way to do-
Thomas LaRock (00:18:12): Take a loan so you can have better credit. And you need to have three or four accounts open and maintain a certain balance month to month in order to maximize your chance to go more into debt.
Kasper De Jonge (00:18:25): So, this lady was explaining this all to me in the couple first weeks. There was a lady that you get assigned to. And she explains all these things. And I was like, "What?"
Rob Collie (00:18:32): I've actually been thinking for you a long time that the United States like the credit score system like we should call it what it is. It's like, "Are you a good milkable crop scorer?"
Kasper De Jonge (00:18:42): Yeah. Well, I mean this show, are you spending nice money? And if you're spending a lot of money, then, you're getting better.
Rob Collie (00:18:47): Yeah. Are you a good consumer that we can extract a lot of money out of? There's a little bit of a loophole in it in this system which is that if you're constantly paying off all of your debts all the time like really quickly, you get a good credit score. But you're not very profitable for the credit agencies, et cetera. That's the way to really stick it to the man, is to-
Kasper De Jonge (00:19:08): Yeah. You have to be a little bit depth.
Rob Collie (00:19:10): Yeah. It's a weird system.
Kasper De Jonge (00:19:12): I definitely like the US. There's lots of opportunity there. And the vibe that you get and see... I've never lived anywhere else. But the vibe you get in Seattle is just amazing. I don't see anything like his here in the Netherlands. You go to a birthday party. And you have people from Twitter and Facebook and Google and Amazon. And they're all there. And they're all talking.
Kasper De Jonge (00:19:32): And I mean there's energy. There's really energy there that I love. I haven't seen that anywhere else.
Rob Collie (00:19:37): That's pretty much universal the United States. I found that same energy in Cleveland. I'm kidding. I did not. Seattle is a bit special in the US, I think.
Kasper De Jonge (00:19:45): Yeah. And that, and then, of course the nature and the trees. And there's so much space. We don't have a lot of space here in Netherlands. The same amount of price I can buy a house here. But it's in the street where I can see neighbors everywhere.
Rob Collie (00:19:56): And how does the Netherlands feel about global warming? How does the Netherlands feel about sea level rise?
Kasper De Jonge (00:20:03): Actually, we're not that scared of it.
Rob Collie (00:20:05): No?
Kasper De Jonge (00:20:06): I live right now under the sea level. My house from the sea level.
Rob Collie (00:20:09): Yes, which was apparently New Orleans fatal flaw at one point with the hurricane.
Kasper De Jonge (00:20:13): Yeah. But we've been having this for five, 600 years. We know how to handle it. We actually made land. We made land. We had a sea.
Thomas LaRock (00:20:22): They're called the lowlands for a reason. And they've been that way for centuries.
Rob Collie (00:20:26): Exactly. Yeah. I mean to be below sea level and on the ocean, you had to make that land. Otherwise, we call that ocean floor otherwise, below sea level.
Kasper De Jonge (00:20:36): No. That land's actually because they didn't have more lands for people to build houses on. So, they just cleared up... Its inner sea, it's not out... We have an inner sea, if you will. And that's where they just cleared up that land. And people live on it now. There's whole cities on it.
Rob Collie (00:20:51): So, I guess what you're saying is that if you've been fighting off the ocean for centuries, a sea level rise that's six inches or whatever, even something extreme like that, it's not going to bother you. It's just a rounding error.
Kasper De Jonge (00:21:06): Well, obviously, I mean I'm just over generalizing a little bit because everyone is concerned about it. But I'm not necessarily we're concerned about the raised sea levels. It's about everything else that is a concern. And we're a very small country. So, there's not much really we can do.
Thomas LaRock (00:21:21): The ecosystem, they'll change as a result, right?
Kasper De Jonge (00:21:25): Yeah.
Thomas LaRock (00:21:25): Like shellfish or some something that doesn't exist there anymore as a result because of warmer waters and things of that nature. That's a bigger concern.
Kasper De Jonge (00:21:32): Yeah. That's a bigger concern. Yeah. Two weeks ago was -15 Celsius. A week later it was 15 Celsius. And this week, we're back to minus two. We've never seen this before.
Rob Collie (00:21:43): Oh, it's just so interesting though. Isn't variety the spice of life? I'm just kidding. I'm just kidding. [crosstalk 00:21:49] I'm just kidding. So, I think about this through the lens of Florida, especially South Florida. It should be pretty damn concerned about sea level rise. I don't think it's going to be too long before insurance companies are going to take a very different stand on oceanfront property in Florida.
Rob Collie (00:22:06): And that will be the place where this starts to really come home to the average American.
Kasper De Jonge (00:22:12): I actually read in the Atlantic, there was an article about Hawaiian beaches are disappearing because people are shoring up. All the rich people are making nice walls. So, their property won't get gone. But there's no beach left. The beach are disappearing.
Rob Collie (00:22:26): They're kind of making their own little micro-Netherlands around their property.
Kasper De Jonge (00:22:31): Exactly.
Thomas LaRock (00:22:32): I see pictures from Andre Common from time to time about he goes to the beach in the Netherlands somewhere. And low tide, I swear it's got to be a kilometer, Rob, from where they park. And they just keep walking because it's just this slow gradual decline. And so, the beach just seems to extend forever before you get to the water. It's amazing.
Rob Collie (00:22:56): Wait. Did you just translate into Kasper's units? Did you say kilometer? Look what a metropolitan man of the world, Thomas.
Thomas LaRock (00:23:03): I try to make our guests feel as comfortable as possible. You know this, Rob.
Kasper De Jonge (00:23:08): Yeah. I didn't do that. I was like, "No, the sea level's going to rise by six inches. Damn it."
Thomas LaRock (00:23:11): You did.
Rob Collie (00:23:12): It's not centimeters.
Kasper De Jonge (00:23:14): All my cars in the US, I set to kilometers. I never could get used to miles.
Rob Collie (00:23:18): But how'd you deal with the speed limit then? You're just, "I'll just tell you what. I'll just drive really slow. I'll just drive 55 kilometers per hour and a 55-"
Thomas LaRock (00:23:24): Stay with traffic.
Rob Collie (00:23:26): There you go. That is the move. The speed limit just doesn't really matter, does it?
Kasper De Jonge (00:23:29): Not in Seattle.
Rob Collie (00:23:32): Really, not anywhere. 465 around Indianapolis. There's just certain stretches where it's been declared an Autobahn. There's never any cops and-
Kasper De Jonge (00:23:42): Oh really? I have more mental traffic wise in Seattle, is never a point that you can actually drive fast enough to get in trouble.
Rob Collie (00:23:49): That's true. That's true. How long were you in Seattle because you're back in the Netherlands now?
Kasper De Jonge (00:23:53): Yeah. Back in Netherlands now. I was in Seattle for six years.
Rob Collie (00:23:56): Wow.
Kasper De Jonge (00:23:57): Yeah. It's quite a long time. But a second kid got born in US. I don’t like US part. But there are certain areas that in us that I think I would feel better to raise my kids in the Netherlands then. I just hated the fact that when I brought my kid to school, she had active shooter drills, my daughter.
Kasper De Jonge (00:24:20): And then, six years ago, she's now 12, she still remembers. She was telling me after like, "We had to go into some closet." So, active shooter drills. And we were living in Redmond. Redmond, it's all Google. It's all Amazon. It's all Microsoft. And there were little kids in my daughter's class that didn't have money to eat because their parents were the cleaners. They didn't have money to eat. And we had to pay as parents, you give some money to the school. And then, the school gets them some breakfast.
Kasper De Jonge (00:24:50): And a couple of streets over, there was lots of things that happened to deck over time. A couple of streets over, there was a party and someone got killed, shot by accident because there was a gun laying around. These things don't happen here.
Rob Collie (00:25:02): I completely understand what you're saying. But especially if you've had the benefit of growing up someplace different where the sense of we're all on the same team and we're all going to sort of take care of each other. I mean it's not a yes, no thing. It's a dial like how much do we acknowledge it. How much do we take care of each other? How much do we protect each other?
Rob Collie (00:25:23): If you grew up in a place where that dial is sort of cranked relatively high and then you come to some place where it's not and I know that we're not, the United States has got a relatively low score on the take care of each other thing. And that's not me being anti-patriotic or un-American or whatever. It's just sad. I really think we need to as a country be doing more to crank that dial the other way. Every little bit helps. I can imagine is that really sort of the primary reason for going back?
