The OG Internet Tech Celebrity, w/ MrExcel's Bill Jelen

Raw Data By P3 Adaptive - A podcast by P3 Adaptive - Tuesdays

Categories:

Bill Jelen, known to many as Mr. Excel, was one of the first Internet Tech Celebrities. He turned his online resume into a highly recognizable brand and is a legend in the data community. We get his story, in his own words.  We hope you'll enjoy this candid conversation with Mr. Excel, Bill Jelen   Buy Bill's book Mr.Excel 2021: Unmasking Excel, containing his favorite Excel tips!   Stuff We Mentioned: Hipster Ipsum Copy Filler Tool Dax Studio We Report Space-Bill's Other Website! Rob's Wheel Of Inquisition   Episode Timeline: 3:15 - Mr. Excel's Origin Story 18:05 - The Excel vs Power BI user base 35:35 - Dax is Badass! 47:25 - Lambda Functions in Excel, Mr. Excel asks some Python questions, and Rob gets cynical 1:16:50 - Bill's other passion We Report Space and the story of Tickling Keys 1:28:30 - More on Lambda functions and Let functions and the immense potential there Episode Transcript: Rob Collie (00:00:00): Welcome friends. This week, we have the one and only Mr. Excel himself, Bill Jelen. Chances are, if you're listening to this, you have been to Bill's site, mrexcel.com via search engine more times than you can count. You might have read some of his books because he is factually the most prolific author of Excel books in history. One of the things that I find most fascinating about Bill is that every technical community, these days, is a community. We're familiar with the idea of, essentially, celebrities, MVPs, community figures, advocates, internet stars. Rob Collie (00:00:35): And Bill was doing that long before that was a thing. He was doing this in the 1990s. When websites still looked like the old Windows help files. He might have been the only person that I'm aware of doing this thing that early. He's absolutely a fixture in the data community as far as I'm concerned. And just also a phenomenal person. He's been a really good friend of mine for a number of years. We'll sometimes talk early in the morning when I'm still up and he is already up. We had a great time, wandered all over the place as you'd expect. I hope you enjoy it. So let's get into it. Announcer (00:01:13): Ladies and gentlemen, may I have your attention please? Announcer (00:01:17): 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 (00:01:35): Welcome to the show, Bill Jelen. Bill Jelen (00:01:38): It is great to be here. Holy smokes. Rob Collie (00:01:41): How you doing today, man? Bill Jelen (00:01:42): It is a beautiful day down here in sunny Florida. 73 degrees. Rob Collie (00:01:47): It sounds awesome. A little sort of Oceanside experience for you. We have freezing rain today. Bill Jelen (00:01:54): That's special. Freezing rain. That's that Indianapolis thing. A little bit further south from Cleveland, the snow changes to freezing rain. Rob Collie (00:02:02): Yeah. I highly recommend it. Bill Jelen (00:02:03): Yeah. Rob Collie (00:02:04): I don't even know what freezing rain was when I ... I didn't know it was a thing when I lived in Florida. Bill Jelen (00:02:10): Great. Rob Collie (00:02:10): Oh, it's just wonderful. The things that ... What was it, the Jimmy Buffett Changes in Latitudes? Bill Jelen (00:02:16): Yes. Rob Collie (00:02:16): That brings you freezing rain. I think he might have been referring to going South. Bill Jelen (00:02:22): Yeah. It's something else. Rob Collie (00:02:22): With a latitude change, it might have been- Bill Jelen (00:02:24): Yeah. Rob Collie (00:02:24): So Mr. Excel, you might be, if not the first, one of the sort of original, the OG internet tech celebrity figures. Everyone's trying for that today. That's a thing today. That's the lifestyle. A whole generation is attempting that right now. Bill Jelen (00:02:44): Right, yeah. Rob Collie (00:02:45): But you, I mean, you were doing this long before it was cool. Bill Jelen (00:02:49): It was easier when there was no competition. Rob Collie (00:02:51): Well, I- Bill Jelen (00:02:52): Yeah. Rob Collie (00:02:52): I agree. Bill Jelen (00:02:53): Yeah, right. Rob Collie (00:02:54): But at the same time, there's a reason why there was no competition. Because no one else thought to do it, or had the energy to do it, or thought that there was ROI. So in some sense, it was harder to start early, right? It was harder to be one of the early people otherwise there would've been more. But yeah, once you're going, it's different. So let's go back to the origin story. The Marvel comic book, or DC, if you prefer- Bill Jelen (00:03:18): Right. Rob Collie (00:03:19): Origin story of Mr. Excel. You were not always Mr. Excel. You were once just Bill Jelen. Bill Jelen (00:03:25): That's true. Rob Collie (00:03:26): Where does it all begin? Bill Jelen (00:03:28): And the nasty secret is that I was originally Mr. Lotus 1-2-3. Thomas LaRock (00:03:32): Oh. Rob Collie (00:03:32): Oh my God. Bill Jelen (00:03:33): Yeah, it wasn't until 1995, when the company I was working for switched over from Lotus to Excel. And then Mr. Excel started in November of 98. So I only had three years of Excel under my belt. Before that- Rob Collie (00:03:44): Whoa, you were three years in before you became Mr. Bill Jelen (00:03:48): Mr. That's right. Yeah. Rob Collie (00:03:50): So were you literally Mr. Lotus 1-2-3? Did you call yourself that at one time? Bill Jelen (00:03:55): No, the whole thing is, I was the guy at the company. There's one of these at every company who knows everything about spreadsheets. Originally Lotus, and then Excel. And it wasn't because I was a spreadsheet geek. It was because we had a miserable tool. They paid $100,000 for a reporting package that did not work. But thank God it had the most important button in the world, export to spreadsheet. Yeah, right? And so then there ... I'm supposed to be using this $100,000 tool, but I'm back in a cube in the back of the finance department. We're using Lotus 1-2-3 to turn everything around. Right? Rob Collie (00:04:31): We think of that button now as, export to Excel. Did they have the foresight in the Lotus 1-2-3 days to call it export to Excel? Was it already called export to Excel, because they knew sort of where it was going? Bill Jelen (00:04:41): No. I'm going to geek out here. It wasn't even an exported .WK1. It was a .WKS. A Lotus 1-2-3 version 1 file, which was limited 2084 records, or some crazy low number like that. Yeah. So the first thing we do was convert that to proper WK1 file with a 16,384 records. Woo! Rob Collie (00:05:02): Get out of here. Bill Jelen (00:05:03): Yeah, right. How would you ever fill that up? Rob Collie (00:05:07): That's too many. Bill Jelen (00:05:07): Yeah. Rob Collie (00:05:11): Yeah. That's great. So I've heard this story before, by the way. This expensive reporting package that doesn't do what you want it to do. But usually there's more zeros on the price tag. 100,000. That's- Bill Jelen (00:05:26): Yeah, that was 1990. I don't know. That was 1995, 1996 dollars. So- Rob Collie (00:05:28): Oh, so in today's dollars it'd be like a million? Bill Jelen (00:05:31): Probably. Yeah. Rob Collie (00:05:32): Yeah. Great. Was that really sort of the origin of your spreadsheet engagement? Bill Jelen (00:05:38): That's how I became really, really good at spreadsheets. Just using this $100,000 tool and then secretly a $99 spreadsheet to produce everything. So that trial by fire taught me a lot. Rob Collie (00:05:49): I'd always thought that, and I've known you a long time now, I'd always thought that you were just sort of in the accounting world. And of course you use spreadsheets in the accounting world and you got good at it. But to hear you tell the story ... Seriously for me this is, I think, the first time I've heard this, that this really started out for you, the primary driver anyway, was a need for reporting a BI need essentially. Bill Jelen (00:06:13): Essentially. Yeah, that's right. And it was ad hoc. I mean, we were never running the same report. It was always ad hoc, which I loved. I loved ad hoc. I loved every day. You'd show up to work and see what fire is burning and what fire you could put out by being the guy who could get the data from the mainframe into, well, spreadsheet. Rob Collie (00:06:31): We have a name for that kind of workflow in the industry. At least at P3 we have a name for it. What fire's burning every day, the different questions every day, the things that twists and turns and all that, we call that reality. Bill Jelen (00:06:45): That's a great name. Rob Collie (00:06:47): You like that? I mean, that's just life. That's just how the world works. That's really, really cool. Seriously, I did not expect, at least this early in the conversation, to be learning something new about the Bill. Bill Jelen (00:06:59): Right. That was it. So I'd been at this company for 10 years and I'm the first generation, I think, that you didn't go someplace and work for your whole life. My dad showed up at a place and worked there for 35 years until retired. And so this place I was working at, they fired people every Friday. HR coming towards your office with a box. It's like, "Ur" Rob Collie (00:07:19): It's just part of the ritual. Bill Jelen (00:07:22): Yeah. Rob Collie (00:07:23): It's like Stalin. It just goes down the row of officers and just shoots every 10th one. Bill Jelen (00:07:26): Right. Right. And you were just hoping that you were not the one that would be let go that Friday. And then they announced. They called everyone out to the parking lot. They made a big announcement that there was a hostile takeover bid. And our hated competitor, our dreaded competitor, was trying to buy us out. But don't worry, we're going to find a white night. We'll find someone else to buy us. Because we hate those guys. And I'm like, "Oh man, this is bad. I'm going to have to go look for a job." I haven't interviewed in 10 years. I don't even have a resume. So in my twisted mind, it was the Friday for Thanksgiving in 1998, I launched Mr Excel. I said, "Send me your Excel questions." And the website said, "I'll write up the best one every week here on the website." Bill Jelen (00:08:08): And the best ones will show up in a book. But what I was secretly thinking was if I could get 50 people to send me Excel questions, then when I lost my job in a couple of months after the hostile takeover, I would have 50 places to send a resume. I mean, anyone who sent an Excel question in, clearly was using Excel. So- Rob Collie (00:08:24): That's right. Bill Jelen (00:08:25): It was a job. It was a job hunt. But then they backed out of the deal. They didn't buy us. They didn't buy us. So now, all of a sudden, I got people sending me Excel questions every day. What am I going to do with this? So that's how it started. Rob Collie (00:08:38): Oh my God. I love that. So that's why I started Power Proof at pro.com originally. Was like an online resume in a way, and had no idea. I mean, I kind of had an idea, but it was a squishy motivation. I love how the accidental things like this- Bill Jelen (00:08:53): Yeah, absolutely. Right. Rob Collie (00:08:54): Yeah I know. Bill Jelen (00:08:54): Right. Rob Collie (00:08:55): It didn't even turn out to be real. The takeover wasn't even real. Bill Jelen (00:08:58): They did come back and get us two years later. And that actually is really lucky, because in those two years ... They say the way to build traffic, you can either buy your way or ... It takes time or money. Either buy ads or just slowly grow organically. And those two years give us a lot of time to grow organically. And then when it came time that they offered us a buyout, I said, "All right. Well, I got six months' severance package here. Let me see what I can do. Let's just hang a shingle out to see if I can get Excel consulting gigs." And I had to wake up and hope that something would come in. There was only one day that nothing came in. So yeah, by the end of the six months, it was a go. Thomas LaRock (00:09:31): See that part has always fascinated me. The fact that, until I met Rob, I had no idea anyone could make a living as an Excel consultant. Bill Jelen (00:09:42): It's amazing. It's amazing. And it certainly requires some VBA. Back then you had to do some VBA. And they were tiny little VBA jobs, like 250 bucks or something like that. But you string enough of those together and you're replacing the old corporate salary. So yeah, it was good. Rob Collie (00:09:59): So When did the book start? Bill Jelen (00:10:01): The book. Okay. Here's the worst part about working from home. First off, the cafeteria at home is last night's dinner, right? Whatever's left over from last night's dinner. There's no choice whatsoever. And I missed not having anyone to talk to. There was no one to talk about last night's game, no one to go golfing with, no softball teams in the summer. There's nothing. And so I started joining the Chamber of Commerce, just to have someone to go talk to. You could go out to breakfast once a month at the networking thing. And this is so weird. There was an event called ... It was the Women's Network. I don't qualify for that, but they said, "It's okay, come on anyway." Bill Jelen (00:10:41): So I would go to the Women's Network thing and there was a lady there who did a presentation and she says, "Yeah, I'm a marketing consultant. I just got started. I want to triple someone's business." And I went up to her afterwards. I said, "Well hey, things are going really good. I'm satisfied with where we're at. But yeah, if you want to triple my business, let's do it." So I explained the business model. I said, "I got all these people coming. They post their questions at the message board. 