Raw Data By P3 Adaptive
One of the World’s Most Amazing Humans, w/ Matt Allington
Power BI Consultant, Trainer, & Microsoft MVPListen Now:
Matt Allington is a Power BI consultant, Trainer, Microsoft MVP, author, and a friend! Matt has worked for many years in the retail and CPG industries, has a very deep knowledge of Business Intelligence and Data Analytics, and currently owns a Power BI consultancy ExceleratorBI in Australia.
Here’s the link to Matt’s book-Supercharge Power BI: Power BI Is Better When You Learn to Write DAX
2:00 – Did Matt and Tom the SQL Rockstar previously meet each other?
6:25 – Matt’s origin story is an epic and relatable journey of discovery
14:00 – The importance of process (and we adopt Matt’s superior pronunciation of the word)
25:35 – Business, Sales, and IT (and once again the importance of Hybrid roles)
29:10 – The evolution of BI tools, and the best and worst of the bunch
37:15 – Structure VS Flexibility in BI tools
48:55 – Power BI and Power Query enter the fray, Excel people get angry, and everyone has the same opinion of Power View
53:50- The friction Tom got from adding Business Analytics to conference menus, and the birth of BACon
1:08:45 – Matt’s connection to Rob’s broken leg
1:11:05 – Training remotely in the COVID world, and The Collie Layout Methodology
1:16:20 – Rob takes exception to Matt’s math
Rob Collie (00:00:00): Welcome back friends to Raw Data. Today's guest is Matt Allington. He comes to us from Down Under. Now, Matt has a very particularly validating story for me. I bet my career in 2010 on the future success of Power BI. I just knew it. I could see it. And even though I was convinced that this was of the future, in 2014, when I crossed paths with Matt Allington, something happened. It was still, even though I believe what I was doing, was still incredibly validating.
Rob Collie (00:00:30): This guy was a BI director for Coca-Cola, Asia Pacific BI director. As far as I was concerned, one of the sort of made men of the elite of the traditional industry. And after a brief interaction and exposure to what I've been working on, he called me up and said, "Hey, I want to go do what you're doing." And like I said, even though I had been so convinced of what I was doing and it was already working out, it was going great, it was still a really validating moment and just an incredibly exciting feelgood moment for me to see someone sort of cross over like he was doing.
Rob Collie (00:01:04): Now in today's podcast, you'll hear that he actually... If I'd known more about him at the time, it wouldn't have been quite as much of a surprise to me. So a lot of things that we talk about on today's pod are things that I was hearing for the first time. It was really interesting even to me, after knowing Matt for as long as I have. We had a lot of fun. Hope you enjoy it as well. So let's get after it.
Announcer (00:01:26): Ladies and gentlemen, may I have your attention please?
Announcer (00:01:30): 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 (00:01:49): Welcome to the show, Matt Allington. Sometimes we like to say the, the Matt Allington.
Matt Allington (00:01:55): Thank you. Great to be here.
Rob Collie (00:01:57): Matt, you were at at least one or two of the past business analytics conferences. Tom, had a lot to do with organizing those. Did you guys ever meet at one of those?
Matt Allington (00:02:07): Yeah. I seem to remember passing Tom in a hallway. I actually think Tom from memory, you had just been elected PASS... Whatever the thing is that you got elected PASS leader is it or chairman or something? President.
Thomas LaRock (00:02:23): President.
Matt Allington (00:02:24): President. Okay, yeah. So I think it was that year. I think you'd been elected PASS president. And Rob, I was over at the PASS conference. I want to say it was in... No, it wasn't in Seattle. It was up in Santa Fe. San Jose.
Thomas LaRock (00:02:40): San Jose.
Rob Collie (00:02:41): San Jose.
Matt Allington (00:02:42): Yeah. San Jose. I reckon it was that year. Does that sound familiar? Do you remember-
Thomas LaRock (00:02:46): Yes.
Matt Allington (00:02:46): ...passing me, Tom? You remember I waved and wanted a selfie?
Thomas LaRock (00:02:49): I absolutely remember you.
Rob Collie (00:02:51): Oh, yeah. Matt, that guy.
Thomas LaRock (00:02:54): That Matt.
Rob Collie (00:02:54): The waiver. The mysterious waiver.
Matt Allington (00:02:57): Yeah, the fanboy.
Rob Collie (00:02:58): You're both figures in the community, in the data community, which is sort of coming at it from different sides. I was actually really curious coming into this whether you had to interact. I mean, you're familiar with SQLRockstar on Twitter. Yeah?
Matt Allington (00:03:11): I'm a follower. I'm a follower of SQLRockstar, so yeah.
Rob Collie (00:03:15): Really? And you still agreed to be on the show.
Thomas LaRock (00:03:17): Yeah, absolutely.
Rob Collie (00:03:18): Even though he was here. Yeah.
Thomas LaRock (00:03:19): Yeah, that's weird.
Rob Collie (00:03:21): I'm disappointed. The two of you never really, truly crossed paths other than the waving.
Matt Allington (00:03:25): Yeah. We haven't really ever met other than sort of through Twitter, but the Twittersphere is actually quite interesting because you really sort of... I feel like you can build a bit of a relationship with people and they're not the same way as a face-to-face relationship, but you sort of feel like you get to know people through these social media channels.
Matt Allington (00:03:45): In fact, it was Scott, Scott [Zinkaresky 00:03:48] at that exact same conference that we're talking about that talked me into signing up to Twitter. And you know Scott. You can't stop him, right? So he said-
Rob Collie (00:03:56): No, you can't.
Matt Allington (00:03:57): ... to me that this Twitter thing is the bees knees. Literally, I think even if I check my start date, it was probably on that evening when we were at the bar where I set up my Twitter account or very shortly thereafter. But yeah, I think you can really get to know people. When I have met people at conferences, you say, "oh yeah, I know who you are." Right?
Rob Collie (00:04:18): Mm-hmm (affirmative).
Matt Allington (00:04:20): I'm a real believer of putting a like image on your social media accounts, on your professional, social media accounts anyway. So Twitter and LinkedIn. If you want to do Facebook, something a bit off the wall butt I think that's a bit different. But as far as your professional accounts go, I think it's great to put a like image of yourself. And then when you go to these conferences, people recognize you. Right? And then if you've had a bit of interaction, you sort of feel like you're not strangers. I think it's great.
Rob Collie (00:04:50): Yeah. I was really surprised when I met Tom, for instance, that there weren't like these multicolored stars in the background behind them. It was really off- putting. I'm like, "Could you like halfway put your hands in your pockets? Oh yeah, there is Tom. Yeah." I was trying to think, what would else would we do with our image? Like a glamor shot or something. But no, Tom's got a professional... It's an icon. He's iconic.
Matt Allington (00:05:15): [crosstalk 00:05:15]. Sorry, about that, Tom.
Rob Collie (00:05:18): I was.
Matt Allington (00:05:21): Everyone that has a like image though is like 10 kilos, sorry, 20 pounds lighter and about 10 years younger.
Thomas LaRock (00:05:30): That's a data guy right there. He just did that conversion.
Rob Collie (00:05:33): Look at that, effortless.
Thomas LaRock (00:05:35): Seamless.
Rob Collie (00:05:38): He's a natural. It's a pleasure to watch you work. It depends. I was probably 10 kilos lighter six months ago. There's some peaks and valleys.
Matt Allington (00:05:50): Then you spend a bit of time at home with a few boxes of delivered donuts and [crosstalk 00:05:55].
Rob Collie (00:05:55): That's right. You know how food tastes better when you're camping? Delivered food sort of tastes better too. Delivered donuts are like an extra 20% above normal donuts.
Matt Allington (00:06:04): Extra sweet, yeah.
Rob Collie (00:06:05): So Matt has a very interesting origin story, I think in the Power BI, the power platform space. We are obligated to at least cover this. I've retold your story so many times because I just find it so compelling. Retell it from your perspective.
Matt Allington (00:06:21): Well, maybe I'll perhaps go back a little bit further than I have before when I've talked about this story. Because I mean, the bottom line is I met you, Rob online virtually or via probably telephone back in those days. Nearly seven years ago, right?
Rob Collie (00:06:36): Yeah. It's a long time.
Matt Allington (00:06:37): I'll come back to that. But we were talking before about data, right? This is about data. So the truth is I'm a data guy locked in a business guy's body. That's actually the true story of Matt Allington. Now, if you want to go way back, I remember in high school and in Australia year eight was the first year of high school. We had a computer lesson. Now, this is, I want to say, 77 orders of magnitude. Right? So I'll give you some sort of sense of time. So the PC was released in, what, '80 or '81?
Rob Collie (00:07:15): Yeah.