Kasper De Jonge (00:25:52): It's that and also parents who are getting older. So, at one point, it's hard for them to travel. And, obviously, I was in an incredibly lucky position that I had an amazing team that said, "Yeah. If you want to move back, you can keep your job."
Rob Collie (00:26:11): And in 2010, when I left, that wasn't a thing yet. When I moved to Cleveland, if I had left seven years later or whatever, they probably would have let me keep my job. And my whole life would be very, very, very different because I would not have left. I relocated for family reasons. And I'm glad that I did leave. I think it's worked out better. I've really enjoyed and appreciated the path that my life has taken since then. But I wouldn't have if I'd had the option.
Thomas LaRock (00:26:38): Yeah. Now Kasper, you left before the plague, right? You left before last year. It wasn't a COVID thing.
Kasper De Jonge (00:26:45): No, no. I left in 2016, quite some time ago. No, no. Before COVID, yeah. I miss the work because I don't do exactly the same I used to do because I used to be a product PM actually developing features and working with the developers and brainstorming and all of these amazing things that we would huddle together and do all these things. That I definitely miss because you were talking about your thrones and names and Rob for like Marius. And those are some people that no one have heard about because they never really go speak out at conferences.
Kasper De Jonge (00:27:18): And then, they have no presence online to talk to the community. But they are amazing people. The ideas and the brainstorms and the things that are happening there is just, yeah, next level. And that's the unfortunate thing now with COVID. I haven't been to the Office in a year or more than a year, actually a year and a half. I haven't been to Seattle in a year and a half.
Kasper De Jonge (00:27:39): I used to soak that up like fly to Seattle and stay there for a week and just brainstorm and hear and talk and see what's going on and, yeah. That's not happening. And that I definitely miss because I'm sitting here. I work fully remote. I already since 2016. I changed from the product role.
Kasper De Jonge (00:27:59): We're actually developing product two what we now do was our customer advisory team which I have an amazing team too. We have Adam Saxton, Patrick Leblanc, Matthew Roach, Chris Webb, a lot of other people on our team. They're really good guys.
Thomas LaRock (00:28:17): I was going to mention how Kasper mentioned the names of the people on the team. It was Patrick Leblanc, not Leblanc, Leblanc, Leblanc. He did it perfectly.
Rob Collie (00:28:27): See, it can be pronounced. We can't use Leblanc. We can't use that as a World War II test to trap the Germans.
Kasper De Jonge (00:28:36): So, many people know French.
Thomas LaRock (00:28:36): I bet you could.
Rob Collie (00:28:40): Yeah. I'm positive that I would never be able to pronounce a French sentence the way that Olivier could. That's never going to happen. It just sounds so much cooler from a native speaker.
Thomas LaRock (00:28:50): So, I have a quick question, questions and comments. Kasper, you said you killed Power BI V1. Tell me why you killed? What was wrong with it? What was so grotesque that you just said, "This Franken monster just needs to go?"
Kasper De Jonge (00:29:06): The really good thing there was I mean back and then, it was really Office is everything. And we need to be able to get people everyone uses Office. I mean it's still the case like Excel. I listen to you guys a podcast too with Bill Jelen which is awesome too. Excel is really the place to be. Office is the place to be. So, what we're trying to add on our way of doing things and are trying to... That came up in that call too.
Kasper De Jonge (00:29:31): Office is a different ballgame than our team. It just takes forever to get anything done. And that is what killed us because we wanted to iterate. And we had to compete then because that was the big problem. We had to compete with Tableau because when we started with Power Pivot, we were doing Excel which is great. And a lot of people were using it. And I think you guys mentioned it there in too.
Kasper De Jonge (00:29:55): It's so hard to find Power Pivot. It's so hard to find the data model. It's so hard to find the features that could compete with things like Tableau or Qlik or whatever the tool is at the time. And then, we're building on top of that inside of Office and all of these things. And then, James Phillips was the guy who came in and said, "Guys, this is not the way forward. This is not the way that we can compete because we're going to lose if we do it this way."
Kasper De Jonge (00:30:20): And that's where he really came along and said, "Okay. We're going to do it completely different. We're going to look at things completely." And he had the guts to go do this too. He had the guts to go in against Office, not really that they were kind of going against. But he said, "If we need to compete with these guys who don't have Office," they are a startup team, they have dedicated people working on this amazing product because I mean Tableau was definitely an amazing product, and it still is, yeah, we have to do something.
Kasper De Jonge (00:30:49): So, that's really why the [inaudible 00:30:50] of Power BI got moved out. And we went to a completely separate product where James came in and said, "You need to be able to sign up with your email address, just like he just did the startup trick with us." He said you need to be able to sign up with your email address, your corporate email address, not your Hotmail. We don't want that, corporate email address. You log in. And within five minutes, you have to have some insights to where you can say, "Wow, this is cool." And he did it.
Thomas LaRock (00:31:20): So, I'm going to share my James Phillips story because I have one. I have one James Phillips story. And I'm going to say the year was 2015. I could be off a year or so. And James Phillips drives from Seattle to Vancouver to meet with me and the members of the past board of directors.
Kasper De Jonge (00:31:41): Oh Yeah. I know what this is. Yeah.
Thomas LaRock (00:31:43): And he does that five minutes to wow speech. And I'm sitting there while he's talking. And he says five minutes a while, five minutes a while. While he's doing all that, I log in. I sign up. I can't use my Hotmail account. I have to use my Thomas LaRock or create a new account and all that.
Thomas LaRock (00:32:00): I clicked on I think the Google Analytics widget for the dashboard to get some data. Less than five minutes, I had insights. It was up and running five minutes a while. So, when he gets done talking, I just turned over. I go, "By the way, the five minutes, the wow is a real deal." And I show my laptop. I go, "While he was talking, I'm up and running with Power BI right now."
Rob Collie (00:32:19): At that moment, James is like, "Oh, I like you." You, I like.
Thomas LaRock (00:32:24): Within 10 minutes, he didn't like me anymore.
Rob Collie (00:32:26): Well-
Thomas LaRock (00:32:28): No, actually, it didn't happen then. It happened later. He was there to ask for pass to help him launch essentially Power BI events and user groups. And he gave his pitch. We asked a few questions. We essentially told him we needed time to talk about it. He was supposed to have dinner with us. He decided not to have dinner, started driving home immediately. We took that as a bad sign.
Thomas LaRock (00:32:56): And, eventually, the answer that came back from us as a board was PaaS was simply not equipped to provide the support that he needed to do the launch that he wanted. And I feel like we really failed there. He was handing us an opportunity. But because of who we were and what we could offer, and Rob knows this, we could not execute. And we were not the right place for him.
Thomas LaRock (00:33:26): And I feel bad, but I had no option, but to say, "Hey, this sounds great. But we can't do that. And James, I know you just want us to just take this on and make all this magic happen. And I felt like the best thing was to say no to you now instead of disappointing you later. I'd rather just you start looking elsewhere immediately and get your stuff up and running." And that was essentially what we ended up doing.
Thomas LaRock (00:33:50): But every time somebody talks about James Phillips, they always have pleasant things to say. He does all these great things. He's a great person. And I'm always like, "Yeah." And I'm the guy who kind of disappointed him. So, I always feel like if I ever see James again, I have to just apologize profusely. He's going to be like you. You could have been a part of this.
Rob Collie (00:34:11): Oh, Tom, I wouldn't worry about it.
Thomas LaRock (00:34:13): Yeah. True. He doesn’t remember.
Rob Collie (00:34:14): Everything always works out in the end. I mean things worked out great for pass. I wouldn't sweat it.
Thomas LaRock (00:34:18): Yeah. And I think years later, it's pretty well known that no was the correct answer. I hope James, if he's ever listening to this and we should have him as a guest, I hope James understands why we were just not going to be able to execute for what he needed. And I wish. And then, immediately, you had your own data analytics conference there within the next year or so.
Rob Collie (00:34:46): Data insights.
Thomas LaRock (00:34:47): Data insights. I'm like, "Great. I'd like to go." I couldn't even get in the door. They're like, "Sorry. Sold out." I'm like, "Really? I'm the president of PaaS, just like... Right. You're not on the list.
Rob Collie (00:34:58): Not on the list.
Thomas LaRock (00:34:58): That's how I knew I was just being cut from James Phillips.
Rob Collie (00:35:02): In fact, they flip it over to the other sheet. And they go, "Actually, you're on that other list. You're on the Dutch bad credit score."
Thomas LaRock (00:35:08): I couldn't even get...
Kasper De Jonge (00:35:09): Exactly.
