Alter Everything Podcast

BrianO · ‎07-31-2018

We're joined by Pete Goldey for a chat about turning frustration into innovation.

Guests

Pete Goldey- (@PeterGoldey, Twitter)

Brian Oblinger (@BrianO, Twitter)

Topics

Folklore and Mythology degree
witlytic
Zeel (massage on demand) - use code ALTERYX at check out to receive a special discount
Subprime housing crisis
Freedom of Information Act
Realtor.com
Sperling's Best Places to Live
Example of Pete's "Best Places" work
Money Magazine
Spatial Analytics
Inman Connect
The Big Short
Joe Fernandez
Onboard Informatics

Community Picks

Pete:

Brian:

Transcript

Episode Transcription

BRIAN 00:06	[music] Welcome to Alter Everything, a podcast about data and analytics culture. I'm Brian Oblinger, and I'll be your host. We're joined by Pete Goldie for a chat about turning frustration into innovation. Let's get right into it. All right. We're here with Pete Goldey. Pete, welcome to the show.
PETE 00:29	Hey, Brian. Glad to be here.
BRIAN 00:30	Yeah. It's awesome to have you on. So if you could do us a favor, maybe spend a few minutes talking about kind of who you are, how you got to where you are today, and some information about kind of what your background is in, would be great.
PETE 00:44	Sure. I'd love to. So I think, like a lot of folks who are into analytics these days, my path to it has been not necessarily straight. And there's been some interesting winding in it, but I think some markers along the way that kind of pointed me towards where I am today. So I didn't study analytics or computer science in school. I actually was a folklore and mythology major because it was there, and it's seemed pretty interesting, and it was a lot of fun. But it actually translated pretty well into some of the things that I've been doing since then because folklore and mythology is really about trying to understand why people believe what they believe when they wrote it down. So if you think about sort of ancient religions and things like that, we had things like a sun god or whatnot, and people believed those things because they didn't have a better explanation for it. Right? So you look at what are people's motivations for the things that they hold to be true and try to explain them. And if you translate that - ultimately, it took me a while to realize I was doing this all the time - into kind of business analytics, when people are managing by kind of gut feel and intuition and personal experience, they are often in that sort of folklore and mythology area where they may be on the right path, but they might not be because they're trying to explain something they don't necessarily have great information to point them towards a good answer. And when you can illuminate that or expose information to them, often what they believe might change. Just like people don't-- or most people anyway, don't believe in Thor and Loki anymore as being powers in the world and in the universe.
PETE 02:37	And look, a lot of people believe lots of really interesting things. But I think part of the reason there has been so much difficulty or conversation about religion in general, and people have been struggling with some of their beliefs and are trying to reconcile these issues between maybe what's in whatever religious text they might follow versus what we see in the world or what science is illuminating for us. It kind of comes down to this. Right? And so how you reconcile those things is often, on the business side like the biggest challenge that's there. Because when you have executives or stakeholders that have a preconceived notion of something from their experience or whatever it might be, but your information ultimately shows something different, that could be a real struggle to communicate. And I think that's where great analytics and communication skills come in. And that's kind of where my career has really evolved over time. I started off with a small data company up in West Chester. I think I took a typing test. That's how I got a position there [laughter]. I had no database experience or anything else like that, and I ended up involved with this big data-collection process and built a database in Microsoft Access, at the time. This is back in the '90s. And it was my first experience with any of that. I really enjoyed it and kind of stuck with that company for a while and then moved off into some other things. But it all kind of started for me with wanting to understand how to maybe make better decisions or understand what's going on better. And that's where I started.
BRIAN 04:24	Wow. That's fascinating stuff. And so it's really interesting how the folklore and mythology has sort of faded into data and analytics. So what are you up to today? What's kind of the current gig that you're rolling with?
PETE 04:39	So right now, I'm working with a great company called Zeel. It's really cool. They're a on-demand services and technology company focused on in-home massage. So what's really neat is you can download the app and with a couple sort of swipes inside the app, you can book yourself a massage sort of on your terms. Someone will show up at your house and the real differentiators that Zeel has in the marketplace for this - and they are kind of the first company in the space - is a lot of the security and vetting that goes around. That process, I mean obviously, you can imagine you're inviting someone into your home, and on the other side of that as a licensed massage therapist, you're going to go meet somebody you've never met before in all likelihood. So there's actually quite a bit of technology involved and a lot of sort of analytics, fraud detection, and all the typical business questions that you have with a company around churn and attracting new customers and things like that. So they had some really interesting questions to solve, and I had been consulting with them for a little while and then decided to come on board and lead their analytics and data science efforts.
BRIAN 05:56	Well, yeah, I'm downloading that app right now. And if anybody has any questions about why I'm getting a massage in my office later this afternoon, it's research for the podcast. So--
PETE 06:06	Exactly.
BRIAN 06:07	--Dean or Olivia, if you're listening to this, this is a qualified business expense.
PETE 06:12	Why and it's funny you mentioned it that way, that a company also does corporate wellness programs and things like that. So I mean there's a lot to it, but I didn't want to go into it-- too far in here [laughter]. Before joining the company, I really was focused in the real estate information space, and I think that there are a lot of interesting things to do there. And, of course, we went through the housing crisis and things like that. So I had started a small company with a couple of guys which eventually got sold. But there was a lot of innovation available in that space around analytics and that was really a lot of fun. And that's sort of what I focused on for about 15 years or so between school and where I am now.
BRIAN 06:56	I'd love to drill down on that innovation piece because kind of in reading up on you and doing the research that I do before the show, it's clear to me that you, yourself, are somewhat of an innovator. So maybe tell us a little bit about how you see that and some of the things that you've done that have taken this nation by storm.
PETE 07:16	Well, I hope they have. I like to think that I had a part in changing the way people actually look for real estate online today. But all of that work was kind of under the covers or behind the scene. So the company I started working for was a data provider that was, back in the day, kind of collecting a bunch of real estate information that was really only available on paper and digitizing it and then making it available to multiple listing services which are organizations of realtors that we all work with when you're looking for a home and eventually, also making it available to online real estate websites like Realtor.com or, today, Trulia and Zillow, etc. So the first thing that was kind of innovative there was we had to break down the doors of being able to collect the information and digitize it. So we were collecting this data in Connecticut in all these little towns, and a lot of them either said, no. They didn't want to give it to us. Or realizing that they really couldn't keep us from getting it, they wanted to charge us a dollar a page to make copies even though we were requesting it in digital format. So there was actually a freedom of information case - I think I was 24 or 25 at the time - that I kind of organized for the small company I was working for. And we stuck with it, and eventually won the case. And the result of that case is all these companies you know of today can request public records from municipalities. And the municipalities have to provide it back to them, reasonably, in the format requested and at the cost to reproduce instead of this dollar-a-page concept which was from Xerox machines and things that people don't even-- I don't even know how many listens know that--
BRIAN 09:09	Many moons ago.
PETE 09:10	Yeah. So the project, unfortunately, failed, that we were doing - but we stuck with it - because of the delays in collecting the data, but we kept with the case. And as a result of that case, we were then able to launch a bunch of new businesses, I guess, out of it, because we suddenly had access to this data, and became kind of the middle man between all these municipalities and the ultimate consumers of this data which were end users online. And in between those two points, there was us and then whatever the web property was, like Realtor.com or what have you, that was actually built into the user experience. So that process, that was like a innovation in business process, I guess, more than an analytics, but that kind of opened the doors to what we did later. And what we did later was I think change, again, the way that people perceived looking for real estate online by providing information about things other than the property. So what I mean by that is if you go back in time quite a bit, the only information you could find about a home that was for sale was typically handed to you on a piece of paper from a real estate agent. And it was really just the information about the home, some pictures and bedrooms and bathrooms and things like that. But when real estate went online and you were able to start searching for homes by zip code and price range and things like that, there a lot of other opportunities opened up that people didn't really see. And I think we were the first folks to see that and those were to allow people to look for areas first that maybe met their criteria like they were looking for a certain level of school system or wanted to make sure it met a safety standard or it was pet-friendly or something like that.
PETE 10:58	So we actually pitched Realtor.com which was really almost the only player in the game nationally for real estate search, and we said, "Hey, you guys have this great search where you can type in a zip code and see all the homes for sale. But if I'm moving from New York to LA, I don't know any of the zip codes. I don't even know where to look. How would I even know what neighborhoods I want to live in or think about looking at for properties in?" So we pitched them this idea where we put together a whole bunch of demographics and school and crime and other statistics for them so they could create community pages. And there were some forward-looking product managers over there at the time and they bid on it. And we put this product together, and they turned it on. And they found that 30% of their traffic actually went through looking for areas first and then looking at homes in the areas, so it was an immediate sort of hit. And, of course, we had taken a risk and they certainly had also in putting it out there. No one had tried it before, but it was a hit. And then that, of course, led to us being able to grow that business in a lot of directions.