Matt Allington (00:07:16): So this gives you a sense of the time horizon. Now, we had a computer class once a week and we had this HP something mini-computer. I think they were called mini-computers back then. It had a cassette. You know what a cassette is, right? It's like plastic. It's got holes in it. Spins the tape through. So that's how you used to save your software. I got a library book out and programmed something in Basic. I taught myself Basic to use this thing. I loved it.
Matt Allington (00:07:51): I'd go in after school and program this computer. Then that was really the start of my love of data and computers, but I didn't go down a data and computer and IT path like many people who have that sort of gene tend to do. I'm not saying it's right or wrong, I'm just saying, this is what happened to me. So I often talk about there being a continuum of skills or styles, or genes, or whatever you want to call it.
Matt Allington (00:08:24): So at one end, you have the most extreme IT person that you can think of. This is the person that does the machine coding, right? The person who developed the operating system. The person who wrote in, in this code that you can't even read and understand because it's directly addressing memory and peripherals and things like that. So that's at one end. And then at the other end, you've got this business person who think of the entrepreneurial, strategic CEO that's never touched a piece of technology in their life.
Matt Allington (00:08:56): They do everything on a whiteboard with a whiteboard marker and they sell the vision and then someone else basically makes it happen. So these are the leadership people. I'm not trying to say these things are mutually exclusive. I'm trying to say that there's this continuum. And then everyone falls somewhere in between that, right? So you could at least philosophically put a line somewhere down the middle of that continuum and say, "At this point, you are now an IT person."
Matt Allington (00:09:26): On the right side, you're an IT person. On the left side, you're a business person or a commercial person. I've always straddled that line. I could have fallen either way. I mean, the truth is there's a gray space, right? So even though there's a cutoff point, there is a gray area where you could be a business person with an IT bent, or you could be an IT person with a business bent.
Matt Allington (00:09:49): So if you're an IT person with a business bent, you're a business analyst, right? So you understand the technology. You're actually an interpreter. It's no different to the early days when we had some people speaking French and some people speaking English, and then some people would do both. And those interpreters sat in between and they listened to one person and they translated it and explained it to the other person who couldn't understand.
Matt Allington (00:10:15): That's what a business analyst does. Right? They're an interpreter because they're in this gray area between the most extreme IT person who can't understand business and the most extreme business person who doesn't understand IT. We need these interpreters.
Matt Allington (00:10:30): I've always been in this gray space and I could have fallen either way. Frankly for economic reasons and luck, and whatever, I just spent my time in the business space. So I started working in a supermarket chain called Woolworths.
Matt Allington (00:10:50): Now in the US, Woolworths is a bit different to what it is here in Australia. So Woolworths in Australia is more like Kroger or Publix. One of those more mainstream supermarkets here in Australia. Whereas I think Woolworths is more like Walmart in the US. So just keep that in mind.
Matt Allington (00:11:08): But I started working in a supermarket and I learned about customer service. I remember getting my butt kicked by a manager one day. A customer came in and said to me, "Do you have any of this frozen chicken number 11 or whatever?" And I looked in the freezer cabinet. I said, "No, we don't have any of that." This is early days in customer service.
Matt Allington (00:11:30): So of course she went and spoke to the manager and the manager came and kicked my butt and said, "Don't you ever look in the freezer. I mean, the customer can see there's nothing in the freezer. Your job is to go and service that customer and find out if there's any out the back." Geez, did I learn a really valuable lesson about customer service. Seriously, for probably eight years that I worked in supermarkets casually and full-time over the years, I really learnt what it's like to service a customer.
Matt Allington (00:11:58): So even though I'm sort of in that gray IT data, I'm still a data guy, right? But I learn about business and customer service and the importance of clearly communicating to customers to treating them well, to listening, like listening. Oh my God, is that a skill. That's a skill that's actually through so many different areas, including IT. I mean, good developers listen.
Matt Allington (00:12:22): But the difference is that a good developer that doesn't understand business can listen, and often what they will hear is a technical problem. What they don't hear in between the lines is the true business problem. This is where these people in the gray area, I think have an additional skill.
Matt Allington (00:12:42): I think this is another thing about if you've been in commerce and you've got this sort of IT bent. You learn to be very curious. So when I go and consult these days... I'm jumping forward. But when I consult these days and a customer says to me, "I need this report," and I look at this report and I think, "No, you don't. You don't need that report. You think you need that report, but that's not what you need."
Rob Collie (00:13:07): You think you need chicken number 11.
Matt Allington (00:13:09): Exactly. But have you tried-
Rob Collie (00:13:11): But have you tried chicken number seven?
Matt Allington (00:13:13): Exactly. They're a bit small. But number 12, you get 9% more for less than 9% extra.
Rob Collie (00:13:22): By the way, there's a bit of a cliffhanger in that story you told.
Matt Allington (00:13:25): What's that?
Rob Collie (00:13:25): Was there any chicken number 11 in the back?
Matt Allington (00:13:28): I think there is. And I think history will say yes.
Rob Collie (00:13:32): Okay. All right. Anyway, it wouldn't be much of a lesson if there hadn't been any in the back. Right?
Matt Allington (00:13:39): Exactly. Anyhow, I think I've got off track somewhere along the line. I mean you asked me about my story and you were asked, I mean...
Rob Collie (00:13:45): Hey, off track is what we do here. I mean, we don't spend necessarily all the time on track.
Matt Allington (00:13:52): All right. So let me jump forward. All right. So school was where I fell in love with computers. At some stage after my supermarket life, I joined Coca-Cola. I Spent half of my life there as you know, Rob. So 25 years. I started as a Coke rep and I took that customer service experience as a Coke rep.
Matt Allington (00:14:13): I'm not a sales guy. I never really was, but I was in a sales job. In addition to customer service, what I learned in that job was process. I really learned a good sense of process. It's such an important skill that someone has. Not everyone has to have it, but someone has to understand process.
Matt Allington (00:14:33): I'll give you a good little anecdote here. So is back in the day, right? We had these, we called them route books. So that was like an A4 folder, which would be like, what do you call it? A letter paper folder with cards in it. You'd walk into the customer and you'd count the stock in the stock room, write it on this sort of thick piece of card and then go out. We had this handheld device, which resembles like an old Nokia phone that you sort of used to punch in the numbers and place the orders.
Matt Allington (00:15:02): Sometimes when I stopped to take an order, I'd take one, two, three orders in succession. Always do the most friendly customer last so you can stop and have a cup of coffee before you move onto the next customer. Right? So imagine this. Get out of the car, go to customer one, least liked customer. Number two, middle liked customer. Customer three, most liked customer, have a coffee, take the orders, get back in the car, take out the book, key in the order for the last customer, forgetting that you've been to the other two customers and off you go.
Matt Allington (00:15:34): So what would happen then five days later, you're sitting in the office and you get a phone call from customer number one, least liked customer. Can we bleep on this? "Where's my bleeping order from last week, you useless pile of Coke rep?" What learned is that I had this process that failed in operation. I could've probably gone with lots of different solutions, but the solution I came, the process based solution to solve that problem because of my customer service ethic, I felt terrible that this was my fault. It was my fault.
Matt Allington (00:16:09): And the process based solution I came to was that when I had those three card stops, I just flipped the order of the cards in my book. I put the last card first, the middle card second, the first card third. So I'd get out of the card and then work backwards. I'd go to the first customer, second customer, third customer. But my cards were in the reverse order. Have my cup of coffee, get into the car, forget all about the other two customers. But when I flip the card, there was the second customer. "Oh, that's right. I got to place this order." Flip the card. "Oh, that's right."
Matt Allington (00:16:43): But to me, it is just such a simple example of how process is so important because things can and do go wrong in business. And if you design a good process that protects against the thing, that's what process is all about. It's about protecting against things that go wrong, because if you rely on a human to execute a process, they're open to failure. Right? So I really learnt about process in my 25 years at Coke and had quite a few roles where that was an important part of the job.
Rob Collie (00:17:20): I certainly have been on kind of a journey of discovery when it comes to... By the way, we've been doing a great job of translating between metric and A4 to letter and all of that. And for those of us here in the states, when Matt says process, he's talking about process. That's what we mean. Similar word, but I just want to translate that for everybody. I had a friend that I worked with at Microsoft who was from Canada and he was always making fun of Americans for saying process. He's like, "I just hate how it sounds process, process. It's process. It sounds so much better." I kind of agree with him. Process does sound better. Anyway, the value of process.
Matt Allington (00:18:00): But I think these things go together. I think data and process go together. I think you got a bit of both. I don't think you can have all data and no process. This would be my experience. I could be wrong. Love to meet the person who's into data that doesn't have any process. I mean, just think about a developer, right? A developer is a process.
Rob Collie (00:18:19): I was just going to say-
Matt Allington (00:18:20): What are you talking about? You've got no process?
Rob Collie (00:18:22): Okay. So here's... I'm going to be-
Matt Allington (00:18:23): Oh, it's process?
