Building Community Around Data Analytics

On this episode, Tom talks with Dean Stoecker, co-founder and executive chairman of Alteryx, a technology company enabling organizations to make faster, more confident, data-driven decisions. Together they discuss Alteryx’s unique approach to data analytics, how creating a social experience for its users accelerates proficiency, and why focusing on company culture has always been a top priority.

Transcript

Tom 01:16
Welcome, Dean.

Dean 01:17
Tom, thanks for having me.

Tom 01:19
Absolutely. So your path in building Alteryx is, in some ways, traditional, but in some ways, also very unique. And I, I know a little bit about your past from the standpoint that you grew up in an entrepreneurial family. Can you tell me a little bit about how that shaped who you became in your professional life and how it led into starting Alteryx?

Dean 01:41
Yeah, for sure. We definitely had a different path at Alteryx. I think that if people looked at the Silicon Valley paths, ours was completely different. Ours was a 20 year old overnight success. But it was planned that way. It was planned that way. I can remember, oh, by the way, all three founders grew up in entrepreneurial households as well. So that helped kind of bond the vision for as many years as it took to get to IPO. But I can recall conversations with my father, who was an entrepreneur, when I was six or seven years old at the dining room table. And, he would talk about how bad pay customers weren't helping out or how hedging lumber in tough times was difficult. Because he was a homebuilder. And I think it, it created a sense of discipline, that I don't think you, you don't probably don't get it at an entrepreneurial class, you probably don't get it in necessarily having two or three other startups that didn't make it, although I had some of those as well. But it kind of grounds you in remaining principled and sticking to your guns.

Tom 02:52
I have a picture of my father from the 1970s, jogging with a bunch of his workmates, and that picture is labeled ‘jogging before jogging was cool’. Right? You were in data before data was cool. How did it become, Alteryx become a data play before anyone really understood data? How did that happen?

Dean 03:11
Yeah, that's actually a great, great question. I was in data from my first job, right out of University of Colorado, I moved from Boulder to California, and my wife had a career working for her father's shoe company. And I didn't know what to do. I went on a bunch of interviews, and I had two offers. One was $32,000 a year selling cardboard boxes with an expense account and a car. And the other one was $12,500 a year selling data that nobody understood. And I took the $12,500 job only because I realized that I could help people understand the data. And I didn't want to go to my son's fourth grade class Dad's to work day talking about corrugated boxes. And so it started there and every job I had, before starting Alteryx, from 1980 to 1997, I worked for data companies, and their big ones: divisions of Nielsen, and Dun and Bradstreet. And I learned a very important lesson aside, from the entrepreneurial lessons of discipline from my father. I learned that the ultimate value in content is when it becomes ubiquitous, when everybody can play with it and touch it and ask questions and derive value from it. So I would work very hard at these companies trying to convince them that you had to wrap the data that they sold with a UI/UX experience that made it super easy for people to, in our parlance at Alteryx, drag and drop and click and run and ask any question of the data. And then you had to apply an analytic layer on top of it that would allow you to ask progressively harder questions, things that would initially maybe end up in a Tableau dashboard: descriptive analytics, to predictive modeling, to machine learning, to artificial intelligence, to in our case, when we first started, a specific discipline around geospatial analytics. And these companies didn't understand it. I think they intuitively understood it. But what they did to act upon it, they said, let's take $1 of investment and let's put 33 cents into data. 33 cents into software and 33 cents into analytics. And lo and behold, they became mediocre at everything. And so I finally said, Okay, I'm going to start a software company that's pure play. We're not going to own any content or gonna manipulate any kind of content, big, little structured unstructured data in the cloud on the ground, in spreadsheets. And we're going to provide an analytic process layer to it that allows you to ask any kind of question. And we accomplished it. It took nine years to build the platform. We went from services to products to platform. And then it took another almost eight years for the market to catch up to realize what we had built. So we were, we were ahead of the game for a very long time.

Tom 06:06
Absolutely you were, I also have to imagine living through those eras of technology as well, you must have been focused on the UI component and making it easy for the non technical end user to be able to use this thing, versus the model that was traditional for 20, 25, 30 years, which was let me go talk to that tech person sitting in the basement, trying to give me something I can use, you actually started from kind of a almost a citizen analytics perspective very early on didn't you?