BRIAN 12:04	That's incredible and I think now it's amazing. A lot of these things are just available. It's just an API. Right? You can plug [all tricks?] into an API and hit run and it will bring all that information in. That's a long way from making a request on a per-page cost. Right?
PETE 12:19	Yeah. The landscape has really completely changed. Back then we were doing things like collecting information that didn't exist at the zip-code level and modeling it ourselves. That was a big part of what did. We did that with neighborhoods eventually and other things. But that business evolved in some interesting ways because we were delivering this data in bulk to people, and they were building their websites off of it. And eventually, almost any website you could think of was pulling that content from us. The Realogy brands, Trulia, and Redfin were customers for a while, all the big brokers. But what we found was that there were a lot of limitations. And when it comes to innovating things, I feel that innovation usually comes from a couple place, like pretty different places. But one of those is someone literally just subconsciously connects the dots between things and has this great idea that no one thought of. I think those opportunities happen infrequently, but when they do you get things that nobody's even ever thought of before. But I think what's more likely to happen is that innovation comes from frustration. So whether it's you being frustrated with some sort of process or a client or yours or customers, if you have enough kind of information surrounding the problem, then you can often look at these frustrations and come up with some interesting solutions. So in a way that searching by communities was an example of that. It's was frustrating to try to figure out where around LA or Seattle or which suburbs or which neighborhoods you might want to live in. And that led, ultimately, to some other frustrations at the time that led to some new products.
PETE 14:11	There was a website out at the time, and I think it still exists. Though it's certainly changed and is much better now, Bert Sperling's Best Places to Live. And I remember going there. I lived in New York and I was thinking about moving, and I was like, "Let's see if it can recommend some cities for me." Right? And I plugged in some information and it turned out that no matter what else I put in-- if I entered in that I wanted good weather like warm, sunny weather, Tucson was always number one [laughter] on the list. Like it didn't--
BRIAN 14:41	Imagine that.
PETE 14:42	Yeah. It didn't matter if I wanted the other things: schools or crime, culture, whatever. It didn't matter. So as a data guy, I was thinking there's a better way to do this, right, and there definitely is. Right? And that ultimately led to developing a new product that we went out and pitched again to all these real estate brokerages and websites. We called it lifestyle search which again is now, today, not a new concept. But at the time, back in about 2008, it was, where you would ask the customer, "Hey, what's important to you?" or, "What are your preferences, and how important is that preference?" So you kind of ended up with a multivariant search where you were weighting the inputs and then providing a best-match result instead of this kind of hard-filter, gated process which is like that old website that I had tried. So we built this, and it leveraged everything the company did. Right? So we had all this geographic data and a geographic model. We did all the spatial analysis to say, "Hey, these neighborhoods are in these cities, and they all roll up in this way. And crime statistics are available at this geographic level, but we need to model them down to block groups," for example. So it was really a big analytics project and then on top of that, to build the weight-sort mechanism to provide best results was a fun challenge also. And it ended up in a product that was typically implemented where you put up 10 or 12 different topics on the web page and let customers sort of drag a slider from most to least important. So you put up something like, "Do you want to have good schools? How important is that to you?" And you think it'd be important to everybody, but if you don't have any kids, often good schools go hand in hand with higher taxes. So everything sort of becomes sort of relative. And we did some learning there which is what I think good analytics is all about, which is initially we just put the product out and we found that everybody who had kids always put great schools, low crime, and low home prices. Right?
PETE 16:57	So they weren't actually differentiating among the things that they wanted. What was most important? And we found that we had to sort of limit the number of points they were able to distribute from a "how important it is this" and force them to differentiate, force them to prioritize. And if you can get your users to prioritize their input for you, then it becomes much easier to provide good results. You could always provide results even if they're not prioritizing, but those results aren't going to resonate in the same. Even in cities that have relatively poor school systems, there are always some schools that are better than others. Right? So there are always some areas that have lower crime or higher crime than others even if it's a safe city. So by having them provide these preferences, you can factor that into an algorithm and provide better results and then, of course, test that. So that was another project that was kind of born out of frustration. And I think in the real estate space, that's kind of been the history of what we did with this lifestyle search engine was one. We put out the first search API for property search which allowed our customers to focus on selling homes instead of building all kinds of interesting search which could include these lifestyle characteristics but also things like drive time, commute time, whether it was in a certain neighborhood, and then factoring in all the different demographics and community data that we had. To build that yourself, you'd have to gather all that information yourself, organize it all, then kind of become expert in how to build the search. And a real estate brokerage, that's not their business. So they were very frustrated. They wanted all the data but couldn't figure out how to leverage it. So we heard the frustration and then built a solution for them from a business-to-business perspective.
BRIAN 18:58	Yeah. I love what you're saying about the frustration piece because I often think that the difference between someone who's a complainer or an innovator is simply how they channel their frustrations. Right?
PETE 19:08	Yep.
BRIAN 19:09	Like some people will just simply kind of take it, so to speak, and just, "Hey, that's just the way the world is. I'll complain about it. Every once in a while, I'll grumble, whatever." Some people get frankly pissed off enough to go change the world or try to come up with a solution. So I think you're right on there. I think that makes a lot of sense. When it comes to sort of these listicles of the best places to live or how you use the data, I think one thing that's really interesting there as well is sort of the full scope. You're describing this project or projects, and you're using almost every facet of these analytical products. Right? You have spatial. You have mapping data. You have all kinds of things, the full scope there. Maybe talk about that a little bit. What's it like when you're trying to solve one of these problems, but you're ingesting and using such crazily different disparate data sets.
PETE 20:12	There are definitely some challenges there. I find them to be fun challenges, but I guess that's maybe what's makes me good at solving these types of problems. So the best places to live in these lists is kind of an interesting side avenue to what I've been doing over the years. Money magazine was actually the company that we started. I started with two other gentlemen, and eventually, that company grew to about 50 people. The very first client was Money magazine. They had been a customer of a previous company I'd worked at. And after that company went out of business, oddly enough, and not in any way related to 9/11 but in New York, they went out of business on 9/11. Money magazine called up a month or two later and said, "Hey, we're ready to run our story this year." And we said but, "Okay, but the company doesn't exist." And they said, "Okay, but we're ready to run the story this year." So we put our heads together and said, "Hey, how can we do this?" And they ended up becoming our first customer. And over time we also worked kind of the same sort of stories for other publications, U.S. News, Essence, Forbes, Fortune, a whole bunch of them, Family Circle.
PETE 21:32	And, of course, if you're doing the same type of product for a number of customers, you start to run into questions about how do we differentiate the product. And we actually, I think, had a pretty straightforward but good process for that which was we tried to understand our client, tried to understand the problem a little bit better. If you think about one of these lists, it's really what are the best places to live for the readership of that publication. That's what the publication's looking for. They're not looking for some totally objective or impartial result. And it's hard to even come up with that because the best place to live for one person is not the best place for another. So you have to start somewhere, and then once you know where you're starting, kind of build the process around that, in that context as impartial and objective. So what we would do is we would ask them for their customer marketing profile. And then we would look at that, and we would try to understand "here's marketing speak for these people". How does that translate into real data that's available to us out in the world? And that information becomes different depending on what that marketing profile was.
PETE 22:44	In the case of Money magazine-- and I still work with them, so you should all definitely go out and get their Best Places to Live feature that comes out in the fall at some point, we just finished up the data work for it. We put about 200 different data sources into that, and it's all sorts of things from home prices and current market statistics to crime to availability of conveniences nearby to commute times, you name it. And it was a pretty big challenge bringing these things in. It's become a little easier over the years as data sources have become better organized. But in reality what you have is a problem where you have data that comes in to you at all different levels, state, county. Point data meaning like these businesses are really addresses, schools are addresses. School-district boundaries don't match up with city boundaries all the time, certainly not with zip code boundaries. Lots of data is only available for zip codes. Weather is related to weather reporting stations. So it's really a spatial matching problem.
PETE 23:53	And when we first started using it, this was before Alteryx came out with a product, we were using Oracle Spatial and [Android PiP?] [inaudible] map info product to essentially do a whole bunch of spatial analytics on the data to be able to aggregate to boundaries and to proximity searching, radius and buffer counts, all these things to essentially pull this data together so that I could tell you what's the convenience score, or an index for this small town in the middle of Iowa. And eventually, we redid that process. So I redid that process entirely using Alteryx tools because it's really well suited to that with all the spatial tool set involved. But the problems all remain the same. And organizing that data and pulling it together and creating something intelligent out of it is an interesting challenge.