99% of them get their answer for free. And that's it. They never come back. And I'm counting on one out of a hundred to actually hire me and pay me 250 bucks." And she's like, "Wait a second. You're either charging people $250 or zero?" Bill Jelen (00:11:20): I'm like, "Yes." And she says, "You're missing an important price point. You need a $20 product. If people love you, if you had a $20 product, they would be handing $20 over to you handed fist." So I wrote the first book. The print run. Rob, what was our last print run? Our last print run was thousands, Right? Rob Collie (00:11:38): Yeah. Bill Jelen (00:11:38): It was just some crazy number. My first print run, on Whipple avenue in Canton, Ohio, at the office, Max Printing store, was 10. I printed 10. Cost $10 each. I sold them for 20. Cool 100 bucks right there. Rob Collie (00:11:53): Humble beginnings. Bill Jelen (00:11:54): I took that profit, I printed another 10. And eventually the day I got to print a hundred, it was like, "Woo! Wow, moving up." Rob Collie (00:12:02): You were self-published from the beginning. Bill Jelen (00:12:06): From the Beginning. That's right. Because I hated rejection. I didn't want to go to the publishers. So I just printed, myself. And then once I started getting five star reviews, and then the publishers came to me. And great news, even at Taylor, the marketing consultant, she wanted to triple someone's business. Within 18 months, the book sales had outpaced consulting 5 to 1. And I said, "Oh, look at this." I just write a book once and sell it for 20 bucks. Bill Jelen (00:12:29): I don't have to worry about writing VBA for someone. And the VBA always worked until that person left. And then the new person who took over the job. Yeah. Like, "Hey, your program doesn't work." "What do you mean my program doesn't work? It worked fine for six months. Where's Shelly?" "Oh, she's not here anymore." "When did it stop working?" "Right when she left." I was like, "Ur." And just having to deal with that ongoing tech support was no fun. No fun at all. And the great thing about a book, I've hardly had anyone ever call and say, " Bill, the book won't open." There's just ... That never comes up. That never comes up. Rob Collie (00:13:08): Yesterday, there were words on the page. Today, there's no words. Bill Jelen (00:13:11): Right. Yeah. I had a back cover that got scratched once. So I had to replace the book for that. But it's very, very low probability for bugs to pop up once the book is out. Rob Collie (00:13:22): Yeah. Not a lot of support calls. Bill Jelen (00:13:24): No. Rob Collie (00:13:24): For a book that ... huh. Bill Jelen (00:13:26): And I'm going to whisper this part. I'm not sure I want anyone to know this. I've always said, if you have a great cover and about 10 pages of content, 90% of the people will be happy with that. Because they never actually read the book. They buy the book and they put it on their shelf. So if you just have enough stuff at the front, and then you could just put 190 pages of white space, most people wouldn't complain at all. Rob Collie (00:13:44): Yeah. I think Salvador Dolly, or one of those artists said something similar back in the day, which was like, "Once you've made a name for yourself, once you've established it, after that, it's all just production. Just churn it out. People will buy anything. Once you've started, you just slap some paint on the canvas. And its more important to get quantity than quality." Really cynical, but you kind of get the sense he was speaking from truth. Wasn't he? Bill Jelen (00:14:09): So now, just in the last 30 seconds, I've hatched this grand scheme that I'm going to get a word cloud of all the Excel terms and then use some huge TEXTJOIN function in Excel and just start generating content automatically. Random. Use ram between in the word cloud, and frequency distribution. I bet I can just start, yeah, knocking out. Rob Collie (00:14:29): Have You seen the HipsterIpsum site? Bill Jelen (00:14:31): No. Rob Collie (00:14:32): So you know about Lorem Ipsum, right? Bill Jelen (00:14:34): Yes. Right. Yeah. Yeah. Rob Collie (00:14:35): Fake Latin text that you fill in on a design, like a webpage, before you're ready to write the actual copy. Well, there's a website, it's HipsterIpsum. and it is so funny. It's like, I'm for loco, farmed a table, sustainable. Bill Jelen (00:14:53): That's it. That's it. I just need to hack their website and put my Excel terms in instead. Rob Collie (00:14:57): Yeah. Bill Jelen (00:14:57): And we're good to go. Rob Collie (00:14:59): SpreadsheetIpsum. Or GridsterIpsum. Bill Jelen (00:15:01): GridsterIpsum. Rob Collie (00:15:02): Something like that. I'm VLOOKUP cross reference relative name range. Bill Jelen (00:15:09): Yeah. Thomas LaRock (00:15:10): It's like you're trying to give yourself a rapper name. Bill Jelen (00:15:12): Yes. Rob Collie (00:15:12): Yeah. It is. And a really bad one at that. Thomas LaRock (00:15:14): VLOOKUP. Rob Collie (00:15:17): Yeah. Bill Jelen (00:15:18): There was a service, I don't know, 15 years ago, that you could put a website in and it would translate it to Snoop Dogg. You could send that link ... I said, "Hey, I rewrote my website. What do you think?" You're not serious. Are you? Rob Collie (00:15:32): All right. So you started essentially in BI. Excel is, even today, the BI tool of choice for the entire world. I think Power BI is slowly colonizing that space. And this is sort of the core of what our business does. So at some point though, it seems like you became not ... Once you went sort of solo, the majority of the work you were doing for people and the topics covered in your books and everything, doesn't have that same heavy analytical slant. I mean, there's still some of that for sure. You've written books 100% on Pivot Tables for instance, but you've become a lot more, I think, general purpose in your focus over the years. Would you agree with that? Bill Jelen (00:16:15): And that's because of the questions that are coming in. At this point, the next YouTube video is just going to be the last good question that came in, whatever it happens to be. That certainly goes from taking massive amounts of data down to a one-page summer report to whatever crazy problem someone's having today. Rob Collie (00:16:32): Yeah. There's a lot of that, especially in the early blogging days, right? Right now I'm scrolling to see if there's a function in Excel that starts with MC, MC something. Oh, there's only MD. Nope. There's no MC Excel function. We're going to have to keep looking for our rapper name. If we ever assembled the brain trust to pick the Excel rapper name, this is it. This is the chamber where that happens. Bill Jelen (00:16:53): Sometimes in Excel, you'll use our rapper function. On the video, have a little call out there of some famous rapper. And it's like, not this. This. Rob Collie (00:17:01): Yeah. Bill Jelen (00:17:04): Yeah. Rob Collie (00:17:05): So we are, tangentially anyway, this podcast is usually ... It ties back to Power BI somehow. You haven't really had to make that shift. You have been. I've seen you. You've been dabbling, especially on the Excel side, like the Power Pivot and especially the Power Query side. You've been dipping your toe into those pools. But some of the Excel crowd of yesterday have sort of fully migrated. They now identify as Power BI people. Bill Jelen (00:17:34): I can tell that from book sales. If you go out to Nielsen and look at book sales, the Power BI books are just outselling Excel books by a huge margin. Rob Collie (00:17:44): Really? Bill Jelen (00:17:44): Right. And I- Rob Collie (00:17:44): Across the board. Bill Jelen (00:17:46): That doesn't mean that there's more Power BI people than Excel people. It's the Excel people already know what they're doing. And the Power BI people, the people who are migrating to Power BI, they're the ones who need help. If you've been using Excel for 20 years, you don't need another Excel book. But the people who are just switching to Power BI ... When you said it's slowly moving, I think it's ... Whatever that rate is, that rate is increasing. That's a great thing. Rob Collie (00:18:06): There's also more leverage on Power BI skill than on Excel skill. It's not to say that Excel skill is less valuable, because even I today, I'm constantly throwing together spreadsheets because there's no substitute for a blank canvas that you can compose in any way. And for example, if you want to build yourself a wheel of fortune type app that picks a random person that is then forced to ask a question of you on the spot in a team meeting, there's only one go-to, and that's Excel. Bill Jelen (00:18:35): You're definitely not going to go to Word or PowerPoint for that. Rob Collie (00:18:37): They're not Power BI either, as it turns out. But in terms of institutional impact, a single person effectively applying Power BI can be worth millions of dollars to the bottom line annually. It's crazy, right? A single person applying Excel isn't going to have that kind of leverage. You kind of need a whole culture of people applying Excel simultaneously. You don't get this one to many with that kind of impact, typically. I mean, except for the famous stories where someone messed up a spreadsheet, right? In which case you can subtract. You can subtract millions of dollars from the bottom line in an eye blink if you want. So I think there's, in essence, sort of a higher premium on Power BI learning. I guess what I'm really saying is, is that it's not just the people who know Excel and there's all that. People are more willing to pay for Power BI expertise than they are willing to pay for Excel expertise. At the same time though, there's still many, many times more people who are interested in acquiring Excel expertise even today, relative to Power BI. Bill Jelen (00:19:43): Do you think that there'll ever become a tipping point where the Power BI user base is more than the Excel user base, or will it always be- Rob Collie (00:19:51): No. Bill Jelen (00:19:52): Is there some limit? And you're 40% Rob Collie (00:19:54): I think there's definitely a limit. For the same reason as that leverage, right? You just don't need as many people in an organization slinging Power BI. You need more people in your organization than what IT can provide, but less than what you need from Excel. So It's somewhere in between there. So I think maybe the Power BI audience might be 10 to 30% the size of the Excel audience at equilibrium. That's just me pulling a number, but that's a gut feeling that I've got. Bill Jelen (00:20:24): That's still a ... That's going to be a massive number though. Rob Collie (00:20:26): Yeah. Bill Jelen (00:20:26): Right? So- Rob Collie (00:20:27): If you think of another way it's, take the fraction of Excel usage that is making sense of reports. That's that same origin you had back in the 90s of exporting from one size fits all reporting systems, but we answered no questions. The Excel as BI backstop is a significant percentage of Excel's total usage. It's double digit percentage. It's 10 to 30%, right? That's the space that Power BI is on its way to colonizing. But I still think that's early. I still think it's early. In fact, I would predict that the sales of Power BI books, the sales rate of them is still ... We have not yet seen its peak. Bill Jelen (00:21:09): I would definitely agree with that. And my anecdotal evidence for that is just the number of Excel people that I run into. So up until this year, I was on the road, right? Doing Excel seminars. And so you'd show up in a city and you'd have 100 people from around the city show up, who were all working in accounting. There to get their continued education hours. I would always try engage. When I got to the Power Query point of the day, show of hands, how many of you have used or heard of Power Query? And it was early on, it'd be one or two out of 100. And even at the end, it was 10 or 12 out of 100. So the number of people who still have no clue that Power Query is there ... And for me, power query is the gateway. Bill Jelen (00:21:54): Once you realize that Power Query is there, then you take that Power Query knowledge and move over to Power BI. Because now you can take that data, and instead of spending an hour to clean it, you'd spend 10 minutes on day one and then zero minutes, just click refresh, on day two through day 50,000. Once you learn that you can do that, then to me, the natural next step is, do that with really cool visuals on top. Rob Collie (00:22:18): Yeah. Bill Jelen (00:22:18): Through Power BI. Rob Collie (00:22:19): We're trying to track down one of the originators of Power Query. One of the shadowy figures behind the scenes at Microsoft and get them on the show. Stay tuned. Fingers crossed. We're hot on the trail. Bill Jelen (00:22:32): I would love to be a fly on the wall for that. With a really large megaphone, Right? And just- Rob Collie (00:22:37): That's a fly. Bill Jelen (00:22:38): To shake them and say, "How did you manage to get this product in Excel with no one realizing that it's there?" Excel 2013, it was an add-in. And then Excel 2016, they moved it to the data tab. But if you look at the group, the Get and Transform group, and the group that it replaced, if you're not looking at them side by side, you would swear it's the same thing. Right? So if someone comes along, they look at the data tab, oh, this