Rob Collie (00:18:24): No, no. I'm adopting your word, processes is better. We've decided. We're going to go with process from now on. I'm going to beat up on myself probably too hard, a little too unfairly, but this is something really interesting about our business is that one of the challenges from the very beginning that we knew that was going to be a problem. I knew it was going to be a problem. That doesn't mean that I knew that I had to solve it. It's a little different.
Rob Collie (00:18:48): The problem was going to be that compared to the traditional BI business model, we're burning through projects at a much faster pace. And it's been our mission to grow. Our mission is to scale. We want to be a big organization. We don't want to just be me, which is great because everyone that we've hired is better than I am. And that's awesome.
Rob Collie (00:19:07): So in order to survive and be profitable as a professional services firm, you need to keep business coming in and you need to keep utilization relatively high when the blocks of work, because you're committed to finishing them much, much more quickly than the traditional model. It leads to what Kellan, our president calls the Tetris problem. How do you take all these little smaller pieces of work? And by the way, smaller, I want to be very clear, smaller doesn't mean less valuable. I truly believe that in a shorter period of time, people with our methodology, and Matt, you're the same way, right? Your operation I'm sure moves at the same pace that ours does.
Rob Collie (00:19:47): And this is why you're in this business instead of the prior one, which we're going to get to. But when you scale this, when you turn it into like a 20-person team even, wow, does it get hard. In the last couple of years, I have stopped saying a particular sentence that I used to say all the time.
Rob Collie (00:20:08): I used to say when people ask me, "What do you do?" I'd say something like I run a data consulting business. It turns out I wasn't very good at running it. In some very real sense, it was my idea. I was committed to this idea for a long time and I'm on the path that I saw 10 years ago. So I give myself credit for that. At the same time though, oh my gosh, if we look now at how our business runs internally, the way that we operate is almost impossible.
Rob Collie (00:20:43): The amount of internal process and data, like internal software, internal workflows that are both human, but also in large degrees automated, I actually consider that now to be a form of intellectual property of our company, that we have figured out how to operate at this pace and at scale.
Rob Collie (00:21:06): And it turns out that I'm just really not the person to solve that kind of problem. I thought that just because I knew that it needed to be solved, that's all I needed. And Kellan, who we absolutely need to drag, kicking and screaming onto this podcast someday, has done an amazing job developing that internal IP, that internal... Let's use the cliche, the internal digital nervous system. But it's a mixture of digital and human that keeps us humming.
Rob Collie (00:21:36): It keeps us profitable. It keeps us moving. It keeps us able to grow. And that is as much of a challenge as building like a piece of software and a complicated piece of software at that.
Matt Allington (00:21:46): You're not a process guy, but you're a data guy. I think that's it, right?
Rob Collie (00:21:49): At the same time, I have developed an incredible respect for process over the last couple of years in particular to the point where I actually think of it as a strategic competitive advantage for our company. It's just that I'm not good at it.
Matt Allington (00:22:07): Yeah. You can actually take a patent on a process. Did you know that? Ooh. Yeah, you can take out a patent.
Rob Collie (00:22:17): When I was at Microsoft, I got paid to patent basically anything. Use these three words together in a sequence. We should file a patent on that. In fact, this conversation is probably-
Matt Allington (00:22:26): Patentable.
Rob Collie (00:22:27): ... running a foul of 17 Microsoft patents because right now, we're not allowed to do this.
Matt Allington (00:22:32): I've got a bit of process in me but I spent my life in business when I really wasn't... No, it's not true. I am a business guy. I'm just not a sales guy. Now, let's talk about sales. Now, Rob, you said that you're growing your business and you have aspirations to continue to grow, and I'm sure you will, and be successful. But the person, the best person to sell that fantastic patentable product consulting approach that gives you hundred times more value than the price you pay in nine days, the person to sell that is not the person that's going to deliver it under most circumstances.
Matt Allington (00:23:16): Maybe there's some special people out there. There's definitely some special people out there. But most of the time, the person... You know that person, the person that they love the hunt, right? So they like knocking on the doors. And that first no is just a motivation to get up the next morning and go and knock on another five doors. And then they get the sniff, right? "Oh, that I think I've got a chance here." And then they... That's not me.
Matt Allington (00:23:42): I lived in the sales world, but the funny thing is we're very adaptable. Humans are very adaptable, right? So if you've got a screwdriver and you're facing a nail, you can actually hit that nail into the wood with a screwdriver, right? It actually can be done. Well, that was me, right? So I'm in a sales job and the truth is, I'm a data guy, right? But I'm in a sales job. But my goodness, I had the best data on my customers from anyone in that sales team. Right?
Matt Allington (00:24:15): I could log into the IBM AS/400 and I taught myself IBM Query, I think it was back in those days. And I could actually extract the data that I needed for my customers. Whereas everyone else is getting these six inches, thick pieces of paper. You know the blue and white line paper with the holes down the side?
Rob Collie (00:24:39): Yup.
Matt Allington (00:24:39): And someone used to come around and drop them on your desk everywhere. That was data for every other sales person.
Rob Collie (00:24:45): It was green here in the states. We call it green bar paper here, but that's okay. It's just the color inversion in the hemisphere. It's no big deal.
Matt Allington (00:24:54): Yeah. So we all use that competitive advantage, right? It doesn't matter what line of business you're in. We all use the skills that we have to get the job done. So I could go into a customer meeting and talk data. I mean, I could talk. I can talk to a customer, and that customer service ethic that I talked about before, I'm more than capable of doing that. I can use data to my advantage. But I wasn't that sales guy as such. So I could use data to help me succeed, but I was wasn't the sales guy.
Rob Collie (00:25:29): Either I wasn't aware or I didn't remember that you'd started in a sales role at Coca-Cola. It is kind of hard to imagine you in a sales role. It isn't your personality.
Matt Allington (00:25:39): I'll second that.
Rob Collie (00:25:40): I want to say one other thing was really interesting, which is that you mentioned that sales personality, the one that's hungry, enjoys the hunt and all of that, they don't really want to work for us either because there's no big kill ever. You're not selling the million dollar project with us. We might end up doing a million dollars of business with a particular company over time, but if so, it's a lot of projects. And the sales person isn't really important after the first sell, you know?
Matt Allington (00:26:12): Yeah.
Rob Collie (00:26:12): So we're forced to hybridize in some ways that you wouldn't necessarily expect, but as soon as you said that, I was like, "Oh, that's one of our issues is that we're not selling at Ferrari price points."
Matt Allington (00:26:25): Yeah. But I think those sales people are on that same continuum that I talked about before, right? So there's a sales person that's closer to the center that's got a technology bent who is good at doing those engagements at smaller levels that the big million dollar project is not the most important thing to them, but getting a successful project of a smaller amount and moving on is fine.
Matt Allington (00:26:50): So you just got to find the right person for the right fit. So I was the best data guy on the block. So you may have imagined this person. So we're in an office and we had various little remote offices. Maybe there was 10 or 15 people in the team that I was in and I was the PC guy.
Matt Allington (00:27:10): So everyone sitting in their little cubicle. Something is wrong with Excel. "Matt, come over here. Help me solve this problem." At 20% of my day on any... Maybe this is why I didn't sell so much, I don't know. At least 20% of my day was solving other people's PC data, Excel. In fact, anything. Anything technology related. That's where I spent most of my time. Frankly, that's where I got most of my enjoyment.
Matt Allington (00:27:41): I think every company has got that person. Right? So that person who is more approachable than the average IT support person with the IT support processes. So imagine if you had this Excel problem and you wanted to get it fixed, so what are you going to do? You're going to log a ticket with support and wait for them to contact you. No, you're not going to do that. That's not how problems get solved in the business world. It's, "Hey, Matt. Come here and help me fix this problem." Four minutes later, the problem is fixed and we move on.
Rob Collie (00:28:18): Every team, it's not just every business. Every work group eventually develops this person. It's almost like they just coalesce into existence out of the cube farm. Because if you don't have one, you're just going to keep... Without even realizing you're going to keep hiring people until you get somebody that's willing to put that hat on. And then once they do, you never let them leave.
Matt Allington (00:28:43): And if you think about it, the tool of choice for that person since 1980 has been Excel. It is the tool of choice because you can solve any problem. If you can't solve it with a formula, you can solve it with VBA. You want a mail merge thing? Let's go to Excel. I'll show you how to do mail merge. You want to automate the sending of an email? The same time, I can write that in VBA for you. Sure.
Matt Allington (00:29:13): You don't need to come and spend $20,000 for that solution. We can do it in Excel. I'm not sure if Excel... Well, you talked about or implied the concept of evolution. I actually think Excel evolved with those people. They're mutually dependent on each other. If those people didn't exist, I don't think Excel would exist the way it is today. If it wasn't for those people contacting people like you, Rob at Microsoft saying, "We need this, we need that." It wouldn't be great if we can do this. Those people wanted those features and they weren't available previously.