Thomas LaRock (00:35:10): I'm like, "I'll pay my own way. I'll give you money to attend your event." And they were like, "No."
Rob Collie (00:35:16): Yeah. Under no circumstances am I permitted to let you in here with Mr. LaRock.
Thomas LaRock (00:35:20): That's exactly what happened.
Rob Collie (00:35:22): So, Tom, you asked that question. I think we're getting into the really cool stuff right now. So, let's stick with it. You asked that question about why did you have to kill Power BI V1? And before I even heard Kasper's answer, I was dying to jump in with the snarky joke because it was too tightly integrated with Office.
Rob Collie (00:35:39): So, first of all, we get this out of the way. It was the right decision. There's no two ways about it. It was the right decision. At the same time though, it is a tremendous shame, in my opinion. And it's not the blame for the shame here. It does not lie 100% at all with one or the other. But I've said it a million times. If Microsoft were just... If all it was, was Excel and Power BI, that was the Microsoft corporation, Excel and Power BI, they would rule the world. There would never have been a Tableau.
Rob Collie (00:36:15): All these other tools, there wouldn't have been any room for them because it rounds to a 100% of the important users of these tools, these BI tools. Just round it to 100. They're the Excel people. There are people who are not Excel, people who build BI.
Kasper De Jonge (00:36:32): I mean it's not that easy, I think. I mean, while I do agree, it's not that easy because as you know, beginning especially with Power Pivot, we were struggling with the fact that people are running their business on Excel with macros and VBA, Scripts and-
Thomas LaRock (00:36:46): Totally.
Kasper De Jonge (00:36:46): ... whatever it is. And they're not going to upgrade. And while Tableau is upgrading every month or every couple of weeks and they're doing new features… So, even if we wanted to do it, customers are going to say, "No way. We're going to upgrade Excel."
Rob Collie (00:37:01): I've got no objection to what you just said. I completely agree with all that. But I'm still saying though that if that was the whole company, these problems were solvable.
Kasper De Jonge (00:37:10): Yeah. I mean you can see it, for example, our sales organization, if you compare to our sales organization to Tableau or any other competitor sales organization, they have dedicated... We have some guys too. But in every country, we have a handful of people where they have a whole company full of sales guys who go in and do all these things. Yeah. I mean they have more resources and things to do. But they don't have the integration.
Rob Collie (00:37:35): So, think about if you're familiar with it, the engineering effort that went into Office is Click-to-Run initially, that thing is insane. It is like a Manhattan project of software in terms of compatibility and deployment. It's unbelievable what they've done there. It is such a radical, radical, radical departure from everything else that they've done. Now, it's all a matter of competing priorities. This is the thing, is that Microsoft isn't just those two products.
Rob Collie (00:38:08): But if they were just those two products and you're capable of pulling off something like Click-to-Run, you absolutely, Microsoft, could have solved this alternate Microsoft. Absolutely could and would have solved all of these compatibility problems that we're talking about.
Rob Collie (00:38:21): But they weren't high enough of a mutual priority for the teams with these separate incentives. This is what big companies do.
Kasper De Jonge (00:38:29): Like in the Office world, they were competing with the Google that was coming up with all the documented [crosstalk 00:38:34].
Rob Collie (00:38:35): That's right. That's right.
Kasper De Jonge (00:38:36): And so, that was more important because they don't want to lose everything. And that's what they did.
Rob Collie (00:38:41): So, this is why I'm saying simultaneously, even though I kind of griped about it at the time, I have absorbed and admit that this was absolutely the right move. It's been very, very good. It's benefited my business tremendously. I'm nothing, but a winner in this story. And at the same time, I think we should all still admit to ourselves that it is a shame that marriage sort of had to come unglued because the most impact, the absolute most optimal outcome was down that road.
Rob Collie (00:39:09): It's just that parties involved for good reason weren't incentivized to pay that level of cost to tackle these problems.
Kasper De Jonge (00:39:17): Yeah. I agree. It is tough. But I think it's actually the right decision. And the way that we can move and operate now and we have some amazing people like visionaries on the team, I'm not saying that Office is now visionaries. But they are bound within the organizations and expectations that they have. We said we just tell them, "Every month, probably desktop is going to ship, and you have to update."
Kasper De Jonge (00:39:41): And I mean, sometimes, things break. Well, can you imagine, Rob, that Excel is going to break things?
Rob Collie (00:39:47): Not allowed.
Kasper De Jonge (00:39:48): Yeah. Exactly. And for good reason. And people built their business on top of that. But people like Amir Netz go, "They have amazing ideas. And they can be constrained."
Rob Collie (00:39:59): [inaudible 00:39:59]. So, back to this Office versus kind of the freedom to innovate kind of path, I mean it's a very overwhelmingly net positive trade-off that was made. It still is a trade-off.
Rob Collie (00:40:14): I believe the place where Microsoft is seeing the downside of the trade-off is that adoption of these tools is harder to force top down than it would be if it were bottom up. Bottom up adopt is always fun. So, an organization commits. They buy the licenses. They purchase Power BI. They commit to it.
Thomas LaRock (00:40:34): Well, actually, we do see it most of the times it comes bottom up.
Rob Collie (00:40:39): Yeah. That's the way it has to happen. I agree, even if the organization has purchased and is committed to Power BI. Now, it's not a yes/no thing whether or not they're going to adopt it. Of course, yet the answer is going to be yes. There's a question of how fast and how well. It's a measure of degrees, right?
Kasper De Jonge (00:40:55): Exactly.
Rob Collie (00:40:56): And that alternate reality, I talk about where the two orgs had really ponied up all the resources required to solve all these very difficult problems, oh man, the bottom up adoption would just be like wildfire. That's the thing that we lost. But we didn't really lose it because we were never going to get there as it was, right?
Kasper De Jonge (00:41:14): No. But I think we're getting back to it. I mean I don't know if you've seen any of the new things. Chris Webb has a new blog post out of it. And it's about actually being able to have an Excel spreadsheet and a PivotTable hosted inside a Power BI and running against the data set in Power BI.
Kasper De Jonge (00:41:32): We're getting back to it. And you can have a PivotTable inside of Excel and say, "Okay. Connect to my data set that are sitting in Power BI." So, we're getting really back to it.
Rob Collie (00:41:43): I agree it's not the 100% A-plus score though, right?
Kasper De Jonge (00:41:47): Of course.
Rob Collie (00:41:48): A large percentage of these things that I see the integration points between Excel and Power BI, they're all things that happen after you've already built a model, after you've got a data set published. The thing is how do these data sets get created? It's the author. It's the Excel author that I think we're not collectively as an ecosystem doing enough about.
Kasper De Jonge (00:42:09): Yeah. I totally agree. I don't see that happening.
Rob Collie (00:42:12): What are you going to do, right?
Kasper De Jonge (00:42:13): But we see more and more. This is really one of the areas that we, as Microsoft, are really amazing at and especially, again, I see this as also a change after Satya came along. It's more cooperative. So, all these amazing things like the security stuff that we integrate with, just mind-boggling. No one has this. And it's just because we're working together with all these AED people and MCAS. And there's so many. No one has this. And we just integrate, and, yeah, we just do it. And teams integration.
Rob Collie (00:42:43): It's true.
Kasper De Jonge (00:42:45): It's really amazing.
Rob Collie (00:42:46): Yeah. We're talking about this divorce in a way from Office. But no. Power BI runs in Office tenants, right?
Kasper De Jonge (00:42:54): Yeah.
Rob Collie (00:42:54): It is a shared security and administration framework which is that in itself would have taken an act of God in the days that I was there. That sort of cooperation was never going to happen. It is really, really, really impressive. It's not a secret or a mystery as to why Microsoft is so successful in the business space for business customers and not just enterprise either. It doesn't have to be enterprise. It can be mid-market. It's really hard to find something of better value than Microsoft.
Kasper De Jonge (00:43:27): But I think also this is again going back to the mindset that James Phillips brought along. I mean I keep talking about it. But we started out not having any of these things. We just shipped something, and we created this market. And we created this excitement around things. And then, at one point, we got to this inflection point where, okay, now we're starting to talk to security teams. And now, we're starting to talk in the boardroom instead of bottoms up. People are just excited and doing things and writing DAX formulas and ability to just say... We talked about with Bill Jelen.
Kasper De Jonge (00:43:58): So, instead of actually having to redo the same Excel spreadsheet and copy and pasting and stuff, you can just press refresh. People were excited about it. And they went to their managers and their IT teams and said, "Guys, we want this. And we want this for more people." And they started sharing. And then, oh, now, we need licenses.