looks like just what we used to have. Rob Collie (00:23:05): Yeah. The Power Query team didn't achieve that. Don't worry. The Excel team takes care of hiding it. Bill Jelen (00:23:10): Okay. Rob Collie (00:23:10): And that is actually a really tough challenge. Dave Gainer, my old mentor, who was my boss when I was on Excel, he's not here to defend himself. So let's pick on him. He had this thing. I'm not really picking on him. I think he's right. And this is the challenge. I would always want to introduce things into Excel that looked different, kind of stood out, that were their own thing, that did something. It was almost like self advertising in its difference. Bill Jelen (00:23:33): Okay. Rob Collie (00:23:34): And he was always macking me around saying, "No, we're custodians of this product. And we are not going to turn it into some tourist trap of garish signs all up and down the road." Little hole in the wall, firework stands and all these one off things, right? Rob Collie (00:23:49): Those things have a cumulative cost to the product that really adds up. And so you get this legitimate and well intentioned, integrative, holistic philosophy as a side effect of that. You can't tell that anything is new. The things that should be screaming at you, for example, the 2013 awful, awful aberration that we all are pledged to never talk about, but we're going to talk about. When they renamed measures in Power Pivot to be calculated fields for an eye blink in time, right? And you and I know that calculated fields were an existing feature of Pivot Tables and they weren't very good. They were like less than 1% as interesting and powerful as measures. Rob Collie (00:24:34): I mean, I can't even put it on a numerical scale. It's thousands of times less powerful, thousands of times less relevant. But out of consistency, oh, we'll just call it calculated fields. It's the new calculated fields. But when that ends up happening is, you give it the same name as the boring old crappy thing that everyone always hated. And now why would I ever click that button? Instead, that button should be glowing. Three dimensional, jumping out of the screen and grabbing you by the nose saying, "I am the best thing ever. Click me." But if you've got 40 things in Excel saying. Rob Collie (00:25:03): Click me. But, if you've got 40 things in Excel saying that, what is it on the Incredibles? "When everybody's special, nobody will be" Bill Jelen (00:25:12): No, I've always said there should be a red neon arrow pointing at certain features, and that certainly is one. Rob Collie (00:25:19): I've got an idea. It'll be almost like the Amazon rating system. User enthusiasm for features. Data driven, light up. So now it's first of all, it's factual. The market's deciding. Bill Jelen (00:25:35): That's beautiful. Rob Collie (00:25:36): And you can turn it on and off, and- Bill Jelen (00:25:39): No, no, no. Rob Collie (00:25:40): ..Right there. Microsoft, I hope you're not listening because I just gave one of my best ideas ever to you for free when really you could be paying me large amounts of money every year to generate two ideas like this per year. Bill Jelen (00:25:52): It sounds beautiful. But I just followed it through through its logical conclusion. And what we would have in an Excel user interface is a button that consumed 98% of the space with Autosum, just based on what the users use that's, Would, The whole home tab would just be Autosum. That would be it there'd be nothing else. Rob Collie (00:26:10): You're saying this would drive us to the bottom, this would drive us to lowest common denominator? Bill Jelen (00:26:15): It would have to be user usage of the people I like, I would get to nominate the 100 people that actually control what gets on there. Rob Collie (00:26:23): I like it, a curated, an MVP curated system? Bill Jelen (00:26:28): Just my friends. Rob Collie (00:26:29): Do we want Ken in charge of that? Bill Jelen (00:26:31): Everything would Be shouted then. Rob Collie (00:26:34): Very intelligently though. Like the most intelligent, booming voice ever met, We need to get Ken on here too. Don't we? Bill Jelen (00:26:41): You do need to get Ken on here. Rob Collie (00:26:42): All right Ken, we're coming for you. Hey Tom, I'll tell you an interesting... If we're going to keep wandering whatever we want do. I'm going to tell you an interesting detail about our business relationship with Bill. So it's a fact that in terms of retail dollars, our book deal between P3 and Bill has done millions and millions of dollars in retail sales. It's a cool thing to say. Right? I don't think it's eight figures, but I think we're somewhere in the seven figures of retail sales. Bill Jelen (00:27:13): Retail sales. That's right. Rob Collie (00:27:15): And we've never had a contract. Bill Jelen (00:27:18): It's a handshake. Rob Collie (00:27:19): Yeah. I don't think we even shook hands. I think we were you know, Email. It was a distant handshake. Bill Jelen (00:27:25): I think at one point you said, just in case something happens, we should write all this down. And I said, oh yeah. And then we never did. And that was like four or five years in. Rob Collie (00:27:34): It's just like that voice that comes into your back of your head like, oh come on, we're really being irresponsible. But then the short term incentives, they straighten you right out, we were never going to Sue each other. Bill Jelen (00:27:46): Your book is the best selling computer book that I've ever seen by far here. Of all the books I've I published, you've outsold every book. Rob Collie (00:27:53): It was a good handshake. Bill Jelen (00:27:54): It'd be interesting to go out and find the best selling computer book, something back from the day and see if you've eclipsed that. Rob Collie (00:28:02): I'll tell you what let's do that offline. And if the results look good for us, we'll circle back. If, they don't look good. We just won't bring it up again. Bill Jelen (00:28:14): Let us just cut this part [inaudible 00:28:14] Thomas LaRock (00:28:14): I don't know why you don't just say how good they are. Rob Collie (00:28:16): We just bluff it. It's optimistic. We'll say it now on the podcast. And then if it turns out to not be true, then we'll cut it. Thomas LaRock (00:28:21): If you say it enough times, it becomes true. Isn't that the lesson that we've learned? Bill Jelen (00:28:25): What a horrible lesson it is, but yes.[crosstalk 00:03:27] Rob Collie (00:28:29): My book is more popular than the Bible. Bill Jelen (00:28:32): Rob's book has sold more copies than the DaVinci code...For example. Rob Collie (00:28:39): In parentheses, for Excel. Bill Jelen (00:28:42): You know, that would be another great book. The DaVinci code for Excel. And that's the one where you only need 10 pages of content and 90, 190 pages of Blank stuff. Rob Collie (00:28:49): And 90% of the production cost is just licensing a picture of Tom Hanks for the cover. Right? Bill Jelen (00:28:54): That's it. And most of the sales will be people who accidentally, they were trying to buy the real DaVinci code and they bought us instead. Rob Collie (00:29:00): Tom's sitting down at a computer with a spreadsheet on it and he's asking the photographer, why are we doing this photo shoot? What is it again? He's like, "don't worry about it you're getting paid". Thomas LaRock (00:29:10): So how'd you two meet? We haven't talked about that yet. Bill Jelen (00:29:13): Cleveland, Ohio, Thomas LaRock (00:29:15): You have this incredible non-handshake deal, but what's the origin story for the two of you? Bill Jelen (00:29:21): It was Cleveland Ohio. When David gainer was in charge of the Excel team, I got a note from David. I'm living in Akron, Ohio. He says, one of my guys is relocating to Cleveland, Ohio. You two should go to lunch. Know what I got from Dave gainer? I reached out to Rob and I went up to Cleveland and had lunch. And it was the best conversation I think I've ever had because I, sitting in Akron, Ohio have hatched all sorts of conspiracy theories. I was the Qanon of Excel. You guys on the Excel team, you did this because. And then Rob would say, well, you know the story there is, and then he would give me some boring story about why it made sense to do that. And I hated that because my story was so far, much better. My made up conspiracy theory was better than yours, but it was just great to have someone who was almost equally as frustrated with things as I was, You could tell me the backstory behind it. Thomas LaRock (00:30:23): Does knowing the backstory, did that make things better for you? Bill Jelen (00:30:27): It made me less angry I think. Sometimes. Rob Collie (00:30:30): So many mundane reasons for things, right? Just normal big company, hands are tied type of stuff. It's just, On one hand it's reassuring because it's not nefarious. We weren't hiding in the basement of a pizza parlor in Washington, DC, but at the same time, it's kind of depressing in a way, because those sorts of forces are durable. If they were acting yesterday to do things we don't like, they're going to be acting tomorrow to do the same things. Like you'd almost prefer it to be some sort of one off, twisted thing, because tomorrow's going to be different. But when the causes for things are corporate, well corporate's not going anywhere. And we see it over and over again. Bill Jelen (00:31:13): The other thing I've realized is you cannot be angry when a version one feature doesn't do what you want because at any other company, they wouldn't have put it out. They would wait until they had version five and then put the finished thing out. But Microsoft, especially in Excel, they'll put something out just to kind of get the feedback and they'll send it to the insiders fast, and so it's going to some small percentage and they'll react. People will say, Hey, this would be better for you, do this. And they're not afraid to change things. So it used to be like, that's so stupid, That new feature. And now I'm just like, hang on it'll come around, this five years from now this will be the greatest thing ever. But right now, this is what they have done. And they're letting us play with this part. So. Rob Collie (00:31:53): Let's go back to the origin story of us meeting from [inaudible 00:31:55] I think the statute of limitations is the now expired on us needing to keep this secret. So you were writing about the time that we had lunch that first time you were writing a Power Pivot book, you had one in the works. Bill Jelen (00:32:08): That's right. Rob Collie (00:32:10): It wasn't called Power Pivot yet the name hadn't been released, Thomas LaRock (00:32:13): was it still Gemini? Rob Collie (00:32:14): It was still project Gemini. You had gotten basically to the end of this book or close to it anyway. And your book was loaded with screenshots, from project Gemini where the field list said project Gemini at the top and all kinds was project Gemini everywhere. And then out comes the real name, Power pivot. And you're like [Spanish 00:32:37] Bill Jelen (00:32:41): That had to be a very unhappy day. Rob Collie (00:32:42): Do you remember what followed? I can fill it in if you don't. Bill Jelen (00:32:46): Yeah. Tell me what followed. Rob Collie (00:32:47): So the problem was, is that you had to be done with the book by a certain deadline for printing, but you weren't going to be able to get a hold of a beta of the product with the proper names in it, with power pivot in the UI for like another month or so maybe six weeks. It was, You had a real timing problem. It wasn't so much that you had to redo the screenshots cause that wasn't going to take you too long. It was just the fact that you didn't have anything to screenshot. And so the black market came through for you. Right? Shady Rob Collie said, Hey, in the spirit of benefit to my baby power pivot and to Microsoft, I'm going to give you access to my internal build of power pivot that already had the right names in it. You didn't come to me asking for me to do that. You were just like, Rob again, curses!, What kind of evil star chamber has decided to convene at Microsoft and decide to screw me, Bill Jelen personally. Bill Jelen (00:33:41): Yes, Cause that's always what it was. It was just trying to get me. Rob Collie (00:33:46): That's right. Bill Jelen (00:33:46): That's what everyone thinks. Rob Collie (00:33:47): I said, Hey, I got a solution. So we met up again at that sort of sketchy, suburban bonefish grill, parking lot. And I handed you a disc, and your power pivot book was able to go out with the proper on time and with the proper UI screenshots. Bill Jelen (00:34:05): I owe You a bottle of Rum for that or something. Rob Collie (00:34:08): I mean, I'm sad that you've forgotten this because to me, this is what cemented our friendship. The lunch was fun, but it was just going to be fun. But from that point forward, you owed me. Bill Jelen (00:34:18): I remember the lunch, you just had so much power pivot knowledge. Like you would just throw an offhand sentence out and I'd be like, holy smokes. This is something I completely don't understand, and then furiously, at least in my head scribbling notes of things I needed to go back think about in the book. Rob Collie (00:34:35): And, that's how my book was born. At least part of the story for it was that your book gave a comprehensive tour of all the things you could do in Power Pivot, sort of like each what all the different graphical elements do. But you largely had to take a pass on the Dax language because that was a whole universe of its own, and