PETE 24:49	So I'm working with editors and writers at the publications, and I could output to them a spreadsheet full of data but they don't necessarily understand the differences in the types of data points. Like their scale, what would be the differences in scale? And how should that impact what they're doing? Because they're taking what I'm giving them, and they're trying to create an algorithm that for each place combines all this data and scores it, and says, "Hey, this is going to be in my top 100 or top 10 list, and this place isn't." And then, of course, they have some editorial processes on top of that. They go out and they interview people and do other things which you can't capture everything in the data, of course. So in addition to combining the data together, then there's a variety of work and modeling work around trying to create scores that they can then combine in a way that's meaningful and don't overly value one data point over another. So a simple example there is if you're dealing with home prices, they're typically in hundreds of thousands. If you're dealing with crime index score, they're typically in the hundreds or the double digits. And if you simply added them up and average them or something like that, you wouldn't really be doing justice to either of the data points involved. So that's a process. I've been doing it for about 20 years with them. Their teams change quite a bit over time, but the focus of the stories really remain the same with Money magazine, in particular, and has really been a great success for them.
BRIAN 26:30	That's really cool. I think we're a little-- I don't know. I feel like these days there's so many - I'm using our quotes here - lists or listicles out there on the web about things. And I think a lot of them feel like they're just kind of thrown together or someone just made some editorial decisions about what the list should be rather than fact-based decisions. And it's really cool to hear, in this particular case, all of the great lengths and processes and things that you're doing to really get to what is a objective by the data cut of what actually would be a great place to live if I was in Money's prime market, right, for who they market to.
PETE 27:15	Yeah. It's a fun process. And I've had the benefit, I think, of being able to help influence them to take into consideration new data sets over time. Obviously, the project's gotten a little bit more complex over the years, but recently they focused more on things like diversity. There's a lot of great data out there on diversity, not just racial diversity, but cultural diversity around things like food, economic diversity, diversity of recreational opportunities and things like that, that can really shed some color, shed some light on how an area really feels when you're there. So that was one thing we've added, some tax burden information, and other things like that, that really seem to resonate with folks. I think we try to stay a little bit away from some of the political questions because they don't always align in any way with any of the other data that we're working with for this type of-- for this type of story. But outside of that, it's nice to be able to have an influence on what are the things that they're trying to consider. And kind of a funny story is every year-- because this is a data-driven story, every year I get some calls and emails - because I guess I've been doing this for so long that my name's out there a little bit - from towns that want to be in the list. Right? So they'll say, "Hey--"
BRIAN 28:52	Tucson.
PETE 28:53	Yeah. Tucson. So sometimes we focus on big cities or small towns, so there's sort of this natural cut about whether a place can be a candidate. But one town in Nebraska, every year they made a-- City Council made a resolution, a 10-year plan. They want to get on the Money magazine Best Places to Live list. And they're very persistent in calling me up and saying what do we need to do to make the list. Right? So I, of course, don't divulge any of the secret's source, so to speak. I just say, you just have to be a better place to live. I mean that's really it.
BRIAN 29:31	Yeah. Make sure the sun shines more. Right?
PETE 29:33	Yep.
BRIAN 29:34	Cool. So one thing I'm thinking about here too, we've been talking about housing and the data and all of this for some time. You mentioned before you've been doing this for 20-years-plus. What indicators, if any-- I guess I'm making an assumption - you can correct me if I'm wrong - that at some point along the way here you were doing this work, looking at this data, diving very deep into it. Were you able to detect anything around the housing bubble or the housing crisis? And if so, maybe tell us a little bit about that like when did that become apparent to you, is it did. And was it ahead of the rest of the market?
PETE 30:22	Great question. And, yes. We did see some things in the data. So back in about 2007, 2008, we were processing-- pretty much all the public records in the country were coming through our office one way or the other. So we were able to look at-- and in fact, we had products built around things like home-price trends and volume on the market and things of that nature. So we were able to look at those things and we started to see some, I think, things that seemed a little odd in the data at the time. Now, keeping in mind our business was not about making predictions or anything else like that. In fact, our customers were pretty much people who were selling properties. So this wasn't necessarily the type of thing that we were thinking about building a product out of, but just bring naturally curious, a gentleman I worked with who was actually our director of data operations, Joe Rugolo, and I - and he's moved on to some other great things - started looking at some of our data sets and realized that there were some pretty interesting trends going on that ultimately, as you look back on it, sordidly related to the housing bubble. And we ended up creating what we called a distressed area index which we did at the block-group level. And we originally did this because we thought that it would actually be a great informational tool for government. We were at the time, as a company, trying to see how our data could be useful to people who are making policy or who were trying to figure out what were the areas that needed more funds for something or more affordable housing and things like that.
PETE 32:07	So we were looking at things like what was the ratio between 30-year-fixed mortgages and the adjustable-rate mortgages, the three and five-year ones. That was so popular back then. And when those things were coming due for their adjustment and a bunch of other data points - residential turnover rates, fair market rent - and then some sort of maybe negative indicators like employment or job growth or the lack of that, mortgage delinquency rates which were actually available publicly at the zip-code level, you don't get information about the individuals, but you can see things like the number of people who are behind by a certain number of months. So there's a lot of data out there and, of course, the fun part is figuring out how to put it together. So we put together this index, and it just spit out all these numbers. Right? And it was hard to make sense of them, so we used mapping software and built some heat maps around block groups. And we did it just for a handful of cities, at the time, New York, Los Vegas, and maybe Los Angeles, if I remember correctly. And it sort of would light up the city, here these areas seem like they're at high risk, right, or relatively more distressed based on these data points and others.
PETE 33:30	And in retrospect, looking back on it several years later, I think that that model actually did a pretty good job of predicting the areas that were likely to have foreclosures. So it was a project that we actually presented at the Inman Connect Conference which is a real estate technology show here. It was in New York in 2008 or 2009. And there was a panel about "What do you predict's going to happen to the real estate market in the coming year." And I remember sitting up there with a whole bunch of real financial analysts and industry experts. And, for the most part, they were all predicting kind of rosy colors because it was sort of before the burst happened.
BRIAN 34:13	Sure.
PETE 34:13	And then I remember putting up my slides and saying, "Well, we're seeing some high levels of potential housing distress in these areas," and getting a lot of questions about it. It was definitely a fun panel, and I was kind of the outsider in it being much more of a-- less of an industry person and much more just of the data and analytics person. I wouldn't say we predicted things perfectly or I necessarily would have felt comfortable with banks using our index to evaluate mortgage risk or something like that, but the trends were there. They were definitely there if you were looking for them and knew kind of how to look. And the fun part was really taking data that wasn't so directly related, things like employment or housing starts, things that weren't so directly related to actual mortgage volume and mortgage delinquency, and realizing that they correlated very directly with some of the issues that were going to pop up.
BRIAN 35:14	Sure. Yeah. Clearly, the economics of that situation were far-reaching beyond just simply the house mortgage situation. And it is interesting looking back on that now. I just went back and watched The Big Short again the other day, loved that movie, and it is interesting how most of the people that were "right" to your point, not all the predictions were spot on, but directionally, the people that were "right" during that period of time and were often getting laughed off of stage at these conferences and things were the ones that were saying, "Hey, we're seeing things in the data," and the other side were people that either had too big of a vested interested to kind of say the truth or what they'd saw happening or they honestly, because they weren't data-driven, they were going with their gut. And they say, "Hey, look. The market's great. We're selling houses like crazy. It's all good. I don't know what you're talking about." Right? So I think that event, among others, has definitely contributed towards what we're seeing now. Here we are 10 years later with this move towards data-driven decision making and, "Hey, let's not ignore the signs that we ignored last time," and hopefully that bodes well for the future.
PETE 36:32	Right. Yeah. Maybe it points a little bit back to that whole idea of folklore and mythology. One group of people were believing something because they had a vested interested in it or it's the way it had always been. And then you sort of had new information coming in and it was kind of hard to spot the new information or to realize that maybe it's something you should be looking at. But for anyone who's out there listening and trying to figure out how to communicate data like this that might be complex or might be very localized or other things like that, the real key for us and what resonated with people was ultimately putting it up on a map, right, and having the right tools available to be able to do that sort of thing because most people can't look at the spreadsheet and see the shape of the data. I think that's something I do. And it took me a while to realize that other people don't do that so easily. And so if you can put it into a visualization that other people can relate to, that can take the message you're trying to get across and actually have it resonate with people instead of everyone sat there in a meeting with somebody where you're trying to present something and people's eyes get a little glassy. Usually, a picture does tell a thousand words or a thousand numbers in this case.