Rob Collie (00:29:49): Well, they would contact us and say, "We really need a chicken number 11 added to Excel." And we'd say, "No, no chicken number two. And you're going to love it."
Matt Allington (00:29:58): Yeah, exactly.
Thomas LaRock (00:29:59): So what you've just though in saying since 1980, the tool is Excel. So does that make Excel the screwdriver or the hammer and what's the nail?
Matt Allington (00:30:10): It's the Swiss army knife.
Thomas LaRock (00:30:12): Is it?
Matt Allington (00:30:12): It's definitely the Swiss army knife. If it was a screwdriver, it'd be Access. Right?
Thomas LaRock (00:30:21): Do you ever come across a scenario, a situation where Excel really wasn't the answer and somebody's trying to use it in a way that they just shouldn't be using it?
Matt Allington (00:30:32): Oh, most of the times. Most of the times I would have to say, to be honest. I'm sure I'm preaching to the converted here for this audience, but I always say the best thing about Excel is its flexibility. And the worst thing about Excel is its flexibility. It's so true, because it's a double edged sword. You can do anything you want and that's not always the best thing. I think that's the difference.
Matt Allington (00:31:00): Everyone here loves Power BI, right? So we love Power BI. But to me, this is one of the big differences between Power BI and Excel is the structure, the rigor. The fact that if you have an error in an Excel spreadsheet and you try and load it in Power Query, into Power BI, it fails for goodness sake. What? I've put up with that error in that spreadsheet for years. Now, you're telling me I have to fix it?
Rob Collie (00:31:30): Yeah. It's so sensitive that if your Power Query is trying to delete a column that isn't there anymore, it also fails.
Matt Allington (00:31:37): Exactly.
Rob Collie (00:31:37): It's like, "No, no, wait. We're going to end up in the same place either way. The column is not going to be there." It gets very sensitive.
Matt Allington (00:31:44): Yeah. So I think that's the step change with the Power BI ecosystem, Power BI, Power Query is it's got the accessibility for the Excel world. Rob, this is how we first met, right? Because you'd recent... Not recently maybe. How long have you've been at Microsoft? Maybe five years or something like that?
Rob Collie (00:32:08): I was officially done at Microsoft in early 2010. But really I'd been out since like the fall of '09.
Matt Allington (00:32:15): Mentally, at least?
Rob Collie (00:32:16): Well, I mean I was in Cleveland. We didn't do remote work back then. That wasn't a thing. But I started the blog in the website PowerPivotPro in 2009. So the first post. But I didn't really get serious and start writing decks on the blog or anything until probably January 2010.
Matt Allington (00:32:34): Yeah, right. So it's targeted the Excel audience. So you came from an Excel program manager background and it was the Excel people, presumably with some of the analysis services, people that were brought together to come up with this new thing. But it was supposed to be accessible to the Excel person, right? The person who's grown up. I mean, one of the things that people ask me, not so much now, but certainly in the Power Pivot days of training was, "Can you give me a list of, this is what I do in Excel. This is how I do it in Power BI?" Because this is their mindset, right?
Matt Allington (00:33:10): So how do I do some ifs in Power Pivot? That was a typical question. Now, of course there's a whole world of retraining around the differences and perhaps that's a bigger conversation in its own right. But the marvel of this product is it was targeted at the Excel user primarily, initially. Not uniquely because it's also targeted at the professional end as well.
Matt Allington (00:33:37): But it was targeted at that group while bringing in some rigor and structure and trying to get away from the robustness or the lack of robustness of Excel. So a classic example would be, let's say you've got a 10,000 row spreadsheet in Excel and let's just for simplicity's sake, let's say you've got cost price and sell price in your column. You got 10,000 transactions and you want to work out the margin. How much profit did we make?
Matt Allington (00:34:04): So in Excel, of course, what we would do is we'd go into the next cell and go equals the sale price column subtract the cost price column. And we would copy the formula 10,000 times down the page, hopefully using some control, double click in the bottom right hand corner to extend the range.
Matt Allington (00:34:20): But the thing is we've duplicated, we've replicated that formula 10,000 times. That is a window of potential for failure, because at some point in time, when someone's down in row 9,723 and they accidentally hit the delete key, no one knows that that's happened, right? So that's the flexibility of Excel and the lack of robustness, which leaves the door open for an issue.
Matt Allington (00:34:48): So of course in Power Query, you have to load the data. We encourage you not to replicate that column, and it's built on the fly with a single formula that works on every single row. And if there's an error in the source data, the error will flow through to the report and someone has to fix it.
Matt Allington (00:35:10): We've always had this rigor in the SQL server world, right? One of the early things that I did when I first started out, so this is six and a half years ago. I started with a customer who had data, but no knowledge. And they wanted me to get their data and turn it into knowledge and insights and actionable reports, and those sorts of things. They actually didn't tell me that that's what they needed. They just said they needed help. So I told them what they needed.
Matt Allington (00:35:41): Anyway, the data was in an inaccessible way. So I loaded it up into access. Go to tool, for me. And I think within the first half an hour, I hit the... Do you know there's a 2GB limit in Access? Goodness. So I hit that limit. I thought, "Okay. This is not going to work." So then I installed SQL Server Express and then eventually went to SQL Server. And I had to teach myself these tools.
Matt Allington (00:36:03): Do you know you can't just cut and paste into SQL server? It's freaking incredible. You actually have to use this tool called SQL Server Integration Services. And do you know how freaking hard it is to use that software? It trips you up at every single step. If something goes wrong, when there's an error in my Excel formula, it won't freaking load. It's unbelievable.
Rob Collie (00:36:22): That explains why I never learned it.
Matt Allington (00:36:25): Oh, it's so frustrating as an Excel guy learning SQL Server Integration Services. I mean, SQL was pretty good. I got to say learning SQL was pretty good. Easy language, scripting language to learn. I got the concepts of tables. I'd done some IBM query programming as I mentioned before. So I understood the concept of writing these statements. But SSIS, oh my goodness, how freaking hard is that piece of software. Because it's so disciplined in what it does.
Matt Allington (00:36:54): So this is the difference between Excel and Power BI and Power Query is I don't think the traditional Excel folk understand how free they've had it, how flapping in the wind they've had it all these years.
Rob Collie (00:37:13): This itself is you've really cut yourself and thrown yourself into a shark pool here. Because this is my favorite, my absolute favorite topics. Although, I don't really bite. It's a pretty friendly shark pool. Structure versus flexibility. Just like we were talking about earlier with the IT versus business, there's of course a spectrum there. The story you just told, I think is absolutely true that Power BI brings structure to the Excel world. But at the same time though, Power BI is less structured than the previous version of analysis services.
Matt Allington (00:37:48): Absolutely.
Rob Collie (00:37:49): The previous version of analysis services was entirely too structured. And that was a mistake. So the way I like to explain it to people is that like the old BI tools and the previous analysis services, the multidimensional analysis services, they all had this same sort of three-part hubris about the world.
Rob Collie (00:38:09): First of all, this is a software engineering mistake. It's the type of mistake that a software and only really software engineers would make. Number one, the assumption is the implicit assumption of all these people working on these products has always been that the real world can be reduced to a series of academic mathematical concepts and constructs. You can translate it into symbolic notation, right? You extract the essence, find the bones of it.
Rob Collie (00:38:35): Okay. So you can translate it into some formal concepts. That's assumption number one. Number two is that you can actually get there. You can figure out what those concepts are. You can actually communicate with people to the point where we can make that translation. And then number three, once you've done that, that's the way things are. It's never going to change. Nothing's ever going to change about the business.
Rob Collie (00:38:54): So all three of those things are wrong, every single one of them. And so the formalized concepts of previous analysis services right off the bat, like is this a measure group or a dimension? What are your hierarchies? And these things became such hard strong constructs that two problems would happen. It actually had two bad, bad impacts. Number one was that it didn't fit the real world all that well.
Rob Collie (00:39:20): It was hard to translate. It was almost like you wanted to change your business to fit the software. It was that bad. And change, oh my God, you need to change something. If something about your business changed, you might as well just start over in some cases. But those same formalized concepts also made it a lot less approachable. You couldn't even get started.
Rob Collie (00:39:41): Every time I'd get started trying to learn MDX, I'd end up asking someone how to write an if. And 20 minutes later, we're off in hierarchy hell and this and that. There was no hello world. There was no easy progression. Whereas something like DAX and Power BI has a lot more sort of humble structures to it. We have tables. We have columns. We have measures. We have formulas.
Rob Collie (00:40:11): We have these lines, you draw between tables. We have relationships. They don't have as much overloaded functionalities as you might expect from SQL. So it's a lot more modest about what it makes you translate into. And that allows it to be more approachable, but it also allows it to be a hell of a lot more flexible. So it's like this Goldilocks sweet spot of structure versus flexibility that is really unique.