that's how the world's worst book title ever came to be Dax formulas for Power Pivot because it was originally, My book was meant to be a compliment to yours. That's how it all started. Bill Jelen (00:35:10): Why didn't we go with Dax-Vinci code with Tom Hanks and Rob Collie? Rob Collie (00:35:15): We just didn't have the budget back then. We weren't the multimillion dollar international conglomerate powerhouse we are today. Bill Jelen (00:35:23): Let's talk about Dax, let's, just for a second. If we go back to that tool I was using in finance that started the whole thing, a little bit of Dax in that tool would've solved the problem because it was the calculation on the total row that that tool couldn't do. And the fact that total that Dax can basically say, if this has one value, then do this other thing, That ability is I think why Power Pivot and Power BI didn't get caught in that problem of being able to create reports that no one wanted the fact that you can customize that through Dax. It's just phenomenal. Rob Collie (00:35:59): Look at Bill, just so casually throwing out the if has one value, like the ability for a DAX formula to inspect its own filter context and say, I'm over here now, I'm going to behave differently, I'm going to do a different aggregation at the total. Bill Jelen (00:36:14): That's badass. The fact... that is really badass, that one measure can change its behavior depending on whether it's on the parent [inaudible 00:36:22] or whatever. Rob Collie (00:36:24): There's your pull quote, Dax is bad ass. Bill Jelen (00:36:27): Dax is bad ass. Rob Collie (00:36:28): It is, isn't it? Well that in Power Query, are the two of them combined or something else? Bill Jelen (00:36:34): You know what the Power Query, So in my early life in Excel, I was always doing things in VBA, it was VBA. I don't have to do VBA at all anymore because Power Query gets me 99% of the way there. And some of the worst data nightmares that people have shown up in my seminar, I get there early and I walk around and see people have questions and they'll show me their data, some of the worst data nightmares ever you can solve in Power Query in minutes. And the fact that so many Excel people have no clue that it's there hiding back on the data tab. Rob Collie (00:37:06): So for certain aspects of what you used to do, you're now officially or unofficially retired from doing them. You were the seminar workhorse, man. You were like the hardest work in rock band in America, all over the place doing these seminars. How many seminars would you do a year sort of at peak? Bill Jelen (00:37:25): The goal was 35 a year. I was trying to do 35 a year, sometimes a little bit more than that. And as Mrs. Excel would point out 35 seminars a year, doesn't mean that you're gone 35 nights. It means that you're gone 70 nights, 70 days, and 35 doesn't sound like a lot, but 70, all of a sudden sounds like too many. And after doing that for 17 years, it was too much. And I'm glad that I made the decision to stop at the end of 2019 because had I had 35 seminars scheduled for 2020, everything would've been canceled. It just would've. But it, Like now in retrospect, we can say everything would've been canceled, but early on you would be like, oh yeah, this is still going to happen, we're going to make it. Rob Collie (00:38:05): We were touch and go for like our early March public class. It was scheduled for Charlotte. It came down to like, if it had been two weeks earlier, if it had been scheduled for two weeks earlier, or maybe even a week earlier, it would've happened. It was that close. But yeah, we canceled it. It was right when everyone decided, no more travel. So that was kind of a relief because I didn't want to put everybody at risk and it was nice to not have to make the decision, so that's it. That was the last in person thing that we did as a company, It was a year ago. Bill Jelen (00:38:39): I think the same thing for me because I was still going to do things in Florida. So I think I went up to Gainesville last January or Orlando probably early enough for half the seminar or something. Rob Collie (00:38:47): I Got to see on the biggest stage, one of the biggest of these sort of seminar dates, where were we? Tulsa? Bill Jelen (00:38:53): That was Tulsa, I think. Rob Collie (00:38:55): I got to see you in full high priest. I mean, this is a room with hundreds of people in it, it was a big room and- Bill Jelen (00:39:03): 300 people in Tulsa. Rob Collie (00:39:05): ..And there's more than that in the town. Just so you know. Bill Jelen (00:39:11): Some were busy that day I guess. Rob Collie (00:39:13): They didn't all make it. They all RSVP'd, the whole town. I think of myself as a reasonably good public speaker. I got to go before you, so I thought I killed it. And then you got up there and it was like this is next level, you have a lot of funny stories. Bill Jelen (00:39:28): Yes I do. So when I'm on stage in front of accountants, that was my favorite thing to do, because we were going to have fun. We, they had to be there. They had to get their CP credits, but we were going to have fun along the way. And people, if they start to doze off a little bit, it's time for a good, funny story, and I had all kinds of stories and I would love that eight hours, 8:30 to 4:30 being in front of that room. It was just getting there and then getting home that was miserable and the flights and everything. It was definitely fun. Definitely just so much fun. Rob Collie (00:40:05): Can we hear a couple of your favorites? I mean, do we need to pay extra? Do we need those chart, paying appearance fees? If we get some. Thomas LaRock (00:40:11): What happened to the handshake? Come on. Rob Collie (00:40:13): Well, we can make a deal right here. Bill Jelen (00:40:15): So almost always, I would start off with a story, Again these are accountants, these are accountants or finance people. And so I would almost always start the story about when I was sitting up in accounting and I got a call from a woman down in the marketing department, and I go down and she had been in a trade show and they handed her a floppy disc with a list of all the people who had been at the trade show, area code in column A phone number in column B. And she over in column C was typing open parenthesis, area code, closing parenthesis, space, phone number. And she had been doing this for like an hour and Had thousands, thousands to go through. And she doesn't use Excel every day. This is probably the first time she's opened Excel this year. And one of her coworkers walks up behind her and says, what are you doing? Bill Jelen (00:41:04): And he said, I have area code in column A and phone number in column B. And now that guy who was also from the marketing department, he doesn't know either. Because he never uses Excel, but he knows me. Let's call bill Jelen upstairs in the accounting. And there must be some fast way to do this. so then I go down to the marketing department, I show her how to create a formula to concatenate those things together. Then double click, the fill handle and bam!, All 5,000 are done and she's just thrilled. so then I walk back up towards the accounting department and unbeknownst to me, I'm whistling a little happy tune because I know that I've just saved her eight hours worth of work. So I get up to the accounting department and I barely sit down and my phone rings and it's this lady. Bill Jelen (00:41:42): She was so happy. The data she had over there in column C, she says, I don't need this data. And of course people were watching me live, so I click on a drag to B and then right click. I don't need this data anymore. And of course, because I hadn't changed the formulas to values, everything is gone, It's all hash rough errors all the way down. I mean that story. And in the course of that story, you're going to learn how to use concatenate. You're going to learn how to double click the fill handle. You're going to learn how to several different ways, eight different ways, how to pay special values, including rob the way. My favorite trick of all time came from Dave in Columbus, Indiana. So just 75 miles south of you right now, Dave and in Columbus, Indiana showed me a way to copy and paste special values without ever doing the copy in some secret menu that was somehow built into Excel that I had never seen in 25 years of using spreadsheets. But somehow Dave in Columbus, Indiana, and knows how to do this trick. Rob Collie (00:42:35): Well that's, I mean, I don't even know what the feature is, but I know the committee that was in charge of hiding it. We would periodically meet at the pizza parlor, smoke filled room and say between puffs, we'd go. Okay. So we're the coolest things in Excel that we're going to hide this week. Bill Jelen (00:42:56): We have reports of users finding this and we need to put the stop to that. Rob Collie (00:43:00): That's right, it's just a matter of time before Bill sees this and it puts a smile on his face. So we can't have that. Now can we?. Bill Jelen (00:43:10): Like that story from Dave, he's Dave in row one of Columbus, Indiana that all of these people who gave me these tips, I remember them like Derek from row six in Springfield, Missouri, they give me these awesome tricks and it's like, how did you ever discover this? But it doesn't matter. There's 750 million people using a cell. Someone is going to accidentally, discover something. And then I run into that person and they give me that trick and then it's off to the races because I'm going to take that trick, And I'm like the Johnny Appleseed of Excel, you know, I'm going to take it from city to city, to city and make sure that people know, that that's there. Rob Collie (00:43:43): I remember one time I was trying to explain to you, In the early days, one of the cool things about Dax, was distinct count, the ability to get count of unique values and, you agreed that that was really cool, but you disagreed with something else I said in which I said, this wasn't possible in Excel before now. And you said, oh contrary [inaudible 00:44:06], and I said, I was like a bar bet. And I was like you can't do that. And you said, yeah, I can. And then you go through the most elaborate series of Rube Goldberg, gymnastics that I have ever seen. And I knew as you were going through it to not even bother committing it to memory, I could just tell that this was at the end. You had a unique count of something and so now we had a dispute, whether that had actually counted as possible. Bill Jelen (00:44:33): Well, no, I remember you go to the source data, add a new column, one divided by the count, if right. And so if there's seven of them then each one gets 14%, it adds up it's one, right? Rob Collie (00:44:44): My brain is still, right now. Busy, not committing this to memory. Bill Jelen (00:44:48): How could you not know this? You are on the Excel team for God's sakes. But it was, was it, I don't know if it was that same lunch or the other lunch? I was having such a good time because we were resonating so perfectly like everything you said I was, I was absolutely down with that. And then you said the sentence that blew my mind. You said "our internal customers will be really upset with Power Pivot because they can't use Get-Pivot data". And I'm like, what? No, Rob, you don't understand Get-Pivot. Data is the worst thing ever. Everyone wants to know how to turn it off, and you said, "that's the people outside of Microsoft. I'm talking about our internal customers, the accountants at Microsoft". And then you proceeded to show me how they use Get-Pivot data. I'm like, that was Rob from my table in Beachwood, Ohio. Rob Collie (00:45:36): Yeah. That's right. Row one of the fish grill. Rob in the front row with the long island iced tea, So now I think that we came full circle on this, on the Wayne Winston episode. Bill Jelen (00:45:54): That God mentioned on Wayne's Wayne's episode. Rob Collie (00:45:56): we think that the reason why everyone internal to Microsoft was, was using that technique was that Wayne had been teaching it. Bill Jelen (00:46:05): He said that. Rob Collie (00:46:07): And I didn't know that until years later. So it's just like, oh my gosh, the power that one person's innovation can have when you share it is intense. It went from Wayne to like this diffuse audience of all these like finance professionals at Microsoft. And I had to go and like sort of circulate amongst them to pick this up, that this was something that they were all doing. It took a long time to sort of absorb that. And then I take that and transmit it to you and you light up and you go, I could imagine there had been some accident where you take Wayne Winston aside one day and say, Hey Wayne, have you seen what you can get? And it'd be like, that's how you'll write your name on a dollar bill and it comes back to you five years later. So if Power BI had existed back in the mid nineties, that's where you would've gone. Bill Jelen (00:46:58): Yeah. That's true. Rob Collie (00:46:59): The Excel route took you a lot of different places than what Power BI takes you, but based on sort of your first really intensive collision with spreadsheets, that's power BI's turf today. That's really interesting. Isn't it? Bill Jelen (00:47:13): It is. It's time and place. When did that happen? When did I happen to end up in that finance department in Akron and need to create those reports and what were the tools available? So yeah. Rob Collie (00:47:23): Have you gotten deep into Lambda functions yet? Bill Jelen (00:47:25): Lambda