Dean 06:39
We did and I would I often characterize our success, as as the epitome of what Clayton Christensen wrote about in Disruptive Innovation, is that you take an old practice, that's relegated to a few, and you democratize it to the masses. And we have seen this so many so many times. I mean, you know, what Netflix did the Blockbuster Video, what Apple in the camera did, much more so than the phone, what what the camera did to Polaroid and Kodak – we did the same thing. We just did it in the data science world where normally if you had a question in the line of business, you would have to go to the data science team. And you'd be required to have a Statement of Work Requirements Document, a technical specification, and six months later, you would get an answer to a question, you forgot what you asked. And so we said, we have to put the capabilities closer to the people who have the context. And that was through UI UX, and a self service analytic prowess that would allow you to march up the curve to, to where you go to over time. And we've seen this with many customers of ours, where people get promoted, because they're solving hard problems. And then they're asked to solve harder problems. And I would rather have, it's not that I don't believe in scientists, I'd rather have the scientists working on edge cases, you know, the autonomous vehicle that's still having troubles, and global warming, and homelessness and all that all the, for good things to, I think the hardest challenges should be relegated to the 2 million scientists and the 50 million citizen data scientists, we're enabling in incredible ways.

Tom 08:15
You know, I went back through some of the interviews that you've done over the last several years, and you've been saying this phrase, and I'm going to going to put out there and I want you to talk some more to it for our for our listeners, ‘data analysis is one of the skills that every human being has to have if they're going to survive this next generation’. You weren't just saying that 12 months ago, I've actually listened to podcast interviews now, where you were saying that four years ago, five years ago, before any of this hype and current era, why did you see that so early on? And then of course, it's influenced the way that you've built the platform?

Dean 08:47
I think if you look at the generations, the industrial age, and the agronomy age, and you go through these ages, and you see you see what happens, and it's pretty clear that data is the new oil, and it's probably one of the biggest assets that enterprises around the world have. And yet, it's, they say that only 3% of enterprise data gets accessed at all. And I read a report recently, where they said that there's $15 trillion of value locked up in enterprise data sources, That means the only thing that's missing is the skills to access it and play with it and manipulate it. And that's tough. Our current generation hasn't had those skills. The funny thing is, they're the ones that are creating a lot of this data, either through social platforms or what they're doing that they're working through NetSuite or Salesforce or ServiceNow, whatnot. When I grew up, the skill was learneing how to balance your checkbook. And while that served me well, without a data science and analytics skill set, I think the current generation is screwed. They're gonna be left behind and there's no reason for it. Other than selling out the human, a lot of a lot of executives at large enterprises around the world. I think it's changed a lot. We went public at 85 million, we'll do a billion ARR this year, making us the first billion dollar pure software company in Orange County and the 41st worldwide. But I think executives have said, no, we have scientists down the hall, or we have systems of record deep in IT. And that just doesn't work anymore. And there's plenty of facts around why you have to upskill the workforce. There's a bunch of studies that have been done around the half life of enterprises. You know, if you started a business back in the, you know, late 1800s, like Ford Motor Company, the half life was 100 years. They made it, they almost went under, and they're still struggling. But as time has progressed the half life has gotten shorter. Do you think for a minute that Sears Roebuck shouldn't have been Amazon, they owned all the customers in the US, all they had to do is turn a digital eye and approach their customers differently. And they got wiped out. Blockbuster had, at one point, 11,000 stores and now today there's one store, and it's the Blockbuster Museum in Bend, Oregon. And they could have even bought Netflix for $50 million, and said, no. And so my belief is that had those executives understood the value of democratizing analytics, to the everyday data worker, the ordinary person who's trying to do extraordinary things, they would have figured these things out, and they would have not suffered through the half life.

Tom 11:35
And so do you feel like, you know, to be competitive in the wind in the next 20 years of business, that democratization, the data literacy of winning organizations is going to be very, very high, as in 99.9% of the of the employee base?

Dean 11:50
It's, I think it's mandatory. And I think we have a digital maturity score that we assigned to to customers, we go on site, and we do these tests, surveys of sorts, to help the customers understand, one through five, what is their maturity level? How, how well, do they understand data as an asset? How accessible does that data become? The hardest part in democratization actually is the first mile, and that is knowing what question to ask. And so a lot of companies just don't have the skill or they thought they had the skill and, and I think they're retrenching, they're recognizing that, you know, especially with COVID, working from home, you've got to, you got to be able to do a lot more than what you were doing just in the office. And so I don't know if it's 99%. I think that certainly anyone who touches data or has KPIs or OKRs surrounding data, one of my surprises in our, in my 24 year journey of running the company is, and I did this intentionally, I'm not sure I fully understood the impact., but I didn't want to become a point solution, we could have clearly gone into pricing analytics or media optimization analytics, we avoided all those things. Because we believe that every functional area of every company worldwide would suffer from the same challenges of not having data literacy. And sure enough, it's true. Today Alteryx is used to hedge fuel and airlines, it's used to run airports in Dubai, it's used to do derivatives modeling and in Wall Street, genome sequencing and med tech. I've never been more surprised at the variety of use cases. And it's why we've got customers in I think 90, 95 countries today.