BRIAN 37:55	Yep. All right. So switching gears just ever so slightly here. A few minutes ago you were talking about starting a business and growing that from scratch up to 50-plus people, maybe talk a little bit about that of being a business owner, seeing it grow. What is the communication like when you have this big idea and you're trying to translate that into a viable operation on which other people's lives are also dependent for their income and things like that?
PETE 38:30	Sure. Yeah. It's a hard kind of road to navigate in some ways. When you're starting off, you have a lot of flexibility in many ways, much more so that as you grow because you don't really have a lot to lose. If it's just you, I mean, sure you have to fend for yourself and have some sort of income, but you don't have people who are clients relying on you or what we call technical depth that you have to support and things like that. So starting off, it was great. We threw a lot of stuff up against the wall and kind of saw what stuck. And of course, we tried to make informed decisions about what we were putting our efforts into. From my previous experience, I had a lot of domain expertise and some great contacts in sort of the real estate information space which was good. I think it's hard, in many cases, as somebody whose maybe on the technical side instead of the relationship side of product development or ideas, right, coming up with ideas, to be able to validate what you're thinking about if you don't have those relationships in place. And my other partner certainly brought that to the table. I was really the product and technology and data person when we started the company. So that was great.
PETE 39:56	And I think I have the skill set to be able to communicate between the technology and the business side, which is also often a challenge for folks. Some people are great at it. Some people aren't. If you're a great-idea person but not great at communicating it, I would definitely recommend making sure you hire for strengths and bring someone on that can really be compatible with you but also add that dimension to what you're doing. That's actually one of the growth strategies that we really employed and I think is really important also when you're dealing with data and analytics is not everybody is going to be good at all aspects of any role that you're trying to put them in. So recognizing what they're great at and then giving them leeway in that area and actually trying to force as much of their bandwidth into the things that they're good at, instead of saying, "Well, they're really not good at recording or doing documentation, so they have to focus on that," and then making them miserable by forcing them to spend a lot of time on something that they're not good at. I think that's a mistake that a lot of folks make. So we always try to do what we call managing to strength. Right? So manage to people's strengths.
PETE 41:14	And on the data analytics side, some people are great at modeling the data, understanding and looking forward, how an application might leverage a data set, or leverage the structures and being able to be forward-thinking enough to future-proof, at least to some extent, what you're doing. Other people don't have that skill, but they might be amazing, absolutely amazing, at taking the output of your systems and understanding what's going on and, right, actually performing the analytics and then interpreting them. So that's one area, I think, as you're growing is to recognize, first of all, what you're good at and what you're not good at, and then make sure that you hire and manage to people's strengths instead of to their weaknesses.
PETE 42:01	The other thing around, I guess, growing and being able to do that in a kind of a measured sort of way is to stay in touch and in contact with your customers. Right? So creating a market is really difficult for a product. To be able to do that from scratch, usually, there have to have a huge PR or marketing budget or somehow have a big lucky break. Right? And you see these companies that they come massively funded well ahead of their revenue as a result of that. That wasn't our situation. So what we did have were these relationships with other companies that had market share. And so originally we tried to put out our own products for consumers, and people loved them, but nobody ever was using them because we weren't out in front of people. We weren't promoting them. We didn't have the dollars behind that. So instead of going down that path-- our path was really around empowering our customers, and that worked really well for us, and certainly, at the point in time when we were all very focused on that business. We were the first people in the space. We stayed ahead of the competitors. We innovated a lot of products. Our customers trusted us, so that allowed us to get them to try new products like that lifestyle search engine where people would find like the neighborhoods in New York that best matched up with whatever they looking for. That was brand new, so when Coldwell Banker put that up on their website, they were taking a leap of faith in us that we had built a good product and that it was going to resonate with their customers. But that was a lot easier for us than, say, building a marketplace from scratch, if that makes sense.
BRIAN 43:47	Yeah. Absolutely. Okay. Anything else you wanted to talk about today? Any other pearls of wisdom we should be sharing with the listeners?
PETE 43:57	Well, maybe just another quick story about ideas coming out of frustration. And there's a guy I worked with who-- Joe Fernandez, is a great success story, really smart guy, worked with us as a product manager, but he had jaw surgery at some point. And he couldn't talk for like two months. Right? This is, again, back in the late 2000s. And he was trying to figure out things like what music should I listen to, what restaurant should I order food in from, or what movies should I go to? And this was sort of really before the heyday of a lot of social networks and things like that now. And he came up with the idea that ultimately evolved into the company Klout which I don't know if you heard of. And they're gone now also, but they were the first social influence measurement company. And he came to me and we sat down, and he was kind of scribbling on some paper because he couldn't talk. But he was going over this idea, is like how can we find-- how can you find out who-- not who like something but who could you trust their opinion and then make decisions for yourself about what they liked. And that was sort of the whole genesis of this concept of social influence in the marketplace and there it was all born out of frustration. Right? And he built a really big company with several hundred people and then eventually moved on. It was very successful from there.
PETE 45:37	But I think the lesson from that is there is an idea there that he came up with because there was a very personal frustration and then the challenge, of course, was trying to figure out from a technology and analytics perspective is there a way to solve it. Right? And he went in and pitched folks like Twitter to open up their API more to give him access so that he could do these analytics on not just what people were posting but on their reach, their response rates people were getting, and all the other things that differentiate what I would call sort of fire hydrant influencer, who's just putting everything out there to somebody who maybe isn't as prolific but has a lot of engagement in what they're doing and ultimately a lot more influence in what other people might be thinking. So innovation, again, I think from frustration is a great place to start. And if you've been frustrated by 10 or 100 different things probably in there there's a couple of great ideas.
BRIAN 46:41	I love that you just told Joe's story. I've heard that story a number of times. And part of the reason why is, I actually, briefly overlapped with Joe at my last company, Lithium Technologies, because they had acquired Klout at some point. So I'm very familiar with Joe and this story, and he's a great guy. And what he's doing today, for our listeners, because I think this is just another-- he identified yet another frustration. And I think it's a-- you'd have to ask him. But my assumption is that the data heavily told him this too. His new company is called Joymode. And what they're doing is they've picked up on this trend that people want access to things but don't necessarily want to make the investment to own them. And so what Joymode does-- they're based out of LA, here, just a little bit north from where I'm sitting right now. And what they do is you get the app or you go their website and you basically say, "I'd love to have a backyard barbecue tonight with a big movie screen and an inflatable flamingo and whatever else, right, but I don't want to go to the store and buy all this stuff, the oversized Jenga game and whatever." And what they'll do is they'll just rent it to you basically. So you say, "Hey, I'd like this package and that stuff," and they bring it to your house, you set it up, you have your beautiful event, and then it gets returned to them. So I think another great example of sort of reading the tea leaves, looking around the corner a little bit and building a business based on some frustrations of, "I want to have these nice things, but I don't necessarily want them forever sitting in my garage collecting dust." So I mean I'm glad you told that story. Joe's a great guy. And we'll put the links in the show notes to his story and his Twitter. He's great on Twitter as well, so.
PETE 48:24	Yeah. I didn't realize that you had overlapped with him. I actually spent some time with him out at the Inspire show in Anaheim about a month ago or whatever that was. And so kind of an interesting thing about his new business, his old business was really a pure analytics business, the Klout. Right? His new business has a lot of parallels to what we do over at Zeel in the on-demand, in-home massage space, in that, it's really a logistics question or a logistics challenge. So if you think about it, he's delivering and then picking up product to thousands of people every day. Right? And he has to figure out what to store, how to store it, how to deliver it, all that sort of stuff, so there's some pretty similar problems there which I found interesting. But the original idea for Joymode, if I remember in my first conversation with him about it, was around what you just said. Someone's throwing a party. Nobody is going to buy and own a bounce house if you buy it and use it three times when their kids are little and then throw it away. That's too big an investment, so I think he was thinking about those things.