Rob Collie (00:40:35): I wasn't part of really anything to do with engine when I worked at Microsoft on Power BI and I certainly had nothing to do with the original analysis services either. But I did get to sit right next to them, mirroring Christian and Marius. Actually, I watched them essentially retrace their steps.
Rob Collie (00:40:54): I had this almost historian viewpoint sitting in these meetings just being so fascinated with watching them sort of... And they just nailed it. It's beautiful. I mean, it is really something to admire what that crew pulled off. It's not often that you get two shots at the same problem over a 20-year timeframe.
Rob Collie (00:41:14): Oftentimes, when you get that second shot, you oversteer, you overcorrect. You miss by the other direction. I'm a software cynic. I'm always beaten up software, but I'm truly in awe of what has been built there.
Matt Allington (00:41:30): Yeah. But I think this is an evolutionary thing as well. We talked about evolution before. So the way things evolve together. I was reading or watching something at some stage about the way that the domestic dog has evolved with humans. So my dog, if I'm looking at my dog and I point something on the floor, my dog looks at where I point.
Matt Allington (00:41:55): Now, if you do that with a wild wolf, the wild wolf does not look where a human points. Dogs and humans have evolved together, or probably in this case, more correctly, dogs have evolved with humans to have this gene of understanding, which has taken this wild canine and turned it into something that has evolved with a human.
Matt Allington (00:42:16): Now, I think software and people evolve together as well. I mean, in many ways you built this analysis services version one and it was complex. But I think most people would agree that we wouldn't have got tabular and decks if it wasn't for the learnings and the shortcomings of the previous version, right?
Matt Allington (00:42:37): So we take what we've learned. You said, Rob the opportunity to do it again from scratch. I mean, I actually talk about that in my training. I said just imagine someone comes and knocks on your door and says, "Hey, Excel guys. Hey, analysis service guys, how'd you like a chance to do it all again from scratch? Will take out all the dirty laundry and we'll just do it better this next."
Matt Allington (00:42:59): I mean, what a fricking once in a lifetime opportunity. And the other thing is that technology is evolving as opposed to software, which is what I was talking about then. So let's look at Apple for a second. You remember the Apple Newton, right? So the concept of palm based computers is not brand new.
Matt Allington (00:43:17): In fact, I had a PalmPilot when it first came out in the same sort of thing that I was talking about before. But the technology has evolved. So the iPhone of course was groundbreaking, but one of the key differences between the iPhone and the previous software that came before that was the hardware had this glass screen that you didn't actually have to push on it with a stylus in order to make input. That technology increase actually was an enabler provided you had the right entrepreneurial vision and right software design.
Matt Allington (00:43:53): I mean, it's an alignment of the planets. I think when we talk about tabular and analysis services tabular and decks and Power BI, it's the new technology that was available as a result of improvements in memory and not so much solid state disk, I guess, but processor speed. The fact that we could come up with these column store database and make them practical for business. I mean, I don't think we would've had that window of opportunity to reinvent SSIS unless that technology was there to support it.
Rob Collie (00:44:27): Yeah. Amir, when he was in graduate school apparently wrote two separate thesis, papers and one of them was on column store databases.
Matt Allington (00:44:39): This would've been years ago, right? Sorry to interrupt.
Rob Collie (00:44:41): Oh, yeah. I mean, I don't know. This was probably in the '80s, right? So he's just been watching for a long time, waiting for the RAM prices to fall into a range where realistic business volumes could fit into main memory. It was like they just passed a certain critical threshold one day and off he went. But he also then had him and the team, they had the experience of what they'd learned about their... There was really no reason why MDX, for instance, couldn't have been brought forward. MDX could have been brought forward and used in the column store, VERTEPAC world.
Matt Allington (00:45:17): Thank goodness that didn't happen.
Rob Collie (00:45:18): I agree. I wouldn't exist. I just wouldn't be here. I'd still be trying to write that first if.
Matt Allington (00:45:28): Yeah. So evolution of software, it's an interesting thing as is evolution of people, right? So I mean, ultimately we've been digressing, but you're asking me about my journey. So at some point in time, I jumped over to the IT side. So I spent eight years in supermarkets and 15 years in sales at Coke, sales and customer service at 23 years training about business. Then I jumped into IT and spent 10 years in IT.
Matt Allington (00:45:59): So that's when we met of course, sort of towards the end of my time at Coke. But for 23 years, I thought I'm an IT guy in the business world. So on the 23rd year, I jumped and I went on the other side of that line and I became in the IT world at least thinking I was an IT guy. But you know what, the grass wasn't greener.
Matt Allington (00:46:24): The grass was different. It was a little bit more manicured and had a little bit more resource against it. But the grass wasn't greener when I jumped the fence. I was just a business guy in the IT world then. There's just different problems, different responsibilities. I mean, I certainly learned an appreciation for IT in those 10 years. Goodness. Did I learn to appreciate IT?
Matt Allington (00:46:51): I spent 23 years hating IT as a business guy. They're the people that took months to respond to your support ticket that the systems always break down, blah, blah, blah. So I certainly got to see the other side of that, right? So you've got a level of responsibility of up time and you need to be able to support things and you can't just let the business guy do the Excel spreadsheet.
Matt Allington (00:47:16): You have to have documentation and processes and recovery processes and redundancy. It's no wonder it frigging costs so much money, right? What you have to put against this stuff, the IT departments are responsible for. But this comes back to my very earlier comment about the fact that you've got this crossover point and then you have this gray area. So the business analyst on one side and the IT style people in the business on the other side.
Matt Allington (00:47:43): The truth is there are a lot of us out there. Now, I'm not saying 50% of the population. Not even 1%. But maybe 0.1%. There's 0.1% of the people that straddle that line. They live in that gray space. They're either business analysts working in IT or they're the in-house Excel type person solving the problems that IT can't or won't, or shouldn't solve.
Matt Allington (00:48:13): So the truth is I'm one of those guys, and I suspect that you are probably one of those guys as well, Rob. I don't really know about Tom. But there's a lot of us out there and we have an important role to play because if you cut that gray piece out of the continuum, there's actually a bit of a grand chasm that a lot of stuff is going to fall through as you try and hand stuff over. I'm not saying that we are the most important people in the world, I'm just saying we're an important part of the-
Rob Collie (00:48:41): Yes, I am.
Matt Allington (00:48:42): No, we're just an important part of the overall process. Right? I think this is where Power BI comes in because Power BI and Power Pivot when I discovered it through you, Rob. I don't even think I discovered Power Query. When I decided to do this on my own, which was, it was early 2014, January or February 2014, and I remember having a chat with you, Rob. I was bouncing it off you and asking what your journey was like. I basically told you I was going to steal all your ideas and just fricking redo it because no one over here has heard of Rob Collie and I reckon I could be the Rob Collie of Australia.
Matt Allington (00:49:25): So I made that plunge at that point in time, but I'd never heard of Power Query. So my business was going to be Power Pivot, InfoPath, and SharePoint. That was my business, right? Customizing stuff, using those three groundbreaking technologies. Now, as it turns out, one of them sort of survived until it didn't, which is Power Pivot. Well, don't start me on SharePoint or some of its newer cousins, but Power Query then came.
Matt Allington (00:49:59): Well, it was already there, the truth is, but most people we haven't discovered it because the branding is pretty poor. So then came Power Query and then... I'm not sure what the history is. You probably know better than me, Rob of why we stopped or why Microsoft stopped on Power Pivot and moved to Power BI. But timing is everything.
Rob Collie (00:50:18): Let me jump in there for a moment. I can tell you that even at the time that we were doing it, we were doing Power Pivot, the Excel team wasn't terribly happy about it. I had left the Excel team at that point and then been recruited. I'd spent a year in purgatory and decided I hated that. About that time, Amir came calling and said, "Hey, I got this thing going on." I'm like, "Ooh." But these are my friends over on the Excel team. Like actually my friends at that point, not just colleagues.
Rob Collie (00:50:45): They were pretty grumpy about what we were doing because there was almost nothing they could do about it because they had an open... Anyone can write an add in so they couldn't stop us. They really just couldn't stop us. They'd have to like go to Bill to stop us. They had to begrudgingly go along with it. They didn't like any of the things we were doing.
Rob Collie (00:51:05): So as soon the SQL team who we think of as the Power BI team today, as soon as they realized that they needed to decouple from the office versions that are deployed in an organization, because upgrading office is such a huge, huge, huge thing.
Matt Allington (00:51:25): Let alone going from 32 bit to 64 bit.
Rob Collie (00:51:28): Yeah. As soon as they realized that they needed to decouple from office and that they needed a more modern graphical canvas in which they could innovate a lot faster than what Excel could.
Matt Allington (00:51:40): So Power View is not going to do it?