Kasper De Jonge (00:44:17): And then, the security team comes along saying, "Oh, what is all this information stuff, all this data that you are putting inside of the cloud? Are you crazy?" And then, we get all this resource. So, now, we have all these features bring your own key. You can encrypt all your data with your own key and stuff like this that we needed to add it. But we didn't.
Kasper De Jonge (00:44:38): Previously, when we were on the SQL server team and we were building all services, we would think about all these scenarios and just ship them. And that's why it took so long. But now, now we just add it afterwards.
Rob Collie (00:44:48): It's amazing. The cloud shift too, right? In the beginning, such resistance on the customer front for putting something like BI in the cloud. And even though they already had their email on the cloud, it was always so funny.
Rob Collie (00:45:03): There aren't that many organizations anymore that have a no-cloud rule. -
Kasper De Jonge (00:45:07): Yeah. There are still some countries where things are a bit less cloud, like, Germany, for example. They're a bit less and easy. They use Power BI report server, for example, on-prem. But the majority, that's a night and day between five years ago and now. People willing to go to the cloud.
Thomas LaRock (00:45:28): I want to make a comment about because we've said a big difference between then and now like Satya and somebody like James Phillips coming in. And for me, I think the biggest turning point or one of them that I've seen between, say, old, new Microsoft, and you guys touched upon it earlier, you mentioned you can't have Excel break. Excel can't break. That would be bad, so, this whole idea of backward compatibility.
Thomas LaRock (00:45:52): And for us over in the SQL world, the SQL MVPs, now the data platform, MVPs, for years, we asked about certain features. And we were told we have to be concerned of backward compatibility. Whatever we're going to introduce, we have to make sure nothing ever breaks. So, along comes SQL 2014 and introduce a new cardinality estimator. And it was almost like the surgeon general's warning right at the top just to let you know, the new cardinality estimator may help your queries run faster. And it may help some queries run slower. You should know that.
Thomas LaRock (00:46:29): And we're going to keep going forward. And that to me was this turning point of a new Microsoft. I'm like, "What about backward compatibility?" They're like, "Yeah. Something might run slower. If it runs slower, we'll give you a flag to make it run as good as it used to be. And you should look at [inaudible 00:46:43]." And that was it. Just move on.
Thomas LaRock (00:46:44): And I see that with Power BI like Kasper said, every time I load Power BI, I feel I don't know how far behind I am on updates. I just know I need to... Sometimes, I don't want to fire it up simply because I know I'm going to have to update it immediately. But I'm okay with it too because I know that's the way I want it to be. I want to be up-to-date with the latest features especially security features, things like that.
Thomas LaRock (00:47:09): So, that whole notion of backward compatibility which for years was almost a way I thought for PMs at Microsoft to just not do things because it was hard. And it might break something else. So, I'm just not even going to worry about it. Yeah. That sounds like a great feature. But I ain't going to do that because something might break. And that is no longer the mindset.
Rob Collie (00:47:27): The whole backward compatibility edict in its day was important because without it, those same program managers were more than happy to just wreck the world with their new ideas. It was to control the drug addict that would destroy things.
Kasper De Jonge (00:47:46): Shipping software at Microsoft is just a whole different ballgame than shipping software anywhere else. At least, I've never worked in actual other product companies. I was a developer too. And I worked for hospitals and other things. And we wrote software.
Kasper De Jonge (00:48:01): But building products that has been used by millions of people, you really have to be careful. And you have to think about so many more things than just backwards compatibility is one. Even in Power BI, we don't break things. The APIs, we won't break. And if we do so, then, it's probably not meant to be. So, we still have to think about backwards compatibility a lot.
Kasper De Jonge (00:48:20): But it's different. We can take more a little bit more risks than, for example, Office. When I want to give you a new Power BI desktop file, I check it in now. And it's going to be there a couple of hours from now in your Office store. If there's something wrong with the Power BI service, we can go in and fix it like in an hour. That's not possible in your Office. We're set up differently than Office. And we're set up ways in ways that are different than where before.
Kasper De Jonge (00:48:43): But still, there's a lot of other things too. A lot of people are very surprised by it. If you write software at Microsoft, you have to think about... And Rob was probably there in the real days like OAPI. You have to document everything that you do because the EU in, I don't know exactly one of us, like 2000 something said, "You cannot have APIs that you're not telling everyone about because it's unfair." And they lost this big lawsuit.
Rob Collie (00:49:08): Yeah. Internal APIs that Office would expose, the SQL team could use, or vice versa that were undocumented for the rest of the world. They took that fun little toy away from us. We weren't allowed to do that anymore. We weren't doing that. You'd think that cases where we were doing that were meant to be sinister. They never were. They were never meant to be sinister.
Rob Collie (00:49:27): It was just like, "Look, if we can allow something to work without having to carry the backward compatibility burden of it forever for the rest of the world, that would be awesome." Well, we got that toy taken away. It was taken out of our toolbox as well. And I understand why. It is something that could be abused. It wasn't a sinister thing that we were doing. But, yeah. We had documented everything from that point forward [crosstalk 00:49:50] every last little bit like your file format areas. It just became this massive clerical problem in addition to the software engineering.
Kasper De Jonge (00:49:56): It's not just this. It's also about they're saying that Microsoft runs on trust. And it's really something that everyone it needs to be secure. I mean, sometimes, we're not doing it as much as we can. But we try to be accessible for everyone. If you use a Microsoft product. And you have a disability or something, you can still use it. We will try our best to be able to use it.
Kasper De Jonge (00:50:18): That's definitely not going to be the case with every other product. Secure by default. Lots of things that we have to think about that other companies really don't have to think about which makes it good too. You can trust it.
Rob Collie (00:50:29): Yeah. Microsoft does get held to a higher bar than a lot of other companies do. And part of that is the outcome of the antitrust case and the court battles with the EU. I mean it's not a 100% source of it. It is that Microsoft your reputation, the reputation of Microsoft is just super, super, super important to preserve. And we're still watching this next wave of technology companies.
Rob Collie (00:50:54): So, a lot of them are still operating in that before era where they haven't really been held to account yet. Their turning point's coming.
Kasper De Jonge (00:51:03): I mean it makes total sense if you ask me. If you're a startup company, you're just going to go. And you're going to listen to feedback and listen to customers.
Rob Collie (00:51:10): But we have tech giants today that aren't being held to the same standard that Microsoft is being held to.
Kasper De Jonge (00:51:17): It's because of the legacy.
Rob Collie (00:51:19): They're in different businesses. So, Kasper, I had a question. If we think of the Power BI CAT team as these smoke jumping ninjas that get called in to solve problems or navigate obstacles for a customer, things like that, how many people are in that talent pool? How many candidates are there? If a new customer question or obstacle comes up, how many potential people can that be assigned to? How many people like yourself?
Kasper De Jonge (00:51:46): So, we don't have a lot of people on the team. And there's multiple different people who are doing different things. I think, probably, there's around 10, 15 people worldwide that are actually talking to customers in scenarios like this. But again, really, the big thing that we're trying to do is because Power BI is growing so massively, it doesn't scale.
Kasper De Jonge (00:52:08): So, we are also trying to scale in different manners in different ways because we want to help more people because, every day, we get emails like, "Oh, my customer, you need to talk to my customer because he has trouble connecting with SAP, using single sign-on with Kerberos." And who knows?
Rob Collie (00:52:27): Yeah. It's sort of Adam when we had him on the show. He was talking about back in support, he could answer support calls or he could go out and blog proactively about the 90th percentile of problems. I think I see where you're going with the scaling thing. Let's talk about that.
Rob Collie (00:52:42): Before we go there, when a problem comes up or a question comes up from a customer, that for whatever reason, the CAT team is going to put one of you on it, how is that assignment process made? Is it just based on availability or is it based on geography or is there some sort of matching of skill set?
Kasper De Jonge (00:52:59): All these things are escalations. Escalations, that's the first thing. It happens like an email comes from James or an email comes from Scott Guthrie or Satya saying, "This customer is having this Power BI problem. They reached out to me. This usually how it goes. And it reached out to me and, yeah. We need to get it solved."
Kasper De Jonge (00:53:18): So, then, we go and look. But usually, it's not really something that we can really do. It's just making sure that maybe sometimes, it's a support case that didn't go the way it's supposed to go or there is a really bad bug for them that just hits their edge case. It hits them, and really hasn't found anywhere else. And the support case got lost.