PETE 49:34	And it turns out because he's so analytics minded and focused and has really concentrated his business, in part on the data collection side to understand how people are using the products, it turns out that what people really used his service for the most are things like waffle irons and vacuum cleaners, right, and tools and things that you probably have a whole bunch of sitting in your basement. And they don't seem like the type of thing that would work in this kind of sharing "I don't need to own it mode" but actually really do. How many times a year do you actually make waffles? How many times a year do you need a wet vac, or whatever else it is. So his business has evolved and pivoted from his original idea also which is-- I only bring up because that's super important is to understand when to listen to the information that you're gathering and make a change in direction especially with a small business. If you're a big business maybe you can split off a new warranty thing or do something else like that. But when you're a small business and you're trying to figure out your part to go down, analytics can really be your friend if you're collecting good information and you learn how to intuit direction from it, and I think Joe is a master at that.
BRIAN 50:51	All right. So let's wrap up with our Community Picks. What has been interesting to you lately that you'd like to point people to?
PETE 50:58	Well, a couple of things, so one is that I'm kind of an avid reader, and I really think it's important to read, not just things in your sort of space, but things outside of it. So for me, probably as with a lot of kind of data geeks out there, I like to read fantasies, science fiction sort of stuff. But there's a great service for finding new books and finding it at, I guess, a discounted rate called BookBub, and every day I get an email from them. It's got a list of books in it that the authors or the service that's selling it have discounted for that day. And sometimes they're even free. Right? And you pick your categories that you might be interested in. In addition to the science fiction side of things, entrepreneurship, small business, applications, things like that are in my list. And I don't always find something that I want, but it's a pretty good way to build up a library at a low investment. So that's a really cool service especially if you're selective in the topics that you let them send you stuff about. So you don't get too much and it stays on point for you. So that one's pretty cool.
PETE 52:15	Another thing I listen to a bunch of podcasts. I'm one of those commuters on the train who's got the headphones on, and I'm not listening to music mostly. I'm listening to podcasts. One of the ones that I love and is actually a great way to get an introduction to some analytics and data science topics is Data Skeptic podcast. Kyle and Linh Da run it, and they do a couple of things. They do interviews kind of along the lines of this one on data science topics, but they also do a whole bunch of what they call mini-episodes. And they're a husband and wife team. And Linh Da does not have expertise in the analytics and data science space. So typically they'll bring up a topic like clustering or the curse of dimensionality or spatial analytics and Kyle will sort of introduce it as a problem to somebody who's not experienced in that area. So it's a great intro for people who don't have a background in things. And I've often gone and listened to those mini-episodes when I'm thinking about a new model or something I don't have experience with before, and it can be pretty helpful. And I also use that in my training. I do a lot of training of folks in the analytics space at the company I work with, and this is a great resource for them. So those are two.
PETE 53:40	And I guess the third one that I had mentioned which is kind of a shout out right back at you, Brian, is the Alteryx Community and the training area have really improved over the years. Right? And right now I think there's a lot of great resources out there. So one of the things I'm working on now at Zeel is distributing kind of analytics know-how and tool sets out into some of the other departments. And so we've got a group of folks from marketing and product and even the sales team who are starting to use some of the tool set. And not only can they self-serve a lot of resource there, but I've actually found, from the perspective of holding kind of training and information sharing sessions, that having them do some of the weekly challenges that are up there-- and there's like 120 of them now, so you're not going to run out any time soon, having them do that and then talking through how there are different approaches to solving the same problem and why one might have chosen one approach over another is really useful for learning. Right? You often get an introduction to a new tool because there's several hundred on the tool palette these days, and I don't use them all. Right? So sometimes someone uses something I've never used, and I get to learn from it too. So I think that's another great resource that if people aren't using, I definitely recommend it.
BRIAN 55:05	Yeah. Those are great picks. Thanks for the shout out to Alteryx Academy there. Like you, I'm also a podcast listener, so I guess I'll follow your lead here and give a couple that I think might be interesting to our audience. One is Podcast Your Data, and that's our friends over at InterWorks. They're actually a partner of Alteryx. But they have a great show where they put it out twice a months and they talk about different analytical challenges and software and platforms and things like that. So that's a good one. I also get a lot of value out of Freakonomics Radio. I don't know if you've ever listened to that one, but Stephen Dubner who was actually one of our keynote speakers at Inspire a couple years back, they sort of gloss over analytics specifically as a topic. But you can tell that a lot of the things that they say and talk about are rooted in that, and how they're using different data sets and data points to look at the sort of economics of any given topic on any given week. And they're pretty prolific. It was just like show, after show, after show that gets published. So that's another one where you'll never run out of content. And it's so widely varied what they talk about that I think you can probably cruise through the list and find something that's immediately important or interesting to you. So those are two good ones. And then also, like I said, we'll throw Joymode in there because I think it's a fantastic service, and Joe's a great guy, and it would be great for everybody to go check that one out.
PETE 56:31	Yeah. They're great, and look, there's so much great stuff out there. It's too bad that there's not unlimited time to go and listen to those things. The Freakonomics Radio is a great one also. And I think that Joe and Joymode are getting ready to do some expansion. So hopefully some of your listeners outside of LA will be able to check out the service some time soon too.
BRIAN 56:55	Awesome. All right, Pete. Well, hey, this has been fantastic. You are a crazy, insightful guy, and I'm so, just, trilled to have the opportunity to talk to you. So thanks for your time. Thanks for being on. Thanks for sharing with us. Where do people go to get more of you? Are you a Twitter guy? Are you blowing up the Twittersphere? Where do they go to find more Pete?
PETE 57:17	Well, first, thanks. And it's great to be part of the community here. So you can find me in the Alteryx Community, of course. I am on Twitter. I'm definitely not prolific there, though I would absolutely respond to folks, and I should probably be doing more. My consulting company is called WitLytic, and I'll send you the link to it because it's a little hard to spell. So you can find me there also, but my handle almost everywhere is Peter.Goldey, so I'm not that hard to track down. And I love starting up conversations with folks especially if you're thinking about some new idea. You're stuck. You're trying to validate something. Talking through a challenge is really the most fun for me. Execution is the most important thing to getting your business going but brainstorming it is sometimes the fun part.
BRIAN 58:11	Yes. And, as always, we will put all of the links to everything we've talks about and ways to get a hold of Pete in our show notes at Community.Alteryx.com/podcast. So thanks so much, Pete, for being on. Again, I just can't think you enough. This has been great, and we will talk to you again soon.