functions are awesome and I want to embrace them, but what's funny is so many other people are embracing them better than I am, The recursion in Lambda functions is just wild. And then the people who we could let inside of Lambda functions. Rob Collie (00:47:44): Wait. This just came out. Bill Jelen (00:47:46): Yes. Rob Collie (00:47:47): Okay. So I only know of them through, through the stuff I've been doing in the past few years with Python. And so when you asked him just now Lambda, I was thinking we're going to start talking Python. I got really excited. So I just did a quick search. It just came out in December or there's a post from December. So it was just released in the latest version. Bill Jelen (00:48:06): Just released. It's still in beta. Thomas LaRock (00:48:07): It sounds like it's probably does the same thing because you also said recursion. So sounds like it's similar to the functionality I would guess from Python. Bill Jelen (00:48:15): We would have to assume it is. The Excel team was excited because they said it makes Excel Turing T-U-R-I-N-G complete, whatever that is now means that you can do everything. Every calculation in Excel, thanks to Lambda functions. Rob Collie (00:48:28): You can now mathematically provably. If something's Turing complete, you could use Excel as the software language to build literally anything. If you wanted to build windows, the windows operating system, you could build it in Excel now, it wouldn't be a good choice but you mathematically could. Bill Jelen (00:48:46): So I was wondering why they called it This. Rob Collie (00:48:49): Tom is lighting us up. Thomas LaRock (00:48:50): So, I want-[crosstalk 00:23:51] Rob Collie (00:48:53): [inaudible 00:48:53]...Church. Thomas LaRock (00:48:54): I'm finding a lot of overlap in this stuff I'm doing with Python. Python uses arrays in data frames, things that we would just consider to be tables. A lot of Python is simply being data janitor. You're going to get all these different data sources and how the data is and you're going to bring it in and you're going to sanitize it for whatever purpose you're going to have down the line, which could be data science, or it could just be building a pipeline. So it outputs in CSV, which then ends up in somebody's Power BI. Thomas LaRock (00:49:28): Python is just this brilliant little thing, but I see so much overlap because when you guys talk about Power Query, well there's a whole part of Python, which is just scraping the web So instead of using Python to scrape the web what if I just put that into Excel and call it the Power Query and they just did all that stuff for you, you didn't have to even worry about coding, you just pick and choose, so I live in this world where I have enough knowledge of both tools to see the real power of using them together. I try to do some data science machine learning stuff, I'm not going to spend a whole lot of time trying to. Thomas LaRock (00:50:03): =stuff, right. It's just, I'm not going to spend a whole lot of time trying to clean my data using Python. I'm just going to go to Excel, make it all pretty, figure out from there and then bring it back in as a data frame, and that's going to save me a bunch of time. But anyway, I'm excited now, Lambda functions, this is interesting. Bill Jelen (00:50:16): The other really weird thing, just weird coincidence, about the Lambda functions is everyone knows the mathematician Turing. His professor was a guy named Alonzo Church and he was the inventor of Lambda calculus. And so the Lambda function is named after this guy, Alonzo Church, and Rob, Alonzo Church ended his career at Case Western University and probably, he died in 1995, he probably ate at that bonefish grill that you and I were at. Thomas LaRock (00:50:43): Probably. Yeah. Bill Jelen (00:50:44): I mean we were all within 10 miles of each other at that point. Rob Collie (00:50:48): You remember that old, it was a BBC history show called, Connections? Bill Jelen (00:50:53): No, I don't. Rob Collie (00:50:54): Oh, it was awesome. Bill Jelen (00:50:54): This would make a great episode, wouldn't it? Rob Collie (00:50:56): The host of this show would trace these weird coincidental chains and overlaps, like the thing we just talked about. And Wayne Winston was at Microsoft teaching matter. But then it wouldn't be Rob Collie that crossed paths with those people, it would be someone else famous. And then Thomas Edison happened to be brother-in-law to the bla-bla or whatever. Right? Bill Jelen (00:51:17): Yeah, right. Rob Collie (00:51:18): And sometimes it would span centuries and I think he would bring it back. He'd actually bring it full circle. He'd connect back to the original. It was insane. It was almost like, he had to be making this up. Bill Jelen (00:51:29): Well, so with your GETPIVOTDATA trick, you said that came from Microsoft finance. And I was in Australia, having dinner in Australia and a seminar, and this is one where the Excel team sent someone who was Carlos Otero, who was on the Excel team, but had worked in accounting at Microsoft. I said, Hey Carl, so I've heard this story about Rob, did you guys use GETPIVOTDATA this way? He's oh, we do that all the time. Yeah. So it was we are validation. You go, Rob, didn't just make that up. Rob Collie (00:51:54): Yeah. Bill still not trusting me all those years later, despite our handshake agreement. He's like I'm still going to fact check him. Bill Jelen (00:52:02): Might just, it came up. All right. So Tom, I appreciate that you could say Python. Python scares the heck out of me, not because of the learning curve, but I think Rob, you call it the learning cliff. I've bought a couple of Python for Excel user type things, and it just seems so intimidating to get up that curve to do anything useful. Have you actually, what are the steps? I'm an Excel guy. I know Excel really well. Just what are the first three baby steps I have to do? So, that way I can say I've done something in Python. Thomas LaRock (00:52:31): So let me see if I can get this right. I think around 2016 or so, is when I said to myself, I really, I want to make an effort to pivot slightly in my career as a data professional and the pivot I want to go to is data science. So I have a background in math. I have a master's in math and a lot of what I see in data science was stuff that I was doing in school many years ago, with paper and pencil, right. It's just stats. It's a T value Z value. I'm like yeah, I remember this stuff. So, back in the day, I would've, what we call data science, back in the day was getting a job with a big six as an actuary. That's essentially what our data scientists of today are doing, except there's other cool things, because you can apply it to lots of different projects. Thomas LaRock (00:53:15): So Microsoft and edX had this online courses that you could take. And I blogged about how my journey, where I basically sucked up as many data science certifications I could, through that edX, and a lot of it was at the time heavily reinforced with Azure products. So I got to learn a lot about Azure Machine Learning studio and things of that nature. And there's all sorts of little things in there. For this course, we're going to use Python for this, and they walk you through the one little thing that they need you to learn on that day. Right? So I think I finished in 2017 or 18 or something like that. And since then I've been just trying to apply it. Because how else are you going to learn, except by actually doing it. So I'm all right, so what can I do? So there's some websites there's Kaggle, which has data science competitions. Thomas LaRock (00:54:05): For example, there's a competition, I think the TSA put some money up where they said, when we put your stuff through the x-ray help us build a better algorithm for detecting objects and things like that. And first prize is $500,000 and I wrote about, and I go, so let me know the next time you make $500,000, tuning a sequel query. Because I'm going to be over here doing this stuff, cause that sounds awesome. So what I did next was I started getting into DataCamp and their online Python courses. They're easily digestible for me and I've been consuming them as fast as I possibly can. Now your original question was, what were the three things? And I've given this long winded answer of how I got involved. How I've gotten to this point, is all of those classes that I've been doing is simply reinforcing similar concepts over and over and over again, but by different instructors. Thomas LaRock (00:54:57): So I'm learning it in a slightly different way. So somebody says, hey we're going to do data visualizations today. That might be important to you, but probably not to Mr. Excel. But it might be important to somebody and you end up with a similar class, taught six different ways and it just gets reinforced. Because the concepts and the terms become more and more familiar, but they're showing you a slightly different way to do it. So for example, Lambda is really just a short way to do a recursion function or any function. You don't have to do it that way. You could write a whole function if you want and call it from anywhere. But now you get into this thing of, yeah but Lambda functions are actually more efficient in terms of processing and memory. Bill Jelen (00:55:39): Right. Thomas LaRock (00:55:40): And so now you start getting, now there's a whole course on, hey you want to tune your Python code? Let me show you this stuff. So for you, Mr. Excel, I would tell you that the first thing I think you should maybe, given all your experiences, you should get into understanding the difference between a data frame and an array, when you would want one versus the other, how are they useful. The other things I would tell you is they do have pivot table. You can pivot your data. I think for you, it would end up being so clunky compared to what you do in Excel, but you would understand the concepts of it. And then I guess, really the question is, what would you want to learn Python for? Why do you think Python is something you would need? For me, it was all about the data science. Bill Jelen (00:56:26): I need to be able to walk into a bar with a laptop, put the laptop on the bar, get five people around me and do something in Python and have them all go, that's what I want. Thomas LaRock (00:56:36): The thing with Excel is that it's visual. Python isn't going to have that same visualization. But I do think that there's a Python studio to help with that. Bill Jelen (00:56:44): Okay. Python studio. Thomas LaRock (00:56:45): I don't know if that, that's not what it's called, but I think there's some way for you to have that visualization. Otherwise it's just going to be code. Rob Collie (00:56:52): I've Got a shortcut. Ready? Probably know what's coming. It's going to be cynical. Don't learn it. Bill Jelen (00:56:59): Oh, come on. All the cool kids, all the cool kids are learning it. Rob Collie (00:57:02): I know. I know. And yet with the tools you've got, you can do a tremendous amount of the actual valuable things that you would do with Python. And you can do things that you really couldn't really particularly do very well with Python either. Here's a theory of mine. This is more factual, I think, than the previous opinion where I said, don't learn it, right? There's something real here. Just a question to what extent you agree. So there are certain kinds of tools. There are tools that are sort of purpose built and still flexible. And this is funny, a little bit humane, not a lot humane, but a little bit humane. In the same way that Mac, Apple will approach a problem, a consumer problem and solve it in a way that is easier for people to learn and people will kind of want to learn it, at all ages. Rob Collie (00:57:55): And then Microsoft will walk up to the same problem, for example, signing you in, and end up with a sign in process that is itself super complicated and requires you to sort of anticipate what the machine is trying to do in order to navigate your way around it. Apple would never do that. Now in that same way, Microsoft has a gift of approaching, as a culture, of approaching complicated business systems, complicated business products. And over time anyway, making them approachable to sort of this lower end developer who doesn't think of themself as a developer. The success of Excel under the Microsoft banner is the single greatest example of this. Everyone's slinging formulas in Excel or writing VBA, but even just writing formulas is a programmer, is a developer, but they don't know it. And a general purpose language like Python, is built for and by the computer science set, that was my background, right? Rob Collie (00:58:58): I'm a computer scientist, right? And so I think a tool like that, because it's meant to be everything, any programming language that allows you to draw visualizations and clean data, is so general purpose that it's going to have one of these steep learning cliffs. And it's going to be a cool kid thing, because everybody that's part of that club, worked hard to get into it, and then kind of lords it over everybody else. It's just sort of how people are. I think that something like Python is something that most people are either going to learn at the very beginning of their career. If you're not a computer scientist, right? If you're a computer scientist, you're going to eat this kind of stuff up every time it comes along. A new programming language being unleashed on the world is like a Christmas present to you. You're like oh