Rob Collie (00:51:42): No. It was a... They just-
Matt Allington (00:51:45): Silverlight? Silverlight is not going to do it?
Rob Collie (00:51:46): No, no. It turned out none of those things actually panned out very well. Power View was the one thing that actually United the two clans. In the end, everyone at Microsoft hated Power View. The SQL team hated it. The Excel team hated it. It was universally despised. They try not to even talk about it. It's like, "We don't remember that cousin that we had."
Rob Collie (00:52:07): So as soon as the SQL team took their eyes off of Power Pivot, it was over at that point because the Excel team never wanted it. In my opinion, that's a perfect example of how a big company can go wrong in a place where a small company wouldn't. Think about it. There are tens of millions worldwide of people who sling Excel in a BI capacity. The Vlookup and pivot crowd that I talk about, these people, like you say, percentage wise is not many. But in absolute numbers, it's enormous.
Rob Collie (00:52:41): There's one or two on every airplane you ever get on. Microsoft owns those people. They own the people who run the world's data. Full stop. They have full capture of the audience that matters. And yet the other half of the company has to go the long way around to try to market to those same people because the office team has different incentives and different goals, and different competitors and everything. It's like if Excel and Power BI were a company, can you imagine how dominant that combined company would be? I mean, nothing could stand before it, nothing. But because they don't meet until Satya, it's like, whoa, whoa, I don't know. What are we going to do? We can't figure this out.
Matt Allington (00:53:34): I think you take PowerPoint with it as well because you've got to be able to cut and paste those reports into PowerPoint. Right?
Rob Collie (00:53:39): Of course. We're even getting signals from the market now that a lot of places, Power BI is just being used as a slide presentation and it's just mind boggling, right? It's too far the other way. You've been talking about tweeners, these sort of like hybrids a lot. Even in the young history of this podcast, this has come up a lot. And you asked if Tom was a hybrid. Absolutely. We wouldn't even be able to communicate. Tom and I wouldn't have been friends all these years. If you know, we weren't like the reach across the aisle type of crowd. Tom, back when you were part of that organization and you were advocating for business analytics being on the conference menu, you took a lot of grief for that. That was a lot of backlash maybe in terms of intensity, maybe not in terms number of people, but that wasn't an easy road to spearhead that conference.
Thomas LaRock (00:54:29): No, there was a lot of friction. What was unfortunate when I look back at it is that it was clearly where the industry itself was heading. It was just very clear. Silos were being disassembled. They were being broken down. There was a lot of cross-pollinization happening and you found yourself increasingly being brought into a lot of different conversations and for good reason.
Thomas LaRock (00:54:59): Now, my experience is mostly the Microsoft ecosystem, but the fact that the letters SQL made their way into no less than 72 different products essentially meant, "Hey, well, this is SQL analysis services. That's SQL. That's database. You're the DBA. So you should just be able to help me with this. It's like, what? SSIS. So all of a sudden, guess what, if you were doing SSIS, you know what you are, you are what we now call data engineer.
Thomas LaRock (00:55:31): So there was a lot of this cross-pollinization that was happening and we saw that, we recognized it and we wanted to serve our members. Also, I was in charge of marketing at the time, and it became one of my responsibilities, one of my tasks to update our mission statement, which was to serve the Microsoft Data Platform Community essentially. It was no longer just a focus on SQL Server, it was on the entire platform. And that was step one.
Thomas LaRock (00:56:04): Step two was as part of that, we wanted to have this event and we were trying to attract a new audience. We could have done a couple of different things. We could have just had the event as part of our main flagship summit. We could have tried smaller events, but we said, "You know what? We want to do something a mid-size tier." The first one we tried, we put in Chicago, which we thought was centrally located. Chicago and Illinois, and Wisconsin is kind of a bed of analytics. You've got a lot of people in Cleveland, right? A lot of-
Rob Collie (00:56:34): Just up to our eyeballs, yeah.
Thomas LaRock (00:56:36): Yeah. So we thought that was a decent location to have the events, but for some of our membership, and these are discussions that we should have that were necessary, right? Hey, we're devoting resources to this new thing. Is that really what we should be doing? You know what those are fair conversations to have, but some of the venom that came out as a result, and what you're doing is just flat out wrong. You don't know what you're doing. You're looking on going, "I am not taking anything away from what you already have. I'm simply trying to add a new seat to the table for a new audience."
Thomas LaRock (00:57:19): Because having those conversations is going to make everything better for all of us. And yet, as you said, Rob, maybe it wasn't a large number. Maybe it was just the volume of a few, but that was frustrating because you would think those people would have, I don't know, enough judgment, enough common sense to understand that we're all in the same boat. And it'd be nice if we were all rowing in the same direction. They're sitting there drilling a hole in the bottom of the boat and you're like, "Hey, what are you doing? Hey, mind off. I'm only drilling underneath my own seat." You're like, "But we're all in this boat."
Rob Collie (00:58:03): I even started to feel a little guilty. Not for you. Right? I started to feel a little guilty. I was helping to an extent in my own small way. I was coming to these conferences and speaking and offering my advice and all that. There was this vocal minority who was so pissed off about it like they were being injured. They were so dedicated to it that and I'm like, "Maybe there's something here." Maybe they have a point. Right?
Rob Collie (00:58:26): So I made a point on Twitter of reaching out and engaging with one of these fine people to find out if I... And within 30 seconds, I'm like, "Oh, no. No, no. This is a bad person. This is just a bad person." Tom is on the right track. I never felt guilty again. I rededicated myself to the cause 2X.
Thomas LaRock (00:58:52): I look back and I always feel we could have done things slightly different, but at the end of the day, the direction was definitely there. What we called, it was business analytics, which I didn't really like that name. I always thought-
Rob Collie (00:59:06): But it was BACon.
Thomas LaRock (00:59:07): Yeah, it was the BACon. It was the BACon. I actually just wanted to call it the data conference and we would've had data con. And I wanted it open to everything. I sat there going Power BI, I don't think existed yet, but I was like, Tableau should be here. It's all about the data and we should have built the data conference.
Rob Collie (00:59:27): You're just not always right, Tom. It turns out BACon was great. I'm messing with you, obviously. But I love that conference. I think it's my favorite. And my second favorite is probably the financial conferences that I go to. It certainly isn't the pure tech conferences aren't as interesting. But not even as good in my opinion for drumming up business for meeting potential clients that you can help. You want the tweeners. You want these hybrids. That's what you want. You want the people who are able to see the value of both when they're combined and that's where we can add the most value.
Matt Allington (01:00:02): I'm pretty sure the first one I went to the first BACon conference was 2015. So I started on my own in April 2014. I have this recollection that the conference was like within weeks of me starting. So I didn't go to that first one, but I went to the second one. I can't remember. Was there one in 2016 as well, Tom? I can't remember.
Thomas LaRock (01:00:24): I'm going to say yes.
Matt Allington (01:00:26): Yeah. [crosstalk 01:00:27] I reckon I went to two and then Microsoft came out with the Data Insights Summit. And then of course, Power BI was starting to happen. The numbers were not the same as summit, obviously. I mean, you're trying to build something. You had a vision there and then BACon was no more. You know what, I've never been back to a PASS event since BACon. I mean, I would've gone to every BACon conference that you had, but I've never been to summit.
Matt Allington (01:00:55): I don't know. Maybe summit's got everything that BACon had. You know what? I don't know, but I've never been there. I'm actually your target audience... Not your target audience. I'm actually the target audience, but I've never been to a past summit. But I loved BACon.
Matt Allington (01:01:09): Then I went to the Data Insights Summit, and I loved it. The first year I couldn't get there. Remember, they had a thousand tickets and they sold out. I couldn't for love or money buy a ticket to that conference. Anyway, so I missed the first year. Then I went to the second year and I just absolutely loved it. Then Microsoft, in their wisdom, they combined the data summit with Dynamics.
Matt Allington (01:01:33): Let's put Microsoft Dynamics. And instead of it being a data conference, it can be a business conference. As far as I'm concerned, they completely marked it up. Someone from Microsoft might be listening, but I think they should be focused... I think there's a place for a data thing. We're the gray people, right? We don't care about ERP and CRM.
Matt Allington (01:01:56): Okay. I'm sure Microsoft would love some of the success of Power BI conferences to rub off on their aspirations to have become an SAP competitor. I'm sure they would like that, but let us have our conference. Give us a data conference. I'm happy if it's going to be a... I mean, if you're going to have a data conference, why not be the Power BI data conference? I think it's nice to have it sort of product agnostic and include Tableau. But I think the writings on the wall, Rob, you talked about that Excel Power BI company before.
Matt Allington (01:02:28): I mean, the truth is it's obviously Microsoft. I think the writing is on the wall. I think you're pretty ballsy if you're going to bet against that product. If you're a CIO and you're going to say no, we're going to be a Tableau. We're going to be a micro strategy shop because they've got mobile. You're pretty courageous these days, I think. So where's our conference? Microsoft, give us back a data conference.