Tom 13:40
Again, data is ubiquitous, and it's ubiquitous. It touches every industry, right? But every industry is generating it at a different velocity and veracity. And so it is going to be unique. So the platform play, it makes sense for why you've had such a diverse base. I'd be remiss if I didn't ask you dean about, given that your one of 41 who’ve built a billion dollar software pureplay, I love asking leaders like yourself about the leadership journey you've been on, right? Obviously, it's really different running a really small startup up, to the $50 million business, to the $500 million business. Talk a little bit of how you thought about building the organization and the culture that you thought it would take to get to where you’ve gotten.

Dean 14:28
Great question. I often reflect on my undergraduate at University of Colorado, and my MBA program at Pepperdine. And in all of my 24 years running Alteryx, there's only one class that really stood out every year. And it was organizational design. And, you know, when you're three people, it's easy, because you can yell at each other over the wall. When you're 30, it's a little bit more challenging. When you're in two or three different locations, you get to 3000. And, you know, there's bifurcation of duties, and there's oversight teams, and then it's just a much more complex web of an organization. But I think the magic is there. And I worked very, very hard every day to try to get to try to build a culture that was enduring and sustaining. Because I knew at some point I would retire, but I wanted to make sure that that culture remained to be that, to me, that was my legacy there. And it's hard to do, it's just really hard to do. So we created a customer centric organization, and everyone devoted all of their attention not to internal happenings, but to external happenings. And it began there. And then I think you build out programs, from ESG programs to our Alteryx for Good program – things that  provide a purpose more than just serving data analysts around the world. They're doing good with analytics. Earnings call analysts on earnings calls, they would ask me, what's your biggest fear? And my fear is keeping our culture. I think that because we did it in an unconventional way, we had to have a culture that was more durable. And, it's interesting. I know there's ways to do it, Palmer Luckey did it in a different way. He sold Oculus to Meta for a billion dollars and had no revenue, or little revenue. And that certainly happens. But unfortunately, entrepreneurs get their minds twisted towards those journeys. And the vast majority of really successful companies, they're more like my journey. I actually wrote about it when I retired. I wrote a book called Masterpiece, The Emotional Journey to Creating Anything Great. And for me, culture was probably the thing that stood out the most frequently. We address culture at every staff meeting, we address culture, fact, I only really had three KPIs my entire career there. And I'll put them in this order: NPS from employees, that leads to NPS from customers, that leads to net revenue retention, for Wall Street. So culture was kind of at the bedrock of everything we did.

Tom 17:05
That's interesting, and not what you typically hear from people who've done what you've done. Before we go to, because I'm gonna get to Alteryx for Good, but I gotta make a stop in between. And it's really about the very specific effort to build a community. So internally we call it the culture, but externally, you really doubled down on building a community. And building a community around your users, getting them to connect, it's not a cost, right? It's an investment. Talk about where that mentality came from. Right? Because not every, certainly in the tech business where we tend to under appreciate relationships and think more about transactions, it's not typical. You double down on something that's atypical, where'd that come from?

Dean 17:47
Yeah, we did community very, very different, which is why we've won so many worldwide community awards. Alteryx, for the people who don't understand Alteryx or haven't seen Alteryx, it is a very sophisticated drag and drop, click and run code free for the analyst code friendly for the PhD. It has 300 standard tools in it. We call Building Blocks, and you can configure these tools in 300 factorial. So we're talking probably 17, 18, 20 billion combinations of how you could string these two things together. And because we were doing work in so many different industries and use cases, there was no way we were going to scale our support organization linearly, it would be a deathtrap. And so I knew that the customers would rally around the culture that we had built. So we went from our internal culture that looked out to customers, and we started the community in 2015, was the launch. And I think the key difference is that I insisted that the community report to me, and because again, a lot of it had to do with my two KPIs, NPs from employees and NPS from customers. And I also insisted that the community be embedded in the platform itself. Because why should you have to leave the product you love to get to the community you so desperately need? And, and so I got a lot of kickback from product management about that. But Surprise, I got my way. And that's all it took. So today, there's a half a million members in the community. It's ridiculous, we’ve gamified it, there's badges for almost every thing you do, we've reduced our trouble tickets or deflected our trouble tickets into the call center by 92%. 