PETE 58:30	Awesome. Thanks so much, Brian. [music]

BRIAN 00:06 [music] Welcome to Alter Everything, a podcast about data and analytics culture. I'm Brian Oblinger, and I'll be your host. We're joined by Pete Goldie for a chat about turning frustration into innovation. Let's get right into it. All right. We're here with Pete Goldey. Pete, welcome to the show. PETE 00:29 Hey, Brian. Glad to be here. BRIAN 00:30 Yeah. It's awesome to have you on. So if you could do us a favor, maybe spend a few minutes talking about kind of who you are, how you got to where you are today, and some information about kind of what your background is in, would be great. PETE 00:44 Sure. I'd love to. So I think, like a lot of folks who are into analytics these days, my path to it has been not necessarily straight. And there's been some interesting winding in it, but I think some markers along the way that kind of pointed me towards where I am today. So I didn't study analytics or computer science in school. I actually was a folklore and mythology major because it was there, and it's seemed pretty interesting, and it was a lot of fun. But it actually translated pretty well into some of the things that I've been doing since then because folklore and mythology is really about trying to understand why people believe what they believe when they wrote it down. So if you think about sort of ancient religions and things like that, we had things like a sun god or whatnot, and people believed those things because they didn't have a better explanation for it. Right? So you look at what are people's motivations for the things that they hold to be true and try to explain them. And if you translate that - ultimately, it took me a while to realize I was doing this all the time - into kind of business analytics, when people are managing by kind of gut feel and intuition and personal experience, they are often in that sort of folklore and mythology area where they may be on the right path, but they might not be because they're trying to explain something they don't necessarily have great information to point them towards a good answer. And when you can illuminate that or expose information to them, often what they believe might change. Just like people don't-- or most people anyway, don't believe in Thor and Loki anymore as being powers in the world and in the universe. PETE 02:37 And look, a lot of people believe lots of really interesting things. But I think part of the reason there has been so much difficulty or conversation about religion in general, and people have been struggling with some of their beliefs and are trying to reconcile these issues between maybe what's in whatever religious text they might follow versus what we see in the world or what science is illuminating for us. It kind of comes down to this. Right? And so how you reconcile those things is often, on the business side like the biggest challenge that's there. Because when you have executives or stakeholders that have a preconceived notion of something from their experience or whatever it might be, but your information ultimately shows something different, that could be a real struggle to communicate. And I think that's where great analytics and communication skills come in. And that's kind of where my career has really evolved over time. I started off with a small data company up in West Chester. I think I took a typing test. That's how I got a position there [laughter]. I had no database experience or anything else like that, and I ended up involved with this big data-collection process and built a database in Microsoft Access, at the time. This is back in the '90s. And it was my first experience with any of that. I really enjoyed it and kind of stuck with that company for a while and then moved off into some other things. But it all kind of started for me with wanting to understand how to maybe make better decisions or understand what's going on better. And that's where I started. BRIAN 04:24 Wow. That's fascinating stuff. And so it's really interesting how the folklore and mythology has sort of faded into data and analytics. So what are you up to today? What's kind of the current gig that you're rolling with? PETE 04:39 So right now, I'm working with a great company called Zeel. It's really cool. They're a on-demand services and technology company focused on in-home massage. So what's really neat is you can download the app and with a couple sort of swipes inside the app, you can book yourself a massage sort of on your terms. Someone will show up at your house and the real differentiators that Zeel has in the marketplace for this - and they are kind of the first company in the space - is a lot of the security and vetting that goes around. That process, I mean obviously, you can imagine you're inviting someone into your home, and on the other side of that as a licensed massage therapist, you're going to go meet somebody you've never met before in all likelihood. So there's actually quite a bit of technology involved and a lot of sort of analytics, fraud detection, and all the typical business questions that you have with a company around churn and attracting new customers and things like that. So they had some really interesting questions to solve, and I had been consulting with them for a little while and then decided to come on board and lead their analytics and data science efforts. BRIAN 05:56 Well, yeah, I'm downloading that app right now. And if anybody has any questions about why I'm getting a massage in my office later this afternoon, it's research for the podcast. So-- PETE 06:06 Exactly. BRIAN 06:07 --Dean or Olivia, if you're listening to this, this is a qualified business expense. PETE 06:12 Why and it's funny you mentioned it that way, that a company also does corporate wellness programs and things like that. So I mean there's a lot to it, but I didn't want to go into it-- too far in here [laughter]. Before joining the company, I really was focused in the real estate information space, and I think that there are a lot of interesting things to do there. And, of course, we went through the housing crisis and things like that. So I had started a small company with a couple of guys which eventually got sold. But there was a lot of innovation available in that space around analytics and that was really a lot of fun. And that's sort of what I focused on for about 15 years or so between school and where I am now. BRIAN 06:56 I'd love to drill down on that innovation piece because kind of in reading up on you and doing the research that I do before the show, it's clear to me that you, yourself, are somewhat of an innovator. So maybe tell us a little bit about how you see that and some of the things that you've done that have taken this nation by storm. PETE 07:16 Well, I hope they have. I like to think that I had a part in changing the way people actually look for real estate online today. But all of that work was kind of under the covers or behind the scene. So the company I started working for was a data provider that was, back in the day, kind of collecting a bunch of real estate information that was really only available on paper and digitizing it and then making it available to multiple listing services which are organizations of realtors that we all work with when you're looking for a home and eventually, also making it available to online real estate websites like Realtor.com or, today, Trulia and Zillow, etc. So the first thing that was kind of innovative there was we had to break down the doors of being able to collect the information and digitize it. So we were collecting this data in Connecticut in all these little towns, and a lot of them either said, no. They didn't want to give it to us. Or realizing that they really couldn't keep us from getting it, they wanted to charge us a dollar a page to make copies even though we were requesting it in digital format. So there was actually a freedom of information case - I think I was 24 or 25 at the time - that I kind of organized for the small company I was working for. And we stuck with it, and eventually won the case. And the result of that case is all these companies you know of today can request public records from municipalities. And the municipalities have to provide it back to them, reasonably, in the format requested and at the cost to reproduce instead of this dollar-a-page concept which was from Xerox machines and things that people don't even-- I don't even know how many listens know that-- BRIAN 09:09 Many moons ago. PETE 09:10 Yeah. So the project, unfortunately, failed, that we were doing - but we stuck with it - because of the delays in collecting the data, but we kept with the case. And as a result of that case, we were then able to launch a bunch of new businesses, I guess, out of it, because we suddenly had access to this data, and became kind of the middle man between all these municipalities and the ultimate consumers of this data which were end users online. And in between those two points, there was us and then whatever the web property was, like Realtor.com or what have you, that was actually built into the user experience. So that process, that was like a innovation in business process, I guess, more than an analytics, but that kind of opened the doors to what we did later. And what we did later was I think change, again, the way that people perceived looking for real estate online by providing information about things other than the property. So what I mean by that is if you go back in time quite a bit, the only information you could find about a home that was for sale was typically handed to you on a piece of paper from a real estate agent. And it was really just the information about the home, some pictures and bedrooms and bathrooms and things like that. But when real estate went online and you were able to start searching for homes by zip code and price range and things like that, there a lot of other opportunities opened up that people didn't really see. And I think we were the first folks to see that and those were to allow people to look for areas first that maybe met their criteria like they were looking for a certain level of school system or wanted to make sure it met a safety standard or it was pet-friendly or something like that. PETE 10:58 So we actually pitched Realtor.com which was really almost the only player in the game nationally for real estate search, and we said, "Hey, you guys have this great search where you can type in a zip code and see all the homes for sale. But if I'm moving from New York to LA, I don't know any of the zip codes. I don't even know where to look. How would I even know what neighborhoods I want to live in or think about looking at for properties in?" So we pitched them this idea where we put together a whole bunch of demographics and school and crime and other statistics for them so they could create community pages. And there were some forward-looking product managers over there at the time and they bid on it. And we put this product together, and they turned it on. And they found that 30% of their traffic actually went through looking for areas first and then looking at homes in the areas, so it was an immediate sort of hit. And, of course, we had taken a risk and they certainly had also in putting it out there. No one had tried it before, but it was a hit. And then that, of course, led to us being able to grow that business in a lot of directions. BRIAN 12:04 That's incredible and I think now it's amazing. A lot of these things are just available. It's just an API. Right? You can plug [all tricks?] into an API and hit run and it will bring all that information in. That's a long way from making a request on a per-page cost. Right? PETE 12:19 Yeah. The landscape has really completely changed. Back then we were doing things like collecting information that didn't exist at the zip-code level and modeling it ourselves. That was a big part of what did. We did that with neighborhoods eventually and other things. But that business evolved in some interesting ways because we were delivering this data in bulk to people, and they were building their websites off of it. And eventually, almost any website you could think of was pulling that content from us. The Realogy brands, Trulia, and Redfin were customers for a while, all the big brokers. But what we found was that there were a lot of limitations. And when it comes to innovating things, I feel that innovation usually comes from a couple place, like pretty different places. But one of those is someone literally just subconsciously connects the dots between things and has this great idea that no one thought of. I think those opportunities happen infrequently, but when they do you get things that nobody's even ever thought of before. But I think what's more likely to happen is that innovation comes from frustration. So whether it's you being frustrated with some sort of process or a client or yours or customers, if you have enough kind of information surrounding the problem, then you can often look at these frustrations and come up with some interesting solutions. So in a way that