Rob Collie (01:02:54): Yeah. We should come up with an acronym that ends up spelling MATT.
Matt Allington (01:02:57): Yeah.
Rob Collie (01:02:58): We call it at the MATT conference. It's like what are we really doing for Matt. I would go to that conference.
Matt Allington (01:03:05): Would you speak at MATT?
Rob Collie (01:03:06): It sounds like a great conference. Oh, I would totally speak at MATT. Yeah. I do a pre-con at MATT.
Matt Allington (01:03:12): Yeah. So, yeah. Interesting. It's very interesting.
Rob Collie (01:03:15): I'm going to share a little bit of a theory with you, Matt, that we could just as easily talk about offline, but hey, why not talk about it in front of the public? So I think I'm finally starting to see the grand plan. I think I'm finally starting to understand the Business Applications Summit. At first, my reaction was the same as yours, like, "Oh, come on. Just a silly rebrand and a combination of conferences." But I think that the story is starting to become a little bit more clear.
Rob Collie (01:03:42): So I was thinking about the other day, and for the first time ever, I realized that BI is actually a form of middleware. It's something that spans multiple systems. You can almost never get any reasonable insight just out of a single system. Even at our company to get an accurate cash flow forecast is like a five, six system problem.
Rob Collie (01:04:03): And they don't all come from the same vendor either. You know? So it was kind of a weird shift for me to think of BI as middleware. But then there's like transactional middleware, whether it's a customer transaction or not. You make a change. Multiple systems might need to be notified of the same change. If an employee is hired, they need to be added to multiple systems probably. Right?
Rob Collie (01:04:25): So you end up with this transactional middleware. So I don't have to go do five separate things. I can do one thing, and it kicks off all of the workflows to touch five different systems and things like that. Salesforce has been describing themselves for a while. At least they were... I don't know. Some number of years ago was like the operating system for your business.
Rob Collie (01:04:44): I think that it's no accident that Dynamics and Power BI, and all of the power platform, middleware stuff like the power apps and power automate and all that kind of stuff, I don't think it's any accident that they all now report to James Phillips. And then Salesforce's acquisition of Tableau starts to make more sense as well. This is sort of like an existential grand battle of the heavens that's taking shape. I do believe as you were hinting at, you do not bet against Microsoft in this sort of combat.
Rob Collie (01:05:21): This is what Microsoft does. No one does this like Microsoft. It's game over. It's lights out. It's just a matter of time. So the SAPs of the world... Do you really think? Does any of us think for a minute that Microsoft plans to long term seed the enterprise, the ERP market to SAP? No. Dynamics is mid-market. Yeah, but not for long. But Microsoft does know. There's something at Microsoft, especially the Satya Microsoft gets is that not 100% of the software at your company is ever going to be Microsoft, right? There's going to be lots of systems "best of breed".
Rob Collie (01:06:05): The random hodgepodge that you happen to evolve into your history. They get that. So it's almost like the multi-silo, multi-system, nervous system. Power BI belonged to James Phillips before he was given any of this other stuff. So I think that what we're seeing is that they're positioning themselves to do the same thing for the middleware market that they've been doing for BI, right, is to slightly make it one level more agile, one level more democratized. And it's all part of the same thing is it's all about spanning systems. Now, I've put my cards on the table.
Matt Allington (01:06:46): I think we're in violent agreement, Rob. I would 110%... Can you do that? You're a data guy. 110% agree with what you just said. But here's the thing. It doesn't mean we need a conference together. I'm talking about the conference. I'm not talking about the business strategy. I 100% agree with everything you said. It's brilliant, but let us have our own conference. Otherwise with Tom, with his BACon conference like why haven't all those people gone to summit? Why didn't they? Because they need their own space. Come one. Let's have some focus. Let's have some love.
Rob Collie (01:07:22): I agree. Do I really need to eat lunch next to a Dynamics?
Matt Allington (01:07:25): Exactly. Yes, we need them. They've got to sell the product. I get that. They've got to balance the books.
Rob Collie (01:07:31): Yeah. They've got to collect the data that we ultimately have to analyze. It's got to go somewhere.
Matt Allington (01:07:35): I like anchovies. They don't like anchovies. Come on. So I mean, three and a half thousand people I think was at the last Data Insights Summit, or maybe that was the one... I think it was the last Data Insights Summit. Something like that anyway. 3000, whatever. That's plenty. Let's just focus on that. That's what I'd love to see. I'd like to see Tom's original vision of having something just for the data people and all right, it's becoming a Microsoft thing. Well, I love backing a winner. Don't get me wrong.
Matt Allington (01:08:05): I mean, there's plenty of things you can do with your money. I would prefer to put mine on a winner and go for it and go hard. But yeah, I just think we need... We're all the same. This is what we've been talking about today, right? Is that we are the same types of people. I mean, I am an introvert. I know you're an introvert, Rob. I don't know Tom that well.
Matt Allington (01:08:24): But I tell you what, when I get to those data conferences and someone comes up and says, hello, I get them. Not the Dynamics people, right? Sorry Dynamics people. But I get the data people.
Rob Collie (01:08:36): I have some great pictures. The pictures of that San Jose hotel lobby hanging out. We need that again.
Matt Allington (01:08:44): Absolutely.
Rob Collie (01:08:45): I agree. But we're not doing any of that anytime soon. You asked earlier if there was a 2016 Business Analytics Conference and I know for sure that there was because I absolutely destroyed my leg right before that. I think it was your daughter, right? Didn't she drew a sketch of my mangled leg from a photograph?
Matt Allington (01:09:06): I don't remember. I think maybe I was writing a blog or something. Oh, that's right. Because I got thrown in the deep end. I think you were supposed to be presenting at pre-con and basically two days before, I think I got a phone call saying, "Hey, can you do my pre-con?" I don't know if the rest of the world knows this Rob, I got a little secret out here. Rob's idea of planning for a pre-con is in the bar the night before. He says to himself, "What am I going to talk about tomorrow?"
Matt Allington (01:09:37): So Rob shared with me all of the preparation that he'd done for that pre-conference and then basically I spent the next couple of days trying to put together what I wanted to present and it was with [inaudible 01:09:50] right? It was an [inaudible 01:09:51] night at that point.
Rob Collie (01:09:51): Yeah, it was.
Matt Allington (01:09:51): Anyway, so you break your leg and I think I must have done a blog article. That was back in the day when everything was super green on your site, right? Everything was green. So I got my daughter to do a sketch of Rob lying on his deathbed who was foot up in plaster. And that was probably it.
Rob Collie (01:10:10): She did a bang-up job.
Matt Allington (01:10:11): If I search it, I'm sure it's there.
Rob Collie (01:10:14): I still have it. It's in our clip art folder here. I mean, I did her a favor and she didn't have to approximate any angles of where to put my foot. I did a nice 90-degree. All she had to do was just rotate the foot 90 degrees from where it's supposed to be and then draw it like that.
Matt Allington (01:10:30): Yeah. Oh, I see you from your website that you did your knee in 2013 as well. So there's a bit of a...
Rob Collie (01:10:36): Yeah. It's the same leg.
Matt Allington (01:10:38): [inaudible 01:10:38] Okay.
Rob Collie (01:10:41): That leg is permanently smaller. My right leg is now... There's nothing I can do. My right quad is now just permanently smaller than my left quad. I'm just going to be lopsided the rest of my life. No more scooters.
Matt Allington (01:10:53): Yeah. But they were the good days, those conferences.
Rob Collie (01:10:57): They were indeed. Wow, this has been great. Tom, you've been scribbling notes seemingly, furiously the whole time. Tom saves them up.
Thomas LaRock (01:11:06): I was going to ask one thing since you guys do this training, I subscribe to the idea that when it comes to training, that telling ain't training, right? You have to put your hands on something. That's to me what training really is. And I've been skewered for this as well from professional trainers who tell me otherwise. I'm like these are two different things, right? So if I deliver a lecture, that's one thing. But if I really want somebody to learn, I'm going to have them put their hands on something and use their mind and their hands together.
Thomas LaRock (01:11:39): So I wanted to ask about how you guys structure your classes. I assume that the people in the class are interacting with Excel and Power BI as you're teaching your classes. Yes?
Rob Collie (01:11:49): Absolutely, yeah.
Matt Allington (01:11:50): Absolutely.
Thomas LaRock (01:11:51): So the question I really have for you guys right now is we mentioned a hostage.
Rob Collie (01:11:58): Yeah.
Thomas LaRock (01:11:58): Right?
Rob Collie (01:11:59): Hostages and volunteers. Two different kinds of students.
Thomas LaRock (01:12:02): Since you've switched to remote, and I'm assuming, Matt, you're doing more remote training these days as well.