Tom 19:44
Now addressed in the community?

Dean 19:45
They’re now addressed by other customers, who are more than willing. We've proven, if nothing else, we have proven that analytics is a social experience, that you can't do it alone. And this is part of the upskilling effort is that we go on site to large organizations around the world, and we'll have Datathon or hackathons or contests. And it's just amazing what the results are, people in marketing solving problems for the people in finance, and vice versa. And, and all of a sudden, you create this community of people who love data, we freed up their minds and got them out of complex v lookups in Excel. And I've had people twice in my career telling me they wanted, they loved our product so much, they wanted to name their baby Alteryx. SBut the most important part for he community was not just that we strengthen the community and that we reduced trouble tickets by 92%, but we found out through a bunch of analysis, as we're an analytics company, that the customers who had their associates involved in community, they expanded with us three times more than customers who are not part of that community. So we know that a great community effort around a durable culture leads to value for both us and our customers. And in most organizations who have communities, community reports to somebody in customer success or marketing, which is probably the worst place that can probably be, we just did it differently. Because we didn't want to lose that, that durability of culture.

Tom 21:27
That's fantastic. Okay, now we're gonna go so from that gets created Alteryx for Good. Talk a little bit about that with us.

Dean 21:36
Yeah, so your Gosh, you're touching on, like, the most important things that I ever, the most important decisions I ever made at the company. 

Tom 21:48
Dean, I tried to do my homework, man. 

Dean 21:49
These are not the typical questions you would ask the head of SAP, right. But it's modern technology for the modern analyst. And so you have to ask different kinds of questions. Well, this goes way back actually, when I was an undergraduate at University of Colorado 1976. I was a sophomore, it was the conference on world affairs, which is still held in Boulder every year. It was the nation's bicentennial. And my father was a huge, who also went to CU back in the day, he was a huge Buckminster Fuller fan. And I don't know if your listeners know Buckminster Fuller but if you don't, you should read up on him. Engineer, architect, systems theorist, studied space in chemistry. And he did everything, didn't make any money. But one of the first autonomous vehicles, or self parking cars, I guess you could say, the geodesic dome. And I went to listen to one of his sessions, and it was called Everything I Know by Buckminster Fuller, so I knew it was gonna be a long session. And I had to go to class. But right before I left for class, he said, the most profound thing that stuck with me until I launched Alteryx for Good. He said, by the way, I just want you to know that we're building all the right technologies, for all of the wrong reasons. And we are never going to be able to take care of spaceship Earth very well, nor for very long, if we don't see it as a common cause. It has to be all of us or none of us. And that just gnawed at me for years. Because as you're building the company, you want to do good work, you want to be philanthropic, you want to try a bunch of stuff. But I didn't have any money. I was self funded for 14 years. So I didn't have money to invest in any of this. But in about 2012, we started offering our software to universities. We were offering it to 501 C three nonprofit organizations, because, you know, trying to give back is probably harder than making money with analytics. And so it was concretized in 2016, called it Alteryx for Good. And today we have, I don't know, there's hundreds of 501 C threes that are solving all kinds of world problems from eradicating malaria in Zambia, to saving the whales, to helping endangered species in the Amazon forest. We even sent teams a few years ago, we sent teams to Health and Human Services, to crowdsource problems or crowdsource answers to the opioid epidemic. 

Tom 24:25
And it's not just given the technology platform, it's also a way for members of your organization to go volunteer with societal problems.

Dean 24:32
Oh, yeah. Yeah, there are lots of people who just learned Alteryx over a ham sandwich, but there's other people who have to actually understand the entire pipeline of analytics, from knowing where the data lives, knowing what kind of a structure it is, knowing what quiet questions you can or can't ask, or can or shouldn't ask, and then just helping them kind of get started. And it's been a lot of fun. We always give Alteryx for good awards to organizations around the world that do incredible things. And that's part of the culture, that's part of the bond that allows that culture to be durable.

Tom 25:09
So the the title of this podcast is life in an increasingly digital world. And at the time, you and I are talking, we're at the peak of inflated expectations about AI, large language models. So let's put the hype aside and say this is going to likely be one of those moments where we're going to overestimate what it can do in the short term, but underestimate the long term. Where do you see this period of technology's ability to impact society good and or bad? Where do you see it taking us?