searching by communities was an example of that. It's was frustrating to try to figure out where around LA or Seattle or which suburbs or which neighborhoods you might want to live in. And that led, ultimately, to some other frustrations at the time that led to some new products. PETE 14:11 There was a website out at the time, and I think it still exists. Though it's certainly changed and is much better now, Bert Sperling's Best Places to Live. And I remember going there. I lived in New York and I was thinking about moving, and I was like, "Let's see if it can recommend some cities for me." Right? And I plugged in some information and it turned out that no matter what else I put in-- if I entered in that I wanted good weather like warm, sunny weather, Tucson was always number one [laughter] on the list. Like it didn't-- BRIAN 14:41 Imagine that. PETE 14:42 Yeah. It didn't matter if I wanted the other things: schools or crime, culture, whatever. It didn't matter. So as a data guy, I was thinking there's a better way to do this, right, and there definitely is. Right? And that ultimately led to developing a new product that we went out and pitched again to all these real estate brokerages and websites. We called it lifestyle search which again is now, today, not a new concept. But at the time, back in about 2008, it was, where you would ask the customer, "Hey, what's important to you?" or, "What are your preferences, and how important is that preference?" So you kind of ended up with a multivariant search where you were weighting the inputs and then providing a best-match result instead of this kind of hard-filter, gated process which is like that old website that I had tried. So we built this, and it leveraged everything the company did. Right? So we had all this geographic data and a geographic model. We did all the spatial analysis to say, "Hey, these neighborhoods are in these cities, and they all roll up in this way. And crime statistics are available at this geographic level, but we need to model them down to block groups," for example. So it was really a big analytics project and then on top of that, to build the weight-sort mechanism to provide best results was a fun challenge also. And it ended up in a product that was typically implemented where you put up 10 or 12 different topics on the web page and let customers sort of drag a slider from most to least important. So you put up something like, "Do you want to have good schools? How important is that to you?" And you think it'd be important to everybody, but if you don't have any kids, often good schools go hand in hand with higher taxes. So everything sort of becomes sort of relative. And we did some learning there which is what I think good analytics is all about, which is initially we just put the product out and we found that everybody who had kids always put great schools, low crime, and low home prices. Right? PETE 16:57 So they weren't actually differentiating among the things that they wanted. What was most important? And we found that we had to sort of limit the number of points they were able to distribute from a "how important it is this" and force them to differentiate, force them to prioritize. And if you can get your users to prioritize their input for you, then it becomes much easier to provide good results. You could always provide results even if they're not prioritizing, but those results aren't going to resonate in the same. Even in cities that have relatively poor school systems, there are always some schools that are better than others. Right? So there are always some areas that have lower crime or higher crime than others even if it's a safe city. So by having them provide these preferences, you can factor that into an algorithm and provide better results and then, of course, test that. So that was another project that was kind of born out of frustration. And I think in the real estate space, that's kind of been the history of what we did with this lifestyle search engine was one. We put out the first search API for property search which allowed our customers to focus on selling homes instead of building all kinds of interesting search which could include these lifestyle characteristics but also things like drive time, commute time, whether it was in a certain neighborhood, and then factoring in all the different demographics and community data that we had. To build that yourself, you'd have to gather all that information yourself, organize it all, then kind of become expert in how to build the search. And a real estate brokerage, that's not their business. So they were very frustrated. They wanted all the data but couldn't figure out how to leverage it. So we heard the frustration and then built a solution for them from a business-to-business perspective. BRIAN 18:58 Yeah. I love what you're saying about the frustration piece because I often think that the difference between someone who's a complainer or an innovator is simply how they channel their frustrations. Right? PETE 19:08 Yep. BRIAN 19:09 Like some people will just simply kind of take it, so to speak, and just, "Hey, that's just the way the world is. I'll complain about it. Every once in a while, I'll grumble, whatever." Some people get frankly pissed off enough to go change the world or try to come up with a solution. So I think you're right on there. I think that makes a lot of sense. When it comes to sort of these listicles of the best places to live or how you use the data, I think one thing that's really interesting there as well is sort of the full scope. You're describing this project or projects, and you're using almost every facet of these analytical products. Right? You have spatial. You have mapping data. You have all kinds of things, the full scope there. Maybe talk about that a little bit. What's it like when you're trying to solve one of these problems, but you're ingesting and using such crazily different disparate data sets. PETE 20:12 There are definitely some challenges there. I find them to be fun challenges, but I guess that's maybe what's makes me good at solving these types of problems. So the best places to live in these lists is kind of an interesting side avenue to what I've been doing over the years. Money magazine was actually the company that we started. I started with two other gentlemen, and eventually, that company grew to about 50 people. The very first client was Money magazine. They had been a customer of a previous company I'd worked at. And after that company went out of business, oddly enough, and not in any way related to 9/11 but in New York, they went out of business on 9/11. Money magazine called up a month or two later and said, "Hey, we're ready to run our story this year." And we said but, "Okay, but the company doesn't exist." And they said, "Okay, but we're ready to run the story this year." So we put our heads together and said, "Hey, how can we do this?" And they ended up becoming our first customer. And over time we also worked kind of the same sort of stories for other publications, U.S. News, Essence, Forbes, Fortune, a whole bunch of them, Family Circle. PETE 21:32 And, of course, if you're doing the same type of product for a number of customers, you start to run into questions about how do we differentiate the product. And we actually, I think, had a pretty straightforward but good process for that which was we tried to understand our client, tried to understand the problem a little bit better. If you think about one of these lists, it's really what are the best places to live for the readership of that publication. That's what the publication's looking for. They're not looking for some totally objective or impartial result. And it's hard to even come up with that because the best place to live for one person is not the best place for another. So you have to start somewhere, and then once you know where you're starting, kind of build the process around that, in that context as impartial and objective. So what we would do is we would ask them for their customer marketing profile. And then we would look at that, and we would try to understand "here's marketing speak for these people". How does that translate into real data that's available to us out in the world? And that information becomes different depending on what that marketing profile was. PETE 22:44 In the case of Money magazine-- and I still work with them, so you should all definitely go out and get their Best Places to Live feature that comes out in the fall at some point, we just finished up the data work for it. We put about 200 different data sources into that, and it's all sorts of things from home prices and current market statistics to crime to availability of conveniences nearby to commute times, you name it. And it was a pretty big challenge bringing these things in. It's become a little easier over the years as data sources have become better organized. But in reality what you have is a problem where you have data that comes in to you at all different levels, state, county. Point data meaning like these businesses are really addresses, schools are addresses. School-district boundaries don't match up with city boundaries all the time, certainly not with zip code boundaries. Lots of data is only available for zip codes. Weather is related to weather reporting stations. So it's really a spatial matching problem. PETE 23:53 And when we first started using it, this was before Alteryx came out with a product, we were using Oracle Spatial and [Android PiP?] [inaudible] map info product to essentially do a whole bunch of spatial analytics on the data to be able to aggregate to boundaries and to proximity searching, radius and buffer counts, all these things to essentially pull this data together so that I could tell you what's the convenience score, or an index for this small town in the middle of Iowa. And eventually, we redid that process. So I redid that process entirely using Alteryx tools because it's really well suited to that with all the spatial tool set involved. But the problems all remain the same. And organizing that data and pulling it together and creating something intelligent out of it is an interesting challenge. PETE 24:49 So I'm working with editors and writers at the publications, and I could output to them a spreadsheet full of data but they don't necessarily understand the differences in the types of data points. Like their scale, what would be the differences in scale? And how should that impact what they're doing? Because they're taking what I'm giving them, and they're trying to create an algorithm that for each place combines all this data and scores it, and says, "Hey, this is going to be in my top 100 or top 10 list, and this place isn't." And then, of course, they have some editorial processes on top of that. They go out and they interview people and do other things which you can't capture everything in the data, of course. So in addition to combining the data together, then there's a variety of work and modeling work around trying to create scores that they can then combine in a way that's meaningful and don't overly value one data point over another. So a simple example there is if you're dealing with home prices, they're typically in hundreds of thousands. If you're dealing with crime index score, they're typically in the hundreds or the double digits. And if you simply added them up and average them or something like that, you wouldn't really be doing justice to either of the data points involved. So that's a process. I've been doing it for about 20 years with them. Their teams change quite a bit over time, but the focus of the stories really remain the same with Money magazine, in particular, and has really been a great success for them. BRIAN 26:30 That's really cool. I think we're a little-- I don't know. I feel like these days there's so many - I'm using our quotes here - lists or listicles out there on the web about things. And I think a lot of them feel like they're just kind of thrown together or someone just made some editorial decisions about what the list should be rather than fact-based decisions. And it's really cool to hear, in this particular case, all of the great lengths and processes and things that you're doing to really get to what is a objective by the data cut of what actually would be a great place to live if I was in Money's prime market, right, for who they market to. PETE 27:15 Yeah. It's a fun process. And I've had the benefit, I think, of being able to help influence them to take into consideration new data sets over time. Obviously, the project's gotten a little bit more complex over the years, but recently they focused more on things like diversity. There's a lot of great data out there on diversity, not just racial diversity, but cultural diversity around things like food, economic diversity, diversity of recreational opportunities and things like that, that can really shed some color, shed some light on how an area really feels when you're there. So that was one thing we've added, some tax burden information, and other things like that, that really seem to resonate with