Matt Allington (01:12:10): Yeah.
Thomas LaRock (01:12:11): How do you deal with a hostage in a remote situation? Because when you're live, I think the way you would deal with it is a little different. So how are you doing that today?
Rob Collie (01:12:21): Matt? Do you know what we mean by hostages and volunteers?
Matt Allington (01:12:24): I know exactly the typecast. I mean, I've been there actually. Yeah, absolutely.
Rob Collie (01:12:30): Tom's right. I mean, when you're in person, you can sort recognize pretty quickly who didn't want to be there. You can try to recruit them into being interested. And then eventually if you fail with that, then you can just start focusing your attention on people who are actually absorbing it. But boy, remote is hard. You don't get those facial clues. You don't get that feedback. We've made it work for sure. It's necessity and we're still getting good results. But I do think it takes more of a toll on the trainer
Matt Allington (01:13:02): Our experience has been exactly the same. So the process that we do for remote training is that we strongly recommend two screens and a camera, a webcam. I've got two screens at the moment. So one screen has the streaming coming from the teacher and the other screen has got your software application, typically Power BI. And you're actually going through the exercises at the same time.
Matt Allington (01:13:33): We ask people to turn their cameras on. It's a little bit obtrusive, but the difference between able to see on someone's face what's happening and seeing nothing is like daylight. It is so freaking hard to train someone if you don't get any non-verbal feedback. So Jason, who's my full-time trainer, he did his first live training course in 10 months last week. He rang me after. He was so excited.
Matt Allington (01:14:02): He said, "Oh, it's just so much easier to deliver face to face training because you're there. You're experiencing the feel of the room." These hostages are a subset of those people, right? They're the people who never look up. So you ask people to just look up when you're done and they never look up and you say, "Hey, hostage number one. Are you finished?" "Oh, yeah. I'm just catching up on my email while I'm waiting for you."
Matt Allington (01:14:30): Okay. So remember we're going to look up when we're finished, right? I mean, we've made it work too, Rob. Exactly the same. I mean, I learned how to train Power BI from Rob, right? So necessarily my delivery style is we... I remember we had a chat about this Rob once before, but in my book I train people to lay out their tables using the Collie layout methodology, which is the way I learned from Rob originally.
Matt Allington (01:14:59): But it just works. I don't believe we all have to reinvent the wheel. I mean, I could have decided to put the dimension tables at the bottom and the fact tables at the top and call it the Allington layout methodology. Right? But hey, why reinvent the wheel? Rob had this fantastic approach. I can always tell someone who's been trained by you Rob, or your team or from my team.
Matt Allington (01:15:23): I say, "You get these consulting jobs. Can you help me with this? First question, show me your model. Oh, you've been to the Rob Collie Matt Allington school of learning. So fantastic. I know I can communicate with you. So fantastic." When they've got all the tables everywhere, you know we're in trouble.
Rob Collie (01:15:41): Matt, I got to tell you, I have actually never circled back and told you this. I have been now told so many times from students, right? They're like, "Oh, this is the Collie style. This is what Matt talks about in his book." This comes full circle.
Matt Allington (01:15:55): Rob, you could be famous.
Rob Collie (01:15:56): Yeah. Tell you what, Matt, you and I just keep [crosstalk 01:16:00].
Matt Allington (01:15:59): You stick with me, mate.
Rob Collie (01:16:01): ... back and forth, right? Which is like increment by increment we'll pull each other up by each other's bootstraps.
Matt Allington (01:16:06): Exactly. Yeah. But I don't think we have to reinvent the wheel. Right? There's nothing wrong with learning something's good. And then let's let's just move on. We all add something to it. Tom, you mentioned the difference between teaching and lecturing. I mean, I really believe this. In fact, I'm actually just rewriting my book now. So I'm up to the third edition. So Rob, I know you've only done two, so this is opportunity.
Matt Allington (01:16:31): But you and Ken Puls, I mean, you're just such easy targets. So I'm doing my third edition in my book and I just can't... I'm actually really surprised at how much change I'm putting into the book. So I'm currently working on... I just finished all and I'm working on the filter function at the moment. I've changed the way I teach that topic in decks now, because over the years of experience of being in those rooms with the students, looking at the feedback, realizing that it's... You know that look Rob when the eyes just glaze over and you say, "Okay, this message is not going in."
Matt Allington (01:17:12): So you have to find another way. You have to find another way to explain the concepts. So I think we're all evolving. We come back to that same concept we talked about with the software, but you can learn how to deliver a message in a way that people understand.
Rob Collie (01:17:27): I feel compelled to point that Matt as a data guy, can't count. Okay?
Matt Allington (01:17:32): How many additions have you got?
Rob Collie (01:17:34): There's an audio podcast, so not everyone is going to see this, but note to the audience Matt is about to see-
Matt Allington (01:17:39): You're going to say you got three books.
Rob Collie (01:17:41): See the truth. So here's the original.
Matt Allington (01:17:43): I have that.
Rob Collie (01:17:44): This is version one, okay? We can all agree. Okay. That's version one. And then there's the...
Matt Allington (01:17:49): Can't count the PDF as a version, Rob. That's-
Rob Collie (01:17:52): No. Then there's the Australian edition.
Matt Allington (01:17:55): Oh, the mini me. The mini me.
Rob Collie (01:17:58): Which is on the metric size scale. See?
Matt Allington (01:18:04): It's 14% less value
Rob Collie (01:18:08): And then let's not forget the Spanish edition translated by Miguel.
Matt Allington (01:18:13): I forgot about that.
Rob Collie (01:18:14): I mean, I didn't really write it, but okay, fine. As long as we're playing games. And then we have the actual second edition. Then the alchemy book that sold like 600 copies and then we don't even talk about it anymore. So let's just talk about in terms of time scale. You're going to be much more current than any of my books.
Matt Allington (01:18:30): It's amazing how fast this changes. Right? So Rob, your first book was on Power Pivot. My first book was on Power Pivot for Excel 2013. One of my first introductory chapters was it's called a calculated field. Do you remember that?
Rob Collie (01:18:46): Oh, that terrible, terrible moment.
Matt Allington (01:18:48): So they're not measures. They're calculated fields. So the only version of tabular that had a calculated field instead of a measure. But yeah, this stuff changes so hard. It's tough enough doing live training where you can change the slides just before every lesson, which I literally do. But then video training, I know Rob, you've got your own video content. I've got my own video content. Books, they just go out of date so quickly.
Rob Collie (01:19:15): So the thing is the second edition book that's out there and still sells very well is incredibly relevant. There's nothing really about it that's truly gone out of date, but it might look like it has. It has the appearance of not being relevant anymore. If the screenshots are all of Power Pivot instead of Power BI or whatever, right? And the audience knows that. So you feel bad about it. Even if people are still buying it and loving it, they're still just not proud of it. It does look so out of date.
Matt Allington (01:19:41): But the transition is complete from Power Pivot. I mean, I don't know what the future is. I know how he's spending a lot of time trying to stabilize that product, but I don't see my opinion from looking on the outside in is that I don't think that's going very far. I mean, I have two versions of my book, Supercharge Power BI, Supercharge Excel. I updated them both together and the Power BI version sells 99 to one. In fact, I'm probably being generous there.
Rob Collie (01:20:11): Wow.
Matt Allington (01:20:11): So Bill and I have got a warehouse full of green ones and we've reprinted the yellow ones a few times. So suffice to say that the green ones will not be updated this time around. It's all about Power BI. And some people, the people that have been on the journey with us, they get it, right?
Matt Allington (01:20:29): So if I go back through my old blog posts, I talk about... I've got blog posts that talk about data modeling and Power Pivot. And some of the newer people don't even know what I'm talking about. Whereas of course it's still relevant. It's exactly the same. So yeah, it's a challenge.
Rob Collie (01:20:45): Well, Matt, I have really enjoyed this.
Matt Allington (01:20:48): Yeah, me too.
Thomas LaRock (01:20:49): Yes, Matt. It was lovely meeting you again.
Matt Allington (01:20:52): Yes.
Rob Collie (01:20:53): Hey, Matt. Can you give him a wave so he can remember?
Matt Allington (01:20:54): I'll be the guy with a selfie stick saying, "Hey, can I have a selfie?"
Thomas LaRock (01:20:58): Yes, that's him.
Rob Collie (01:20:59): That's him.
Thomas LaRock (01:20:59): That's him.
Matt Allington (01:21:02): Thanks, Tom. Good to meet you, seriously. And it's just great to chat. It's been a while, Rob.
Rob Collie (01:21:08): Matt, I'm really, really happy we did this.
Announcer (01:21:11): 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 Luke P, L-U-K-E-P @powerpivotpro.com. Have a data day.
Sign up to receive email updates
Enter your name and email address below and I'll send you periodic updates about the podcast.