Dean 25:41
Yeah, I think it is hype right now. I think that if you listen to CNBC, no CEO mentioned AI until they started talking about chat GBT. We've been doing AI for a very long time. And there's, there's actually nothing artificial about it. It's kind of built into the DNA of the platform itself. I think that if you look back at history, you know, a lot of the hype cycles didn't turn out to be true. I mean, you remember Alan Turing back in the late 30s or early 40s, you know, he was hired by MI6 to try and resolve German Enigma during World War Two. And they built the bomb, the super computer, and he had to inject human logic to actually resolve it. It was a simple thing and the machine was never going to outsmart the human. I mean, you can go back to Big Blue and the chess contests where the three ordinary analysts with mediocre computers beat the supercomputer with all of the intelligence available. We've been talking about singularity, for example, for, what, 30 years, 40 years, 50 years, and they keep pushing it out. And I'm a believer that AI is going to be important for sure. I think we're going to be in a world of hurt, because there's going to be a lot of nefarious activity with AI, I think that, an election cycle will prove it, where there'll be fake speeches given by people who look just like the candidates, and people won't know what to believe. Look at the models of social platforms in the last election, they were rejecting things that didn't make any sense to reject. Here's a little experiment for your listeners, go just Google AI mistakes in the last five years, and you'll find machine learning challenges that Microsoft where they're trying to be more inclusive in their hiring process, and they found out they were more racist in their hiring process. And so, we're a long way from perfecting this. And I believe that if you amplify the human, we will never reach singularity. Can AI do good? Sure. Initially, we'll see simple things. We rolled out our Aiden AI platform, and it's got some cool things. If you want to, for example, convert all your AR based algorithms and Alteryx to Python algorithms, all you have to do is send your code base for AR and ask chat or any other Bart are probably not Ensign based on their data. But it'll convert your models automatically. If you have an unstructured data set, you can say help me build a regular expression to clean up this dataset. So we're gonna use it to accelerate the experience of analytics. So now we've got more humans that are skilled, and the process is now accelerated. So I think that's where the focus should be, I think, I think that there's gonna be a lot of bad people doing bad things with AI. I wrote up a piece, I was a guest writer, columnist for Forbes, for a couple of years, and I wrote one piece on the ethics of AI. And I referenced a movie, a cartoon movie called Wally. And they talk about, as artificial intelligence begins to emerge, people will become dumber and dumber, and they'll get fatter and fatter. And they'll eventually live in spacecraft because no one will want to live on planet Earth. And I think that's the risk is, is that humans will get less engaged and we can't allow that to happen. The other thing your readers or your listeners should read is a seminal paper by JCR Licklider 1964 called Man computer symbiosis. And he talks he's given he didn't make any money either but he's been credited with a lot of things in tech, the computer mouse, the graphical user interface, online banking. And you would hear Bill Gates talk about JCR Licklider or Steve Jobs, and and he talks about the user experience, that if you just remove the friction between the man and the machine, or the human and the computer, that singularity will never be reached. And I think he's right. Human intelligence is always going to be required for AI to actually do much for planet Earth.

Tom 30:10
The last question, Dean – What can we expect from you in the future? What are you gonna go do next as part of how you want to contribute with the time you have on this planet? 

Dean 30:22
Well, I think any, any entrepreneur, especially successful entrepreneurs, but I think even unsuccessful entrepreneurs have a lot to give back. Lessons Learned. So I started right away I in an effort to try and figure out how to deploy my own balance sheet. I mentor startup CEOs, I currently have eight of them around the world at various levels, some are pre revenue. One is pre product, which is gonna be a bit of a challenge. And these are all AI machine learning startups, in different industries from, disrupting the PowerPoint market to make presentations more immersive, to disrupting the maritime industry, because it turns out the shipping industry is not very enabled. And so I mentor these folks because I think my journey is probably more realistic than Palmer Luckey’s. He's, obviously he's done a great job. And what he's doing now is incredible. But I think my journey is probably more the one that they're going to face. And so I have one on ones every two weeks with these eight startups from Istanbul, Turkey to Tel Aviv to Mexico City to an Irvine. And to me that that's kind of a an obligation. I think that I have. 

Tom 31:38
We'll leave it there for today. Thank you so much for joining us on the podcast.

Dean 31:40
All right, Tom. Thank you. Bye, bye.