Kasper De Jonge (00:53:36): And so, we just need to put people to the right people together because we know the right people, you put the right people together, and things get solved, or they have an architectural ready to go start a new project or something that is big spans all kinds of products and things, and they want to just verify it with someone because we go in, and we tell them, "We're not going to do the project for you." That's out of scope. We're not doing it.
Rob Collie (00:54:02): You can't.
Kasper De Jonge (00:54:03): We can advise you. Oh, we can't. And it just doesn't scale.
Rob Collie (00:54:06): Not enough people.
Kasper De Jonge (00:54:07): And there's tons and tons of other people who can do it. There's this whole partner's site on the Power BI website whereas amazing partners. There's plenty of other people who can do it. We're just there. We're going to go in and give them advice. And this is how you can set it up. And, maybe sometimes, we know a bit more about the future than the rest of the people. So, we tell them, "Okay. Maybe, we should start thinking about it in this way or that way." And then, we usually go back out of the way and give it back to the field.
Rob Collie (00:54:32): So, those emails, well, let's say that come from James or whatever, do they go to Mark? And then, now, we've got to sort of figure out is this a Kasper or an Adam or a Patrick?
Kasper De Jonge (00:54:43): Yeah. I mean usually we're somewhat interchangeable, really, most of the folks. They have the same skills and things like this because we never really have to go really deep. Sometimes, it happens like, "Oh, I have this model, and it doesn't fit in my premium capacity." Okay. It's just, yeah. You open VertiPaq Analyzer. And you take a look into how everything is going and how... Yeah.
Kasper De Jonge (00:55:06): I mean most of the team can handle this, those things. But, obviously, when things get really hairy, we have people like Michael Kovalsky on our team who's really deep into analysis services. Obviously, we have Chris Webb. And also for us, when things get really hairy, we can just call up a developer and say, "Guys, come look at this with us. Something really fishy is going on here," and no one else obviously has that backdoor.
Rob Collie (00:55:28): Yeah. So, mostly interchangeable. Every now and then, there's something deep that you would sort of specialize and say, "Okay. This is a Chris Webb or whatever, right?" That's something I was really interested in coming into this, was as we sort of have, at our company, a similar sort of matchmaking need. We have a new client project or whatever.
Rob Collie (00:55:46): And like what you described, most of our consultants, the average job, the average project that we're engaged with is something that really fits many different members of our team. But every now and then, they'll be like, "Oh, we're going to be coding a power app in this one." Okay. We got to make sure that we get one of these couple of people engaged in it then. So, it sounds pretty similar.
Kasper De Jonge (00:56:03): So, we are thinking about because again, for everything, Power BI is growing. The team is growing. So, we're also thinking about growing our team. So, we need to think about something here too. So, we need some deep experts. For example, we have a guy on our team Serge. When you talk security, he knows everything like TLS 1.4. I don't know. He knows all these things and all the encryption keys and how this works.
Kasper De Jonge (00:56:28): And so, when big security questions comes up, we call in Serge. And then, he goes talk. The InfoSec teams love him. He goes in deep. And he talks about all these things. None of us can do that. And probably going forward, we're going to have a bit more people who have in-depth skills because there's also more and more integration with dynamics. And we have someone who helps us with SAP because again that's a whole different ballgame too. So, we do have some expertise. But, in general, everyone has to know models and VertiPaq Analyzer because that's comes up a lot as you can imagine.
Thomas LaRock (00:57:04): Does your team have a need for someone that is on Twitter a lot?
Kasper De Jonge (00:57:08): I think we all are very much a lot on Twitter. So, I think-
Thomas LaRock (00:57:12): Okay. So, or writes LinkedIn posts. How about somebody who just can be critical of the data they're working with?
Kasper De Jonge (00:57:20): Critical data they're working with. Yeah.
Thomas LaRock (00:57:22): Your model stinks. These nulls are horrible, things of that nature. There's no real need for that skill?
Kasper De Jonge (00:57:28): No.
Rob Collie (00:57:28): All of SQL Twitter just cried out in anguish. That's our core competency.
Thomas LaRock (00:57:38): How about just somebody that has for years been able to just look people in the face no matter what and just say, "No."
Kasper De Jonge (00:57:45): No. I think we're really good at saying no. All of us.
Thomas LaRock (00:57:50): All right. So, I'm halfway there.
Kasper De Jonge (00:57:51): Exactly.
Thomas LaRock (00:57:52): All right. I'm halfway there. I can say, "No. I'm on Twitter. I just have to learn Power BI." [crosstalk 00:57:57]
Rob Collie (00:57:59): We've already talked about this in five minutes, you were cranking out the wow and that James Phillips meeting.
Kasper De Jonge (00:58:05): Exactly. You're ready to go.
Rob Collie (00:58:06): Yeah. Because I logged into Power BI. You're hired. All you need to know is star schema. And I think you've got that.
Kasper De Jonge (00:58:13): That is the part of the feedback in our team. Most of the people in my team, you have credibility. And when we go back to the engineering team and we tell them, like, "Guys, this is really a problem . You need to go fix this because, otherwise, it's going to be a massive, massive problem. Hey, listen."
Kasper De Jonge (00:58:32): And that's really important, I think. And, actually, the one thing that I didn't really touch upon, we do have some specialists too because that's really a big problematic area, I think, in probably the whole of the BI industry, is we all love our data models and our DAX and our joints. But if you really can't tell the story, it's a problem. We have two guys in our team who are amazing at it.
Kasper De Jonge (00:58:59): They can take some model and some questions and some things from customers, and they transform it into a UI. I called it the UI, not just necessarily a report, a user interface and a report that looks stunning. And it just tells the story so amazing.
Kasper De Jonge (00:59:16): And that is definitely an area that I think all of us can learn a lot from.
Thomas LaRock (00:59:22): Hey, Rob, we had Hugh Millen. And what did he tell us? He feels accomplished when something he's done that's by presenting data. He finds something meaningful. The two things he says here is, "Is it comprehensible, and do I get from point A to point B? And do I tell that story?" So, it's amazing how this guy, Hugh, and what he's in his background wherever has that same connection to the Power BI CAT team.
Rob Collie (00:59:49): Yeah. This is another episode where a former NFL quarterback is without realizing, he doesn't know, I don't think, that there are books out there and a whole thing about data storytelling and stuff like that. But he talks the talk because he's learned it in the trenches the hard way with his radio show.
Kasper De Jonge (01:00:07): Yeah. It's pretty amazing. And it's really amazing to see how these guys get it. And we're trying to get a lot of that stuff into Power BI too. How can we make it easier for people to tell the story? Actually, I said we're kind of generalized. Actually, now, I'm thinking about it, there are definitely some specializations in the team. And the more we grow, we need to go continue in that way.
Rob Collie (01:00:25): It really does sound very similar to the composition of our consulting team. I'm not saying that we're the same people or whatever. But in terms of like there's this core that is shared, there's 90% core. But there's a lot. That's specialization. There's a long tail of specialization because there are many, many, many different specialties, core, power query, and data modeling and DAX and navigating the Power BI service.
Rob Collie (01:00:51): I mean that's everyone. That's the table stakes. Everyone's got to be really, really, really good at that. Not everyone needs to be the expert on writing power apps or whatever. I really like, by the way, that you're security expert. Even his name sounds like security, Serge. It's just like if you say it really fast, it sounds security. It's like he was born for this.
Kasper De Jonge (01:01:11): Hey, I think he is. It's pretty amazing all these things that are happening there. And I think the way I sometimes say it, we are tip of the spear. We see things months before everyone else because we hear this from these big customers. We're doing these crazy things with Power BI that no one has ever thought was possible. And they're running into problems. And they're turning into things that no one haven't even thought about.
Kasper De Jonge (01:01:34): And then, we try to help them in a way that makes it work for them. And if it doesn't work, we need to bring it back to the engineering teams. And, for example, a lot of the feedback that we're giving today to the engineering teams, look, I run a meeting every month with all the Power BI leadership team like Amir and Arun and all the PMs are there where we share the feedback that we're getting from the field, that we're getting from the customers, that we see on Twitter, as you know where everyone is on Twitter a lot in my team.
Kasper De Jonge (01:02:03): But also, we interact with people. And they come up. They ask questions and, yeah. We give that feedback back and say, "Guys, now, we're seeing this, and this is a problem." And they're taking that into account when they're going to do their next semester plan. So, Power BI does... We only think we do six months. We work. We plan for six months. We do the work.