my God, I can't wait to try out this new programming language because it gives me some sort of, scratches some itch for me to do these sorts of things. Rob Collie (00:59:48): Most of humanity isn't like that. If you're not a computer scientist, you might learn Python at the beginning of your career. Almost like when you're still in college, when you're young and you still have time to do that because the rest of the job hasn't consumed you yet. I just don't think Python has that, I can adopt it, 15 years into my career, sort of characteristics to it. In the same way that say something like Power Query does or Power Pivot or the DAX data model. What you do of course is not really the point. The joke that I made about Bill, don't go do this. It isn't really about you at all. I don't have anything against Python and I certainly don't have anything against you going and learning it. It's more for the rest of the audience, that's listening. If you're sitting here going, what do I do with my career right now? And you're kind of in the Excel space, the first, second, third, fourth, and fifth things you should be doing, are getting better at the things that Microsoft is throwing at you and then hiding. You don't need to have this FOMO, dear audience, about Python. I'm not saying you shouldn't learn it. I'm not saying that it's not a good thing to learn, but when we consider that everything's prioritized, if you haven't already, kind of, quote unquote mastered or gotten competent at Power Query, DAX and Data modeling, that's your next frontier. And you're going to be just ridiculously effective as you climb that curve. And without having to run smack forehead first into the Python cliff. Thomas LaRock (01:01:15): So Rob, I'm not going to disagree entirely because I actually think you are right about almost all of that, except you could just reversed it and talk glowingly about the need to do Python and not worry about Excel. Bill Jelen (01:01:26): But I wouldn't have. Thomas LaRock (01:01:28): You could have. Bill Jelen (01:01:29): I wouldn't have Thomas LaRock (01:01:31): But bill, if you're looking for something cool, then what you're going to want. You have a unique opportunity, I feel, because your depth of experience with Excel, that you can kind of blend the two together. If you're for example, you could use Python to scrape a website and then say, by the way, this is also available in Excel as Power Query. And here's the difference between the two. Somebody might be impressed by the ease at which you can point Python at some URL and just scrape everything necessary and get data from a Wikipedia article. That's one thing. Another big thing though, is sentiment analysis, right? You, you could go say hook into Twitter's API, pull back a whole bunch. You were talking about a work cloud earlier. You could pull back a whole bunch of stuff and say what's the current sentiment for my book, or something like that. Thomas LaRock (01:02:21): But that's something you're not going to really do in Excel, but that's going to be something that also has value for another person out there. They're going to be like, you know what, marketing specifically, oh I can get sentiment analysis fairly quickly with a few lines of Python code. That's actually kind of interesting. There's always going to be some type of unique application. Like Rob said, it's general purpose, Python, right? And there's so many aspects of it, but you should be able to find something, that I would say, you can bring together both worlds because you could do the sentiment analysis and then build a power BI dashboard on top of it. Bill Jelen (01:03:02): Yeah. Right. Thomas LaRock (01:03:03): One thing my friends had done over at Microsoft was sentiment analysis on movies, specifically Marvel versus DC. And what was funny is you found out through the sentiment, you could kind of use that to predict the amount of revenue a movie might make on opening weekend. Thomas LaRock (01:03:23): And what was funny was if you had movies that had, if there was much more positive sentiment, it was going to make a lot more money. So I'd be there, they'd be actively collecting data and then I would be on Twitter, tweeting something negative, slightly, about the upcoming movies. And then you found the movie didn't have enough revenue. So I would then tell Disney and others, that for a low fee, I could tweet much more positive things, so you could make more money. Clearly there's a correlation between these two events. My positive tweets or my negative tweets in your bottom line. Now I'm not saying I'm holding it for ransom. I'm just saying, it's a thing that you can show other people, which is kind of cool. Rob Collie (01:04:04): You're not holding it for ransom. You're just offering it for sale. It's different. So I thought what you were saying now, I was interpreting it as different, which was. I was hoping you were saying, I guess, that it was doing sentiment analysis on the script of the movie. Thomas LaRock (01:04:20): You could do that too. Rob Collie (01:04:21): And Michael Salvino, one of our previous guests, he loves analytics in sports, but at the same time he thinks analytics are ruining sports, because it's driving everyone towards this single optimal strategy. And a lot of the interest values being squeezed out of sports, like famously in baseball, it's all strikeouts and home runs. No one cares about anything else anymore and no one's optimizing for it anymore. So there's a lot of the most interesting aspects of the game are gone. And here's where I would whisper, and baseball couldn't afford to lose the interesting things. Rob Collie (01:04:52): Cause it was already pretty dull. Can't resist, sorry. I have a tendency to say things that are sort of glib and short and dramatic and nasty, in some ways. Even though in my heart, there's something very different. Let me see if I can circle back. So there was a Twitter thread over the weekend that I wanted to Wade into, but decided I didn't, where it involved Chris Webb and Marco Russo and some other people talking about MDX versus DAX. If you don't know what MDX is, it's the precursor to Dax. It was the thing that was in the formula language, expression language of SQL server, analysis services in the old multidimensional format. And there's a lot of diplomatic or partisan team MDX, team DAX, team both sides. But it was all talking about the strengths and weaknesses of the languages in terms of their technical capabilities. Rob Collie (01:05:48): It was primarily focused on that. Or how tedious they were in certain ways to use. And I wanted to join in and say, well, folks, it isn't about MDX versus DAX. It's not about the technology one versus technology two, it's about their adaptability. So for example, I tried multiple times at Microsoft to learn MDX and I had access to all the people who built it, and I couldn't learn it or I wouldn't learn it. I just rejected it. It was awful. Just hated that thing. 15 minutes into the explanation of how to write an if. And I'm oh, I'm out. I am out, get me out. So that's me and MDX. And then me and DAX, all I do is write a book that sells millions and millions of dollars at retail about tax. That's a striking difference, right? Rob Collie (01:06:42): To me it was sort of a silly argument. Which one's better is MDX still, blah, blah, blah. No MDX is not going away. But in terms of a percentage of the population, it is going to round to zero in relevance, relative to DAX because of its adaptability. So, that's what I'm really saying. Yes, Python is like, hey it's turning complete. You can do anything with Python. If you learn Python, you in theory, don't have to learn anything else. Whereas with Microsoft, you've got to learn DAX to do this and you've got to learn Power Query to do that. And yet, each one of those things, Dax, for each particular purpose is both better than Python for that purpose and more learnable for that purpose. And so I think the Microsoft ecosystem of users, especially the Excel crowd, is sort of a demonstrated group that will incrementally learn sort of one new thing about their tool set every two weeks, and just sort of incrementally pile that up into the point where they're just monsters. Rob Collie (01:07:43): They didn't go and learn Python, at some intensive bootcamp, they've been learning these tools sort of in a drizzling rest, never sleeps sort of way, and years later they're gods. And so I'm really just speaking sort of my own ecosystem, that the people that I know. If you are one of the computer science types or one of the types that's looking to completely change careers and things like that, or you just can't help it. And you're just curious about this other tool, not going to make fun of you. I'm going to salute you for going and learning this thing. I'm just saying that most people aren't like that. And there's a lot more people out there who will learn these other tools like Dax and M and that's all I have to say about that. Bill Jelen (01:08:27): So From the Excel point of view, when you look at the DAX formula language and the Excel formula language, there's so much overlap, that an Excel person can easily just start doing simple things in DAX, and they'll feel like it's their home, they already know 80%. And then you add the good stuff on top. That gives you super power. Thomas LaRock (01:08:46): Bill, have you played with DAX Studio. Bill Jelen (01:08:47): No. Thomas LaRock (01:08:48): Okay. Forget Python, go play with DAX studio. Bill Jelen (01:08:51): Okay. There we are. Thomas LaRock (01:08:52): I'm serious. Bill Jelen (01:08:53): I'm writing it down. Rob Collie (01:08:54): One of the number one things, the thing you just said, just triggered it for me. One of the biggest disconnects between Excel formulas and DAX, is the inability to step through it. The inability to evaluate step by step. And Excel sometimes you do this just by splitting your formula into multiple columns. You sort of see intermediate steps. So you can sort of see where the error happened. Where did things go wrong? But you also have the, evaluate formula step through, feature. And so much of DAX happens in places that's invisible. Rob Collie (01:09:25): But DAX Studio allows you to do things, like that filter function that I wrote in the middle of my formula. I can evaluate just that filter function, forget the rest of the formula and just see what the table it looks like that's hidden under the hood in computer memory. And I never see it in my measure formula, but I can see it explode out under the screen. It's not quite the same as a step through. It's not quite as good as that, but Bill, I think, you spend a week with that or less suddenly you'd come back to me and you would be running circles around me in DAX. Bill Jelen (01:10:01): That's hard to believe, but I'm definitely going to try it. Rob Collie (01:10:04): Apparently. I really have it in for Python. Thomas LaRock (01:10:07): You have a thing for Python. Rob Collie (01:10:09): And R we could, we could substitute R in here and I'd be saying the same things about it. Thomas LaRock (01:10:14): R to me, is just very focused data science. And it does that very well. Python, I think is just more extensible. Python's more widely adopted. You find it has more modules. So that's why I went Pythons instead of R. This all actually really started when R got brought into SQL server. That's when I was all right, this is getting a little more serious. I should start learning a little bit more about this. And then it was Mark Cuban who tweeted one night, that he was doing some Python for some analysis on basketball and stuff like that. And I, why would you choose Python over R? And a bunch of people, not Mark replied, helped me understand a little bit about the difference. And so that's when I made the pivot to Python. Bill Jelen (01:10:58): Wow. Right there. I mean, Python was on the list. R was on the list and TikTok. Those are the big three that I [crosstalk 01:11:05] Thomas LaRock (01:11:07): Oh my God is Python and TikTok a thing. Let me check. Cause if it's not, I want a corner in that market. It has to be. Rob Collie (01:11:14): Yeah, right. Thomas LaRock (01:11:15): Bill says, there's someone doing Excel, TikToks. Bill Jelen (01:11:18): Yeah. 15 seconds or less. And it's impressive. What you can learn in 15 seconds and a TikTok Excel tutorial, two variable data tables. Yeah. Thomas LaRock (01:11:28): This is known as, not Rob's medium. I'm still in the process of saying hello and time's up. Bill Jelen (01:11:38): It's a challenge. It's a challenge. How to teach something in 15 seconds. Thomas LaRock (01:11:41): I'm going to create a new platform called TikTok, TikTok, TikTok. It's this 45 second video. Bill Jelen (01:11:47): Gives you 45 Seconds. Thomas LaRock (01:11:52): And even then it won't be enough. That's just the introduction to the lesson. Today we're going to learn this. If you want to see the rest of it, you go to YouTube. So is it a thing, Tom? Rob Collie (01:12:06): Yeah, there are people. I'm finding there's a lot of Python coding videos. Thomas LaRock (01:12:11): On TikTok? Rob Collie (01:12:12): Yeah. On TikTok. Python itself seems to be about actual snakes in addition to some coding. And now I'm more enthralled by these snake videos. Thomas LaRock (01:12:23): So while we're talking about snakes, let's talk about a statistical concept known as risk. I was reading this on Reddit one day and it's just stuck with me ever since. It's just one of the funniest things. We do it all the time. We talk about this all the time at our house in various forms. They're talking about people who have poisonous snakes as pets. You know how Reddit comment threads go. They're very conversational. It sounds like the same voice is talking to itself. These different people on the internet, but it's the same wise internet voice talking back and forth to itself. And talking about how, whenever you know someone, whenever you meet someone who handles poisonous snakes, they have poisonous snakes as pets. What do you hear? You always hear the same thing over and over again. Oh, it's not that dangerous. Thomas LaRock (01:13:03): As long as you know what you're doing, right. As long as you know what you're doing, it's perfectly safe. It's perfectly safe. The risks are vastly overstated. In fact, this person would go on to tell you, my father was a snake handler like this, my uncle was a snake handler with this. And then you go, oh yeah. Are they still doing it? Oh, no, no, they're dead. Well, how did your dad die? Snake bites. How did your uncle die? Snake bites. It's like human ability to properly estimate risk, maybe in a single trial. But underestimate it over a large number of trials. Thomas LaRock (01:13:46): It's kind of the birthday problem. Right? How many people did you have to have in the room, before it's likely that two people share a birthday. It's ridiculously small, right? It's 24 or 30 or something like that. And the chances of two people having a birthday that are the same, are nothing, like oh I'll run that risk, a fraction of a percent. But you keep up in the trial. It's not the same problem. It's not the same as the birthday problem. But the point is, is that over thousands of trials, that fraction of a percent, if you did the math on, are you going to survive thousands of trials? It turns out that no, you are not. Rob Collie (01:14:21): Well, that was always the thing with the data center, right? If there's a 0.01% chance of a disc failure. So that's one in 10,000, but I've got a data center with 10,000 discs, you know one of them just failed. You just, yeah. That's how it works. Thomas LaRock (01:14:37): Yeah. And it's 0.01% over a particular period of time. And so in a six month period, I mean a hard drive's only got a one in 10,000 chance of failing, but we run those discs for five years and we've got banks of banks and banks and banks of them. Yeah, exactly. We're having one every second now I don't know what it was that we were talking about that led to that. Oh, Python. The snake= Rob Collie (01:15:03): Yeah. I don't know what it was that we were talking about that led to that. Oh, python, the snake handly. Bill Jelen (01:15:04): It's snakes, right? The TikTok. Rob Collie (01:15:06): I mean, pythons. Yeah. They've got a lower fatality rate than cobras. Thomas LaRock (01:15:10): Aren't pythons an issue in Florida right now? Don't they have to go hunt them? Rob Collie (01:15:14): Yes. Bill Jelen (01:15:15): Yeah. And I think there's a reward if you get them, I think. Rob Collie (01:15:19): Yeah. And it's gotten to the point where they've realized, of course, that only the females really matter to the reproductive capacity of the species males, biologically speaking, ecologically speaking, males aren't worth very much, right? So they've gotten to the point where if they catch a male, they don't kill it. They radio collar it so that it leads them to the females. Bill Jelen (01:15:42): Wow. Rob Collie (01:15:44): Right? A male, a live male is actually more valuable as a weapon against the population than they are a contributor to the future of the population. Killing a male does nothing. Thomas LaRock (01:15:57): I'm surprised they just haven't reversed engineered their DNA. And inject them with something that they take back to the nest and just wipes everything out. Rob Collie (01:16:05): Well, and then we get movies like I Am Legend where this engineered thing hops species and just... Thomas LaRock (01:16:17): But what's... That's a 0.01% chance? Rob Collie (01:16:20): Oh, I know. I know. Yeah. It's perfectly safe if you know what you're doing. Thomas LaRock (01:16:27): It's like a one in a billion chance that it'll affect a human. Rob Collie (01:16:31): And so, because it's so safe, we're going to generate all these kits that you can do at home. You can engineer your own virus at home. I'm sure it'll be fine. Thomas LaRock (01:16:39): What's the worst that could happen? Bill Jelen (01:16:42): What's the worst that could happen. Rob Collie (01:16:45): Only bet as much as you can afford to win. Here's something that's interesting. So Bill, you've diversified. When the phone rings and it's you, every now and then my iPhone decides to put under your name, Bill Jelen, to put under your name, the words "We Report Space." Bill Jelen (01:17:03): That's hilarious. Yeah. Right. Rob Collie (01:17:05): I don't think of you as We Report Space, I think of you as MrExcel. Bill Jelen (01:17:09): MrExcel. Rob Collie (01:17:09): And holding macro books. That's your alter ego as far as I'm concerned. And then there's Bill Jelen, but now you're also... Bill Jelen (01:17:18): We Report Space. Rob Collie (01:17:19): For a podcast about data, there's a strange near-100% Venn diagram overlap with people who are interested in rockets and Mars and things like that, so what is We Report Space? Are you out there running around, checking on the moon? News reports from the lunar surface? Bill Jelen (01:17:36): I'm going to blame this on Penn and Teller. I was in Las Vegas doing a seminar. You check in, they're like, "Well, can we book you any shows?" I was like, "Hey, Penn and Teller, let's go see Penn and Teller." And then afterwards, Penn and Teller, they talk to everyone in the audience and they're selling their book. And so I buy their book and on a flight home, it says, "You need to go see a space shuttle launch at Kennedy Space Center once in your life." All right. An STS-135, the very last launch. We came down from Ohio to Florida to see that launch and we got there four hours early and we're sitting there just talking with other people in their lawn chairs who were lined up to watch this launch. And some guy next to us says that his friend right now is at the press site, three miles from the launch pad, thanks to some lottery he won on Twitter. I'm like, "What?" Yeah, he says, "Yeah, there's a lottery on Twitter." All right. Bill Jelen (01:18:23): And so two years later I was just cruising through Twitter and there's a lottery to go see a launch. I'm like, "Hey, yeah, I'm going to sign up for that." It was an early SpaceX launch and they scrubbed, which means they didn't launch. And then, 30 people won this lottery and they showed up to watch this. And then they scrubbed. And then a month later, 20 of them came back, and it scrubbed. And then there was another launch. And by the time they finally launched on their fourth opportunity, there were only seven of us that made it to each one. And three of those seven formed a group that is now We Report Space, right? So I'm here 12 miles away from the launch pads. Bill Jelen (01:19:01): We actually started working, the three of us started working for another outlet, right? Another website. And the three of us quickly found that our attitude towards life is "we don't give a shit." That's our motto. It's just... And the guy that we were working for was very angry. He was a very angry guy and the three of us all run our own companies and we're like, "We don't really appreciate this, he's not resonating with the 'we don't give a shit.'" So we finally just decided to leave and start our own outlet where our overriding credo is "we don't give a shit." Right? You can't make it? That's okay. Right? Or, it didn't happen? That's okay. There's other people that got to that photograph. Bill Jelen (01:19:44): But what's really fascinating, it's a techy thing, the ability to leave a camera too close to the rocket for a human to be, and to have that camera take pictures, and the camera's going to be out in the Florida environment for 24 hours or 48 hours. And to engineer this whole thing to work and take the photos right. And just get some amazing photos. It was just a blast and it's a techy thing to be able to get everything to work right. Again, a great group of people out there at the press site. People have been doing this for 30 years. Generally they all share their information. You say, "Ah, yeah, this didn't work." And they're like, "Oh, you should try this." So just a great group of people down here in Florida that cover those launches and are willing to accept new people showing up, I guess. And generally be cool. Thomas LaRock (01:20:32): I was going to say that, I guess you guys don't know this, I thought, Rob, you knew this, but I did win one of the NASA Tweetups. Bill Jelen (01:20:43): You were at NASA Social, weren't you? Thomas LaRock (01:20:46): I was at NASA Social. So yeah. NASA Tweetup is a thing. It's awesome. And now my buddy, who worked for... He did some contract work with Canon who does all the cameras around the launchpad, and he got a nice photo of me watching. It went up. And then I got to see the pictures that they had taken. But as you described, the weather, the acid that comes down from the rocket layer that just destroys these... So they have to be in these boxes and just the entire process of how you can take, like Bill said, a photograph where no human can live. Rob Collie (01:21:18): And they have to be, those cameras have to be python proof? Bill Jelen (01:21:21): Oh sure. Thomas LaRock (01:21:23): The gators by the launchpad, the gators are a thing. Bill Jelen (01:21:26): And the army... Rob Collie (01:21:27): Don't worry, the python's reading those. Bill Jelen (01:21:29): And the armadillos. And it's why... And Tom, I did know that, I knew that about you, that you had been to, the NASA Tweetup is now NASA Social, that you had been to one. I don't know, maybe Rob told me that or something, somehow I connected that dot. Thomas LaRock (01:21:44): That's awesome that you're... I didn't know you were down there doing work with them. I'm fascinated. I need to learn more now. Bill Jelen (01:21:49): When we say work- Rob Collie (01:21:49): You can find Bill's work at his website, we don't give a shit about space dot com. Bill Jelen (01:21:56): We Report Space dot com, yeah. Rob Collie (01:21:58): All right. So let's just recap for a moment. You started in Lotus 1-2-3 as the export to Excel BI guy, which I'm still going to call export to Excel, even though it was Lotus 1-2-3 at the time. Bill Jelen (01:22:09): Of course. Rob Collie (01:22:09): And then that kind of beautifully metastasized into the entire MrExcel empire that we know today. But along the way, I think we missed something in there. Something really, really unique and important. So I would definitely want to circle back. The first time I had to conduct any business with you that involved money changing hands, I had to make out a check to Tickling Keys. Bill Jelen (01:22:35): Tickling Keys, Inc. That's right. Rob Collie (01:22:36): Tickling Keys, Inc. I expected it to be MrExcel or Bill Jelen, or maybe even Holy Macro Books, but no, it was Tickling Keys. And for a while I was just like, "Oh, that's so neat. He's talking about, in that name, he's talking about himself writing his book. He's just sitting there just tickling the keys and just brilliant Excel is flowing onto the page as he tickles those keys." I brought up to you one time after a year of believing that, and the reality behind that name is way cooler than what I'd imagined. So what's the origin of the name, Tickling Keys? Bill Jelen (01:23:14): Tickling Keys, we didn't start out as an Excel consultancy, or even as an Excel website. I grew up, in our basement there was a player piano from 1920, you know the old player pianos, the things that play themselves? You put the paper roll in there and it holes in the roll, press the keys down. As it goes over the tracker bar using some sort of mechanical vacuum system, you're pumping that whole thing, right? Rob Collie (01:23:36): I've never seen one in person, but I know what you're talking about. Bill Jelen (01:23:39): They were cool, but the problem is all the songs were from the twenties. And when I became a teenager, I was more into rock and roll. And there was no one doing Led Zeppelin on player piano roll. And that just seemed like that was something that had to be corrected. Rob Collie (01:23:53): Terrible tragedy, yeah. Bill Jelen (01:23:53): Yes, right. So I would arrange the roll in Excel. I would get the sheet music, I would build an Excel file that had all of the data that needed to be punched in the role. And then I found a guy down in Richardson, Texas, that you would send him a data file and he had a machine that would actually punch the holes in the roll. But the deal was you had to get five, right. And what are you going to do with five copies of Stairway to Heaven? So I actually started a company that licensed that music and then would sell the extra roll. So I'd make one for myself and then sell the other four. And that company is Tickling Keys. And when I opened the website, this tiny little website, that was going to be the offshoot of the player piano, rock and roll player piano company, that's when... I didn't bother to reform another corporation. I was like, "Okay, this is the new division of Tickling Keys. People think it's a computer and that's fine. It's going to work fine." Rob Collie (01:24:52): And then that would've been the thing, right? You'd just be a billionaire today on