Kasper De Jonge (01:02:23): And then, we're going to plan the next six months of work. So, we're never going to go at further than that. Obviously, there are some longer term goals and visions than things that people want to think about. But all of that feedback comes in. So, all the burning questions that customers have right now, I mean they're probably going to be solved in the fall of this year. That's how quickly things go. And that's pretty amazing.
Rob Collie (01:02:45): I really wonder how hard it would be for me to adapt to that kind of software pace because I lived the old pace and left before the transition. You got to live through the transition with everybody else. I bet that would be a real culture shock for me being dropped in that river.
Kasper De Jonge (01:03:01): It was hard for a lot of people. And there's people who still can. The specs and the spec reviews that we used to do have to do. I mean you had documents and documents, pages and pages of documents describing everything. And then, you have spec reviews where all the engineers would come in the room. And they would beat you up on the stuff that you didn't think about.
Kasper De Jonge (01:03:22): I mean that doesn't happen anymore. It's not there. We have a one-pager. And you have some docs in note. And you work on it collaboratively. And it's get taken as it goes along really.
Thomas LaRock (01:03:33): That might be a lot of fun actually. You describe it that way.
Kasper De Jonge (01:03:34): It is. I think the author was fun too because this stuff I learned from those spec reviews and the things and not just technically, but also communicating things well. And I always tell people after I've seen House of Cards, I said "Okay. This is Microsoft." This is the way it used to be.
Rob Collie (01:03:50): Oh yeah. A lot of maneuvering, machinations.
Kasper De Jonge (01:03:54): Exactly.
Thomas LaRock (01:03:54): That show is eye-opening about power, I think. I learned a lot from the first two seasons there.
Rob Collie (01:04:01): Those specs you talked about, there was actually a moment back in the day when we were doing the dreaded stack ranking meeting that Microsoft doesn't do anymore. Get in a room and have to rank 30 people from one to end in terms of how valuable they've been for the last year. It's just awful. Just the worst thing ever. And our director of program management at the time, he introduced this new wrinkle which was all the managers, we had to print out all of the specifications, all the specs that our people had written over the past year and bring them into the room.
Kasper De Jonge (01:04:32): What?
Rob Collie (01:04:32): And I swear. I swear he took them all and assembled them and almost physically weighed them. And there were all these stacks of specs. Some of them were small. Most people's stacks of specs were sort of in the same range of height. And then, you get to the end of the list, the end of the row, sit on this giant table. And there was this one stack that loomed over all the others and ruled them from on the high like a tyrant. And that was our friend, Alan Folding. Alan Folding stack of PivotTable and related specs. And you flipped through them. And they were not padded with fluff either. I mean Alan had just... He had outspecified.
Kasper De Jonge (01:05:12): I can only imagine.
Rob Collie (01:05:15): And as silly as that exercise sounds, there actually was something objectively valuable about it. You would have been able to see just how much work Alan had done, if you hadn't. I mean, of course, there's all kinds of things. It's really hard to reach the decisions to write a short decision down. And so, it probably wasn't fair to the people who had small stacks. But Alan's large stack of specs was 100% legitimate.
Kasper De Jonge (01:05:42): I can only imagine.
Rob Collie (01:05:45): Yeah. Earlier, the very beginning you said five minutes to wow. You repeated that a few times. I'm just sitting here laughing to myself because to you and I, the word wow means something else too, doesn't it?
Kasper De Jonge (01:05:56): Yeah. World of Warcraft.
Rob Collie (01:05:59): Yeah. So, have you gotten into Classic?
Kasper De Jonge (01:06:01): No, no, no. I'm not touching that thing anymore.
Rob Collie (01:06:04): You're clean. You're clean now.
Kasper De Jonge (01:06:05): Oh yeah. I've cleaned this for many years.
Rob Collie (01:06:07): Yeah. I was. I was clean from 2009 to August of last year. And then, I made a deal with my wife. I will give up watching the entire NFL football season in exchange for being allowed to play World of Warcraft Classic two nights a week with my old buddies just as a social outlet. And she was terrified because it's like you're letting that beast back into our lives. But she hates me being gone for football on Sundays.
Rob Collie (01:06:39): I just disappear for eight hours. She's a football widow. And it's been a positive trade. I've been getting my Warcraft. I've got another rogue. I've been raiding knacks.
Kasper De Jonge (01:06:53): No. I'm not touching that thing anymore.
Rob Collie (01:06:55): Yeah. We want you back. Come on, Kasper.
Kasper De Jonge (01:06:58): Yeah. Adam says the same thing, Adam Saxton.
Rob Collie (01:07:01): Really? Kasper, first taste is free. Come on, Kasper.
Kasper De Jonge (01:07:05): Yeah. I know. I know.
Rob Collie (01:07:09): It's bad.
Kasper De Jonge (01:07:10): I did grab myself an Xbox. So I'm playing with that.
Rob Collie (01:07:13): Fine. I mean if you want to play Xbox, a game that you can turn off whenever you want-
Kasper De Jonge (01:07:18): Exactly.
Rob Collie (01:07:19): It doesn't creep into your life. The only way that I can make it work playing two nights a week, this will tell you everything you need to know about World of Warcraft. The only way that I can sustain an entertaining two nights a week of playing is to buy in-game currency on the black market on a regular basis. Two weeks, I've got to go spend 60 bucks on gold. Otherwise, it's a job. You've got to go farm stuff so that you can play those two nights a week. You can't just play. You've got to have the materials. Yeah. It's a beware. Tom, you're about to say something.
Thomas LaRock (01:07:58): I wanted to be a bit cheeky. I'm looking to blame somebody for something. So, Kasper, are you responsible for giving it the name as your Purview? Was that you?
Kasper De Jonge (01:08:09): No.
Thomas LaRock (01:08:11): No? Are you sure? Are you sure that wasn't [crosstalk 01:08:13].
Kasper De Jonge (01:08:13): I'm very sure.
Thomas LaRock (01:08:14): Somebody at Microsoft did it. And I plan on asking every employee until I find the person that thought Purview was the go-to market name.
Rob Collie (01:08:23): What's wrong with Purview? Come on, Tom. Out with it. What's the problem here?
Thomas LaRock (01:08:26): I feel out of all the possibilities for naming that particular security product, Purview should not be on any short list. Now, we have Microsoft Mesh to go along with the Azure Service Fabric Mesh because if you're going to overload a product name, Mesh is a good one to start with, I guess. Yeah.
Rob Collie (01:08:48): And by the way, Mesh was a previous Microsoft research product that carried that name as well. It was based the forerunner to OneDrive. It was the first peer-to-peer file sharing service that Microsoft put out. They called it Mesh. And we were using it.
Kasper De Jonge (01:09:02): I saw a tweet by Ray Ozzie about it. I think he was the one who came up with the original.
Rob Collie (01:09:07): But I think that's right. Yeah.
Kasper De Jonge (01:09:10): And he said he's on Twitter. Nice name. I like it. I must say in their defense, naming is hard. super hard. I remember in when I was in Power Pivot and I had to come up with names for the buttons or drop downs or, oh my god, it is hard because you do user research. And you ask people. And everyone thinks differently about what names are, what they mean. Oh, yeah.
Rob Collie (01:09:37): Yeah. If you want proof that naming is hard, just look at all the names Microsoft has chosen over the years. I tell people that Microsoft is amazing at building software. They are the best at building software. And they are bad at basically every other thing.
Rob Collie (01:09:55): And naming is absolutely one of them. Naming is hard. But also, Tom's right. Naming doesn't have to be quite as hard as Microsoft makes it for themselves. I think there needs to be a new engineering discipline at Microsoft that is just dedicated to naming.
Kasper De Jonge (01:10:11): You know it's not engineering. It's marketing.
Rob Collie (01:10:13): But there's marketing within the product. Even the product sort of has to market in a way, market its own buttons to the user. It's a tricky thing. And the same people who are deciding how multi-directional relationships need to flow and how they need to behave, five minutes later, they're being asked to name a button. And it's a hard contact switch.
Kasper De Jonge (01:10:34): That has changed a lot too in the last couple of years. Before, we had one designer sitting somewhere in the corner who were doing some research. Now that it's completely gone, we have proper design teams. And we have people who actually know what they're doing instead of getting some PM like me who has no idea around this to come on board and to think about naming. Now, we have actual designers who have actually have a design degree to come and do it.
Rob Collie (01:11:00): I think it shows. I recently upgraded to one of my computers anyway. I did the switch to the new model view. And I feel the arrows on the relationship lines are, now, they're just a little too hard to see whether it's one directional. That block is really big. And the arrow is kind of small. There's still some room.