your player piano rock and roll empire. But if it weren't for Napster, right? Stinking Napster, I mean, the teens all pirating your Tickling Keys player piano music. Kids in dorm rooms, they brought you down, man. So you also told me that, didn't you give one or get it signed by Bruce? Bill Jelen (01:25:24): I haven't gotten Bruce yet, I've gotten a lot of other people in the E Street Band. The big one was Meat Loaf. So I had a roll of Paradise by the Dashboard Light. And the very first time that I ever got to show one of these rolls to someone who was connected to the roll was: Meat Loaf was in town for a concert at Blossom Music Center up in Cleveland, but he did a charity softball game down in Akron, right. And after the game, I went up to the bus and I had two copies of the roll. And I showed it to him and he knew what it was. Bill Jelen (01:25:50): He knew what it was, because he even said, "Oh, this is great. I'm going to fire the band, we'll just put a player piano on stage." But he autographed it. And then he yells back to me, I'm walking back to the car, he yells back to me. He says, "Hey! Hey, it's Meat Loaf, two words. Look at the album." And on the player piano roll, I put meatloaf as a single word and somehow, I gave him one copy and someone on the bus said, "Hey, you spelled it wrong." Rob Collie (01:26:13): Wow. So you said he knew what it was immediately. So you're saying that he recognized it was a player piano roll. He didn't look at it and go, "Oh my God, Paradise by the Dashboard Light!" Bill Jelen (01:26:26): And you say, I think this is the world's only existing copy of a Meat Loaf tune on player piano. And that gets you in. I was at a dinner with a lot of the E Street Band about three tables away from me. And they had decided they weren't going to sign anything. But when I had a player piano roll, I said, this is if you want a player piano roll of a Bruce song. I got Roy Bittan, who's the piano player for Bruce, to sign that. And Max Weinberg, who's the drummer, he instantly knew. He says, "Ah, yeah, you're just X number of years too late." He knew it was 1922. They were really in their heyday. So it's kind of an interesting conversation piece and I've gotten a lot of, under the pretext of, "Hey, will you sign this player piano roll?" I've gotten to meet Mark Cohn from Walking in Memphis and Bruce Hornsby, a lot of the E Street Band, Nils Lofgren, who was with Neil Young, autographed Our House. Rob Collie (01:27:13): That's so cool. And what do these spreadsheets look like? I need to see one, I kind of envisioned originally that it was a bunch of columns formatted to the same width and you're coloring in the squares to make the dots. Bill Jelen (01:27:25): That would've been awesome. Rob Collie (01:27:27): But, it's not like that is it? Bill Jelen (01:27:28): So it was measure number, then beat within the measure, which note? So on a player piano, I think middle C is the 40th note or 44th note. I had kind of in my head that the middle C is this number and you type in the number and then how long it was held down for. And then I probably, I don't know, was it a pivot table or some really cool formulas to take that data and stretch it into whatever, then export from Excel to whatever binary format that the guy in Texas needed. Rob Collie (01:27:58): Yeah I can't stop thinking about this. And then if there's a rest in the music where no notes are being played, do you have blank rows in your spreadsheet, or? Bill Jelen (01:28:11): No, no. I think if you simply don't send instruction to... So it's a little hole punch, right? So you're telling it when to turn on and when to turn off. So if nothing's happening, I think that you just don't send anything at that time code. Rob Collie (01:28:24): I see. Okay. So it's each row in your spreadsheet is time coded as to when it happens. Bill Jelen (01:28:28): Exactly. Rob Collie (01:28:29): Okay. That makes sense. I'm really excited as kind of as a parting shot, as someone who, my profession, even though I use Excel certainly for lots of one off purposes in my day to day business life, professionally, we're in the BI space. And so Power BI is, that's our thing. We don't use Excel for BI purposes as a company. I think LAMBDA functions though, it's one of those unexplored frontiers, in a way, in terms of what it can do. This is one of those things where I'm actually wishing that I had a bit more time to just go screw around with a technology, and that's Excel LAMBDA functions. And I have a bit of a background with it. Rob Collie (01:29:12): I worked on an early sort of specification for this feature many years ago, and it wasn't my idea at the time. So it's not like I'm saying, "Oh, they did my thing." No, it was someone else's idea back then. And it's the same idea has been sticking around for a long time, it's just really hard to implement. But if you think about for a moment, a spreadsheet, like a single sheet, that does a lot of calculations, like an income statement type of forecast, it's got columns for month one, month two, month 24. And it's got a handful of variables that go in, like what's our attrition rate of losing customers? What's our rate of new customers per month? How much are we going to spend on marketing? How many employees are we going to add per month? Whatever, right? Rob Collie (01:29:58): And from these handful of variables, it generates a tremendous amount of intermediate data month to month and it ripples and ripples and ripples along to come up with a handful of outputs. Like, this is how much money this company would be worth if we executed on this plan. What's this internal rate of return? What's its multiple that we could sell it for? How many months until we're cash flow positive? How much money, what's the maximum spend, meaning how much funding do we need? There's a lot of final answers that come out of a spreadsheet like this that you can only to by generating hundreds, if not thousands, of intermediate cells that ultimately then condense down to these numbers, right? Okay. Now, if you want to play with different scenarios, like different sets of inputs, this sucks. What are you going to do? Make a million copies of this sheet and change the variables, or you're going to write some macro that changes the variables and then records the answers on a separate sheet and then goes and changes the inputs again and iterates through like that. Rob Collie (01:31:03): But with LAMBDA functions, as I understand it, if it's the same thing that I was originally, like I was a long time ago, part of planning something along these lines, you could take that whole worksheet that I just described and turn it into a new function, like equal my income forecaster function. And it takes four inputs and it generates three or four outputs. And now you just create a table of all your different inputs, all your different input scenarios. And then you've got a number of calculated columns sitting next to it that are just calls to this function, this custom function of yours and which are the outputs you want. And you just drag fill that thing down and it uses your spreadsheet, this giant spreadsheet on the other sheet as a function. That's intense. And I think that it's one of those things where until you've played with it a bit, you don't really understand its potential. It so reinvents some of the fundamental rules of spreadsheeting that you're going to find yourself able to perform a type of financial modeling analysis, which by the way, Power BI is not really the right tool for a lot of things like this. The type of scenario I just described is a very difficult thing to do in Power BI. Not that you can't do financial modeling and forecast in Power BI, but that particular kind of very iterative ripple forward through time calculation is hard to do in DAX. And so this might be something that the world has never seen in terms of what it's capable of doing. And it's going to end up creating types of business, believe it or not, that weren't possible. Bill Jelen (01:32:40): I think the power here is, give this to a million people who are in a hundred thousand different industries and just watch what happens. Rob Collie (01:32:46): Mm-hmm (affirmative). That's what I'm talking about. Bill Jelen (01:32:47): So people will come up with uses that no one ever imagined. Rob Collie (01:32:51): I'm really excited by that. And it's not often that something comes along where you can say that, think about the accessibility of this. Someone who can build one of those income statement type spreadsheets now has the power to do something exponentially more sophisticated and it's not hard, and the Python person could have done it all along, but it'd probably take them longer, even though they were good at Python. But most importantly, there are fewer Python people and they're not deeply embedded in the business in the way the Excel people are. So this is why we've got sort of this really fertile breeding ground for something, I don't know. I'm really eyebrow raised that this one's come to life. I had no idea it was coming. I'm not one of the Excel MVP insiders anymore so when it landed, I was just like, "Holy cow. Oh wow." I need to go make some time just to screw around with it. Bill Jelen (01:33:43): No, it's cool. The other one is the LET function, LET. L-E-T. Which is, Rob, when you talk about those sub columns, you build your formula and the tiny little columns, the sub columns, and then the final answer out there is the eighth column or whatever. When you start combining that all back into one formula, you find that you're using some of those intermediate results over and over and over again. And so your formula becomes 800 characters because you have to put the LEN of the TRIM of A1 over and over and over. The LET function just, you define that once and assign it into a variable, slap it into a variable right inside the formula, and then can reuse that again. Right and so, yeah, both of those together. Rob Collie (01:34:22): That's a lot like DAX variables, which they didn't have in the beginning. So you can perform that one calculation, in a way if you'd had LET, you wouldn't have needed IFERROR. But you still would like IFERROR, because it's even simpler. Yeah. There's so many things, there's so many places, even now when I'm writing an Excel formula where I'm like, "If this calculation less than zero, then zero. Otherwise-" Bill Jelen (01:34:45): The calculation. So now you're doing the calculation twice. Rob Collie (01:34:50): I just want to scream every time I encounter that, I just want a function that truncates it, floors it at zero. I do that all the time. Bill Jelen (01:34:59): So how do you do that in DAX? Is there a DAX solution to... Rob Collie (01:35:02): Instead of LET, it's VAR, like measure name equals, and then you say VAR, then you make a name, like My Variable equals, here's the calculation, and you can have a bunch of variables and you can say, VAR My Variable 2 equals, my tax rate equals this, right? Then you say, return, you type return. Then you write your regular DAX formula after that but you can plug in the variables above. Bill Jelen (01:35:28): Nice. Yeah. Same thing. That's the LET function right there. Rob Collie (01:35:30): It's sometimes a little funny in DAX, and by the way, you can even use variables as tables. You can have a filter function as a variable, which is pretty hot. However, you've got to be careful because these variables evaluate the raw filter context of the measure or whatever. And then if you use the variable deeper down in your function, in your formula, in a place where it would've had a modified filter context, it won't pick up the modified filter context because it's already pre-evaluated to whatever it was at the very beginning. So you got to be a little careful with it sometimes, but most of the time it's awesome. Bill Jelen (01:36:03): That's cool, that's cool. And when I download DAX Studio, will I... Rob Collie (01:36:08): Oh yeah. You'll be all over this. Yeah. You'll get them. Well, folks. Thomas LaRock (01:36:12): Oh my God. Is there a DAX TikTok? Rob Collie (01:36:14): DAX Studio TikTok? DAX TikTok? Bill Jelen (01:36:17): DAX TikTok, that's it. Rob Collie (01:36:19): All right. I'm filing the copyright now, DAX TikTok. Thomas LaRock (01:36:24): You own that hashtag. Rob Collie (01:36:25): Yeah. I own it. Yeah. We'll just say we own the hat. Just the pound sign. Like the Austin Powers joke- Thomas LaRock (01:36:32): I invented the pound sign. Rob Collie (01:36:33): He invented the question mark. He was prone to making outlandish claims, like he invented the question mark. You know, I haven't been wanting to have you on, Bill, but Tom's been insisting, he's like, "When are you going to get Bill?" Bill Jelen (01:36:48): I think you mentioned this to me before you recorded your first one. But in my head I'm like, "Oh, I should wait a little bit. So that way the show will develop into something." Thomas LaRock (01:36:56): Yeah. We're still waiting. Rob Collie (01:37:00): Maybe this is when we arrive. Bill Jelen (01:37:02): I don't think so. I really don't think this was it. But you know what, I don't give a shit. So... Rob Collie (01:37:09): That's so good. Bill Jelen (01:37:14): Thanks guys. Rob Collie (01:37:15): All right. Announcer (01:37:16): Thanks for listening to the Raw Data by P3 Podcast. Find out what the experts at P3 can do for your business. Go to Power Pivot Pro dot com. Interested in becoming a guest on the show? Email Luke P, L-U-K-E-P at Power Pivot Pro dot com. Have a data day!