Kasper De Jonge (01:11:20): There's lots of feedback already on that one. And that's the good thing though. Yeah. People look into it. And that's I think the good thing. Yeah.
Rob Collie (01:11:27): And you really are on top of this stuff. I am so distant now. It's hard to even imagine a huge chunk of my day. Most of my day is not spent with these tools anymore. It's company stuff that I'm doing most of the time.
Kasper De Jonge (01:11:41): Yeah. I can imagine.
Rob Collie (01:11:43): I'm not one of the consultants. The consultants on our team are far more capable than I ever was. And I'm also distant from the software team. I have this weird... This history of being so tightly tied to all of these things, I keep a prize of some of it. But it's just there's no way I can keep up.
Kasper De Jonge (01:11:59): That's really the fun thing about the role that I'm in and the position that I'm in because I came from the engineering team. I know a lot of the people who were there and built error. So, I have connections with them. I spent time working on projects with them. But now, I spend a lot of time with the community. I see what's going on. We have a huge internal team site over 3000 people from the field who are every day talking to customers.
Kasper De Jonge (01:12:23): And they're asking questions to the probably a CAT team there. And we're trying to answer. It forces me. Today, I had this question about VNets. I didn't know what VNet was. But I had to go figure it out. What is a VNet and how does it work?
Kasper De Jonge (01:12:37): But I do not know DAX anymore the way I used to because I haven't really written it. I haven't really done it the way that I used to be able to. It's more superficial and more broad.
Rob Collie (01:12:51): I sympathize. I'm definitely off peak. My DAX skills are not what they used to be.
Kasper De Jonge (01:12:57): Yeah. And so, much has changed too.
Rob Collie (01:12:59): That's true.
Kasper De Jonge (01:13:00): I have to watch Marco and Alberto's recordings to see what's going on.
Rob Collie (01:13:03): Yeah. The style of people's formulas has evolved quite a bit. There's actually stylistic variants of how you write DAX. After a while, if you can look at someone's DAX and go, "Oh, I know who wrote this," there's that much personalization in terms of how you go about doing things.
Rob Collie (01:13:22): And my way of writing DAX is very, very, very straight to the point, straightforward which isn't necessarily the most efficient in terms of how it runs.
Kasper De Jonge (01:13:31): But often, with the access, this is the way. If you have a certain patterns to solve problems, that's, oh, wait, you're going to go back to the same thing. It's usually the same thing really. Overwriting some filters in a certain way, and that's really what you need to do. And if you need to do some market basket or whatever returning customers, you go online. And you look at some examples. I sometimes come back to your old blogs from 2011. And then-
Rob Collie (01:13:57): That's true [crosstalk 01:13:58].
Kasper De Jonge (01:13:57): This is how you did it.
Rob Collie (01:13:58): Yeah. It is amazing. The power of a DAX pattern is really something to behold. Back when I used to teach classes, I used to tell the Excel crowd this. It's like, "So, if you go on to Google and you search and you find an Excel formula that does something that you didn't know how to do, very often, you don't understand how that Excel formula works."
Rob Collie (01:14:17): When you're googling something you don't know how to do, almost by definition, you're finding a technique that you don't understand. If that technique happens to just fit your data and fit your existing question perfectly today, you can just copy paste. It's a black box. I don't know how it works. But it works. Of course, then, when the road bends and someone asks a slightly different question as they inevitably will, now, you're exposed because you need to edit that formula to fit this new context in Excel because you don't understand it. you have no idea how to edit it.
Rob Collie (01:14:49): Beautiful thing about DAX is that you might not understand the pattern just like you did in Excel. But then, when someone asks a different question, all you do is just rearrange the field list. You just drag, drop. And it responds to the new context and answers your question. And it still works.
Rob Collie (01:15:04): And there are patterns you don't have to understand in order to use them effectively. It's easier and better to steal DAX than it ever was to steal Excel.
Kasper De Jonge (01:15:15): Yeah. By the way, Rob, have you read a calculation group?
Rob Collie (01:15:18): I have not.
Kasper De Jonge (01:15:18): Oh, that's going to blow your mind.
Rob Collie (01:15:21): Is this essentially like the calculated member or calculated set type of thing?
Kasper De Jonge (01:15:26): Yeah. From MDX world. It's a fundamental shift on how to write DAX, and the possibilities and things that you can do.
Rob Collie (01:15:34): So, is this sort of, let's say, I write a measure that is current year-to-date versus prior year-to-date, that's for revenue. And now, I want that same thing, but for-profit margin. Before, I have to copy-paste that whole formula and just change a couple of the base measures in it. This allows me to sort of not play that copy paste game to the end [crosstalk 01:15:58].
Kasper De Jonge (01:15:58): Exactly. One new function that is the trick it's something called selected measure.
Rob Collie (01:16:02): I love it.
Kasper De Jonge (01:16:03): And whatever measure you put into it, you can do whatever you want with it. That's opening up so much different things and so much improvements that we-
Rob Collie (01:16:11): I love this. This is one of those things like when you layer it on as a sort of the software elegant architect type, we'll look at this and go, "Layering this concept on later a Band-Aid, this is so gross." But really, it's genius because if the DAX language had been built from the very beginning to include this concept, it would have been so much more complex at its base.
Rob Collie (01:16:38): DAX could have been built to be the new MDX that was this giant learning cliff at the beginning rather than a curve. I would have face planted into that cliff.
Kasper De Jonge (01:16:46): People tried.
Rob Collie (01:16:47): Yeah. I know they did. They do. But you can still walk up to it today and write a simple sum. You can write a simple if in the same way that you would write in most cases anyway, the same way that you would write an if in Excel. And that was all the difference in the world for me versus MDX. There is no simple. There's nothing simple.
Kasper De Jonge (01:17:06): I don't know. I don't agree with that. I mean I used to write MDX too. Actually, we had an interesting discussion yesterday with a lot of the folks like Christine Wade and Chris Webb. I've talked about MDX versus DAX. I mean they're definitely things that were easy there too. And there were patterns there too. But especially being able to use hierarchies. But the biggest problem is before you get there, you have to get your model in such a shape that you can actually do it. It needs to be hierarchical. Your data needs to be in a certain way.
Kasper De Jonge (01:17:35): You have to define everything and process it. And then, it works. Now, you can write the code.
Rob Collie (01:17:41): Oh, I think I saw. This might not be the same conversation. But I saw a conversation like this going on, an email distribution list with these people recently, MDX versus DAX. And I wanted to jump in and say, "Oh, my gosh. No." If you're someone that was able to learn MDX in the first place, you are not qualified to speak on how easy it is relative to DAX. I understand that once you know it, there were things that were easier to do in MDX than in DAX.
Rob Collie (01:18:08): But the whole point that I wanted to make was like, "But I never could learn MDX. I just couldn't even get in the door." I couldn't get into the door to discover that certain things were easier. The fact that I went from not being able to learn MDX to writing a best-selling book on DAX, it's kind of the end of the story in terms of which one's easier. But at the same time, I understand that there are concepts in MDX or facilities in MDX that we miss in DAX.
Rob Collie (01:18:36): And we're sort of going back, not we. I have nothing to do with it anymore. But we're experiencing this as we sort of go back and we're layering some of those things back in the same way that Excel is a programming language like formulas in Excel or a programming language. But Excel has forever lacked many of the things that we take for granted in "real programming languages." And a lot of those concepts, you sort of see them get layered back in over time with the lambda functions recently and dynamic arrays.
Rob Collie (01:19:06): It's the inspiration for ideas. But none of those come back and say, "Let's reinvent the Excel grid into something more scary." I think that's good. You maintain that learning curve that you can sort of forever perpetually climb incrementally. That's a really good feature of a lot of Microsoft's platforms. And it's certainly true of DAX. It's true of them. These things are wonders of the world.
Kasper De Jonge (01:19:32): I think probably the most fun part of working on the Power BI team is being able to talk to people who were never even an IT person. The guys I went to school with or the girls I went to school with and the people I worked with in the beginning as programmers and all these things, the people I meet now, they're just different people. And they're also still there because when you want to write the really big models and IT projects, it's still the same people that we had before.
Kasper De Jonge (01:19:56): But now, you get people from the left field. They were not working on this. And actually in my team, we have a few. Mark, my manager, he was a finance guy. This is an amazing story. You should probably get him on the show. It will be very entertaining.
Kasper De Jonge (01:20:08): He comes from a finance background. And he did all these amazing things. And I remember back in 2012 or something, he started talking about power view and the way that you can do things, you can visualize it. And he started changing the way they were running because he was a finance manager at Microsoft. He was doing finance discussions.
Kasper De Jonge (01:20:29): And he was having these, I mean you've heard the story many times before, the gazillion Excel spreadsheets. And he replaced it with one Purview on top of some data that they've created. I think that's one of the most cool things that we see in the work that I do.
Rob Collie (01:20:44): And you wouldn't really expect this because I was part of the software engineering team. But that's really me. It's kind of in get shorty, they say to John Travolta, "But you've been a wise guy all your life." And he goes, "Yeah. But I was never that into it." I was not the techie. I'm more into tech than the average person, I suppose. But compared to the average Microsofty, I wasn't the one that had a Windows media PC at home.
Rob Collie (01:21:05): I wasn't coming in every day saying, "Hey, look at this neat little script that I wrote last night." No. So, for me, I would say that in all the ways that are important, my transformation into an advocate for this stuff and a practitioner of it and all that, is almost exactly the same as Mark's story. They're really the same thing.
Rob Collie (01:21:24): I am one of those citizen developers that we talked about with Chuck. Like you say, those are the most compelling people to meet and interact with. They're just the most fun and the most gratifying.
Kasper De Jonge (01:21:36): Yeah. Exactly.
Rob Collie (01:21:36): I can't get enough of that.
Kasper De Jonge (01:21:37): I mean the opportunities that you get, you guys have Shannon and Stephanie on the show too. I went with Stephanie to Kenya. I did a week-long Power BI training for the people who work there. And that was just amazing to see the things that they are now able to do. I mean I don't mind that either. But it's not for the shareholders. It's to help pediatric aids. That's a whole different ball game.
Rob Collie (01:22:06): Yeah. That's not something that on the software team, even though you're the one building the tools that do that for everybody, there's no way you're ever going to get any significant exposure to that real world impact. And that's one of the things that I've really enjoyed about switching to the consulting side of the equation, is that that's basically our whole life now, is living that directly appreciated. We can see the change in people's lives. It's just the emotional component of this job is actually quite a bit more satisfying than building software ever was. And that's a little surprising. But it's indisputable.
Kasper De Jonge (01:22:42): That also has changed significantly. I remember way back when before we also started the Power BI journey, it would have some tap customers. You remember? And then, he would talk to them a couple of times before the release. And that's about it. But now, we talk to customers all the time. And we have people telling, sharing their stories. And that is I think again one of the really nice things of the things that's going on now. You actually see it way more. And it's more way more practical. And I think that's good.
Rob Collie (01:23:12): Yeah. And I believe that. I think sort of the shame for the software team is that most of the time, that's going to be talking about the pain points. That's where the attention sort of should be professionally focused, is on the pain points. In the consulting world, they have a pain point to begin with. But we're there for the triumph. We're there for the revolutionary impact. And those are good feel-good stories. But as the product team, you really can't spend a lot of time wallowing in that. You've got to focus on the problems. And so, t's kind of a happier job being on the front lines of it.
Kasper De Jonge (01:23:51): Actually, I had a discussion with one of my colleagues today who's pretty new on the team. We are the shit-solving team. So, we hear all the problems. And, usually, we cannot do anything about it because we're not the developers ourselves. When you're a PM on the team, and you can get angry and you can walk over to the engineers, you can do something about it. We cannot. And that's sometimes frustrating.
Rob Collie (01:24:17): Even when you can solve the customer's problem, again, just about the nature of your job, you're not there when it actually gets solved.
Kasper De Jonge (01:24:24): We won't hear from them anymore when the problem is solved. It's like they drop off because... Yeah. Thank you.
Rob Collie (01:24:30): That's got to be at times, like a torturous purgatory.
Kasper De Jonge (01:24:34): It does. Also, we have more ongoing connections with a lot of customers too. We have programs where we spend a lot of time with our customers. And we build better relationships with them. Things have definitely changed. But I totally get to just saying, as a consultant, I used to be a consultant too. It's definitely... Yeah. You see the big change that you can get.
Rob Collie (01:24:58): When's the last time you played basketball?
Kasper De Jonge (01:25:00): Well, it's corona. So, no. We're not allowed.
Rob Collie (01:25:03): You'd totally be allowed to play basketball in the United States. I mean at least here, I mean high school football is still going on.
Kasper De Jonge (01:25:10): So, I can't even go. It's almost 9:00 PM here in the evening. I'm not allowed to go. We have a curfew.
Rob Collie (01:25:17): Wow.
Kasper De Jonge (01:25:18): So, I can't go out.
Rob Collie (01:25:21): Oh, your chances of going out and ruining your pass fail credit score go down too, don't they?
Kasper De Jonge (01:25:26): Exactly. So, I was playing basketball every week. I found a really nice team here.
Rob Collie (01:25:31): So, post-COVID, some point in the distant future when we're all in the same place, me and Kellen versus Kasper and Tom, two-on-two.
Kasper De Jonge (01:25:43): Sure.
Rob Collie (01:25:43): Tom's a bit of a ringer. I'm probably the Not probably. I'm definitely the worst of the four of us. I'm not stacking the teams in my favor. I can't stack the teams in my favor because, by definition, I'm always on my team. But that'd be fun. Kellen can play. So can Tom.
Kasper De Jonge (01:26:01): Thomas is not saying anything.
Rob Collie (01:26:03): No. We lost him.
Thomas LaRock (01:26:05): I'm not going to comment. I haven't played in so long. I like hearing Rob just say that I can play. That's all I need right now. But I'm old. I haven't played in a while. And it's not my chair [crosstalk 01:26:19]-
Rob Collie (01:26:19): Me either.
Thomas LaRock (01:26:19): ... you hear squeaking. It's actually my bones.
Kasper De Jonge (01:26:23): I didn't play. All the time when I was in Seattle, for some reason, I just couldn't find... I couldn’t go to the gym and play with these pickup games. I just didn't like it. When I come back to Netherlands, we find a nice team. And I started playing again. But I had to work out because if you're not in a good shape, I mean you're going to get in trouble. You get hurt all the time and all these things. So, I got into shape. And I started playing. I mean it's the highlight of my week really, just playing for an hour and a half with the guys. It's just amazing. I love it.
Thomas LaRock (01:26:52): Yeah. Basketball is fun. I miss it.
Rob Collie (01:26:55): COVID isn't the reason I'm not playing. I haven't played in six years.
Kasper De Jonge (01:27:00): I saw something new on ESPN yesterday [inaudible 01:27:04]. It's a combination of basketball and wrestling.
Rob Collie (01:27:07): Oh really?
Kasper De Jonge (01:27:08): Yeah. Apparently, it's something new that they're starting doing in Russia.
Thomas LaRock (01:27:13): Of course.
Rob Collie (01:27:13): Why would this remind you of me?
Thomas LaRock (01:27:15): I don't know.
Rob Collie (01:27:19): Because I cheat?
Thomas LaRock (01:27:22): I remember the chicken wings.
Rob Collie (01:27:24): You do the chicken wing. You know what? I didn't even know I did that at all until my friend, Ben, pointed out to me back here in the late '90s that like, "Oh, you're always doing this chicken wing." And then, you brought it up in New Orleans. I'm like, "Yup. Apparently, I'm a chicken wing player." Isn't that a big a position in basketball now? They talk about the wing, right? Is that what I would be, the wing?
Kasper De Jonge (01:27:55): The wing, no.
Rob Collie (01:27:56): It's good to talk to you again, Kasper, and good to see you.
Kasper De Jonge (01:27:59): Yeah. We talk more often.
Rob Collie (01:28:01): COVID is on the wind down. We'll be seeing you back over on this side of the ocean at some point. There will be in-person conferences again. I believe it.
Kasper De Jonge (01:28:10): I hope so. I really hope so.
Rob Collie (01:28:12): There has to be. It's almost like we need a party or a victory lap sort of conference circuit just to compensate for all the time we've spent inside over the past year. I really appreciate you making the time especially in the evening for you.
Kasper De Jonge (01:28:25): I really enjoyed it.
Rob Collie (01:28:26): Thank you so much.
Thomas LaRock (01:28:27): It's good to get caught up.
Rob Collie (01:28:28): Thank you, guys.
Announcer (01:28:29): Thanks for listening to the Raw Data by P3 Podcast. Find out what the experts at P3 can do for your business. Go to powerpivotpro.com. Interested in becoming a guest on the show? Email lukep, [email protected] Have a day to day!
Sign up to receive email updates
Enter your name and email address below and I'll send you periodic updates about the podcast.