folks. I think we try to stay a little bit away from some of the political questions because they don't always align in any way with any of the other data that we're working with for this type of-- for this type of story. But outside of that, it's nice to be able to have an influence on what are the things that they're trying to consider. And kind of a funny story is every year-- because this is a data-driven story, every year I get some calls and emails - because I guess I've been doing this for so long that my name's out there a little bit - from towns that want to be in the list. Right? So they'll say, "Hey--" BRIAN 28:52 Tucson. PETE 28:53 Yeah. Tucson. So sometimes we focus on big cities or small towns, so there's sort of this natural cut about whether a place can be a candidate. But one town in Nebraska, every year they made a-- City Council made a resolution, a 10-year plan. They want to get on the Money magazine Best Places to Live list. And they're very persistent in calling me up and saying what do we need to do to make the list. Right? So I, of course, don't divulge any of the secret's source, so to speak. I just say, you just have to be a better place to live. I mean that's really it. BRIAN 29:31 Yeah. Make sure the sun shines more. Right? PETE 29:33 Yep. BRIAN 29:34 Cool. So one thing I'm thinking about here too, we've been talking about housing and the data and all of this for some time. You mentioned before you've been doing this for 20-years-plus. What indicators, if any-- I guess I'm making an assumption - you can correct me if I'm wrong - that at some point along the way here you were doing this work, looking at this data, diving very deep into it. Were you able to detect anything around the housing bubble or the housing crisis? And if so, maybe tell us a little bit about that like when did that become apparent to you, is it did. And was it ahead of the rest of the market? PETE 30:22 Great question. And, yes. We did see some things in the data. So back in about 2007, 2008, we were processing-- pretty much all the public records in the country were coming through our office one way or the other. So we were able to look at-- and in fact, we had products built around things like home-price trends and volume on the market and things of that nature. So we were able to look at those things and we started to see some, I think, things that seemed a little odd in the data at the time. Now, keeping in mind our business was not about making predictions or anything else like that. In fact, our customers were pretty much people who were selling properties. So this wasn't necessarily the type of thing that we were thinking about building a product out of, but just bring naturally curious, a gentleman I worked with who was actually our director of data operations, Joe Rugolo, and I - and he's moved on to some other great things - started looking at some of our data sets and realized that there were some pretty interesting trends going on that ultimately, as you look back on it, sordidly related to the housing bubble. And we ended up creating what we called a distressed area index which we did at the block-group level. And we originally did this because we thought that it would actually be a great informational tool for government. We were at the time, as a company, trying to see how our data could be useful to people who are making policy or who were trying to figure out what were the areas that needed more funds for something or more affordable housing and things like that. PETE 32:07 So we were looking at things like what was the ratio between 30-year-fixed mortgages and the adjustable-rate mortgages, the three and five-year ones. That was so popular back then. And when those things were coming due for their adjustment and a bunch of other data points - residential turnover rates, fair market rent - and then some sort of maybe negative indicators like employment or job growth or the lack of that, mortgage delinquency rates which were actually available publicly at the zip-code level, you don't get information about the individuals, but you can see things like the number of people who are behind by a certain number of months. So there's a lot of data out there and, of course, the fun part is figuring out how to put it together. So we put together this index, and it just spit out all these numbers. Right? And it was hard to make sense of them, so we used mapping software and built some heat maps around block groups. And we did it just for a handful of cities, at the time, New York, Los Vegas, and maybe Los Angeles, if I remember correctly. And it sort of would light up the city, here these areas seem like they're at high risk, right, or relatively more distressed based on these data points and others. PETE 33:30 And in retrospect, looking back on it several years later, I think that that model actually did a pretty good job of predicting the areas that were likely to have foreclosures. So it was a project that we actually presented at the Inman Connect Conference which is a real estate technology show here. It was in New York in 2008 or 2009. And there was a panel about "What do you predict's going to happen to the real estate market in the coming year." And I remember sitting up there with a whole bunch of real financial analysts and industry experts. And, for the most part, they were all predicting kind of rosy colors because it was sort of before the burst happened. BRIAN 34:13 Sure. PETE 34:13 And then I remember putting up my slides and saying, "Well, we're seeing some high levels of potential housing distress in these areas," and getting a lot of questions about it. It was definitely a fun panel, and I was kind of the outsider in it being much more of a-- less of an industry person and much more just of the data and analytics person. I wouldn't say we predicted things perfectly or I necessarily would have felt comfortable with banks using our index to evaluate mortgage risk or something like that, but the trends were there. They were definitely there if you were looking for them and knew kind of how to look. And the fun part was really taking data that wasn't so directly related, things like employment or housing starts, things that weren't so directly related to actual mortgage volume and mortgage delinquency, and realizing that they correlated very directly with some of the issues that were going to pop up. BRIAN 35:14 Sure. Yeah. Clearly, the economics of that situation were far-reaching beyond just simply the house mortgage situation. And it is interesting looking back on that now. I just went back and watched The Big Short again the other day, loved that movie, and it is interesting how most of the people that were "right" to your point, not all the predictions were spot on, but directionally, the people that were "right" during that period of time and were often getting laughed off of stage at these conferences and things were the ones that were saying, "Hey, we're seeing things in the data," and the other side were people that either had too big of a vested interested to kind of say the truth or what they'd saw happening or they honestly, because they weren't data-driven, they were going with their gut. And they say, "Hey, look. The market's great. We're selling houses like crazy. It's all good. I don't know what you're talking about." Right? So I think that event, among others, has definitely contributed towards what we're seeing now. Here we are 10 years later with this move towards data-driven decision making and, "Hey, let's not ignore the signs that we ignored last time," and hopefully that bodes well for the future. PETE 36:32 Right. Yeah. Maybe it points a little bit back to that whole idea of folklore and mythology. One group of people were believing something because they had a vested interested in it or it's the way it had always been. And then you sort of had new information coming in and it was kind of hard to spot the new information or to realize that maybe it's something you should be looking at. But for anyone who's out there listening and trying to figure out how to communicate data like this that might be complex or might be very localized or other things like that, the real key for us and what resonated with people was ultimately putting it up on a map, right, and having the right tools available to be able to do that sort of thing because most people can't look at the spreadsheet and see the shape of the data. I think that's something I do. And it took me a while to realize that other people don't do that so easily. And so if you can put it into a visualization that other people can relate to, that can take the message you're trying to get across and actually have it resonate with people instead of everyone sat there in a meeting with somebody where you're trying to present something and people's eyes get a little glassy. Usually, a picture does tell a thousand words or a thousand numbers in this case. BRIAN 37:55 Yep. All right. So switching gears just ever so slightly here. A few minutes ago you were talking about starting a business and growing that from scratch up to 50-plus people, maybe talk a little bit about that of being a business owner, seeing it grow. What is the communication like when you have this big idea and you're trying to translate that into a viable operation on which other people's lives are also dependent for their income and things like that? PETE 38:30 Sure. Yeah. It's a hard kind of road to navigate in some ways. When you're starting off, you have a lot of flexibility in many ways, much more so that as you grow because you don't really have a lot to lose. If it's just you, I mean, sure you have to fend for yourself and have some sort of income, but you don't have people who are clients relying on you or what we call technical depth that you have to support and things like that. So starting off, it was great. We threw a lot of stuff up against the wall and kind of saw what stuck. And of course, we tried to make informed decisions about what we were putting our efforts into. From my previous experience, I had a lot of domain expertise and some great contacts in sort of the real estate information space which was good. I think it's hard, in many cases, as somebody whose maybe on the technical side instead of the relationship side of product development or ideas, right, coming up with ideas, to be able to validate what you're thinking about if you don't have those relationships in place. And my other partner certainly brought that to the table. I was really the product and technology and data person when we started the company. So that was great. PETE 39:56 And I think I have the skill set to be able to communicate between the technology and the business side, which is also often a challenge for folks. Some people are great at it. Some people aren't. If you're a great-idea person but not great at communicating it, I would definitely recommend making sure you hire for strengths and bring someone on that can really be compatible with you but also add that dimension to what you're doing. That's actually one of the growth strategies that we really employed and I think is really important also when you're dealing with data and analytics is not everybody is going to be good at all aspects of any role that you're trying to put them in. So recognizing what they're great at and then giving them leeway in that area and actually trying to force as much of their bandwidth into the things that they're good at, instead of saying, "Well, they're really not good at recording or doing documentation, so they have to focus on that," and then making them miserable by forcing them to spend a lot of time on something that they're not good at. I think that's a mistake that a lot of folks make. So we always try to do what we call managing to strength. Right? So manage to people's strengths. PETE 41:14 And on the data analytics side, some people are great at modeling the data, understanding and looking forward, how an application might leverage a data set, or leverage the structures and being able to be forward-thinking enough to future-proof, at least to some extent, what you're doing. Other people don't have that skill, but they might be amazing, absolutely amazing, at taking the output of your systems and understanding what's going on and, right, actually performing the analytics and then interpreting them. So that's one area, I think, as you're growing is to recognize, first of all, what you're good at and what you're not good at, and then make sure that you hire and manage to people's strengths instead of to their weaknesses. PETE 42:01 The other thing around, I guess, growing and being able to do that in a kind of a measured sort of way is to stay in touch and in contact with your customers. Right? So creating a market is really difficult for a product. To be able to do that from scratch, usually, there have to have a huge PR or marketing budget or somehow have a big lucky break. Right? And you see these companies that they come massively funded well ahead of their revenue as a result of that. That wasn't our situation. So what we did have were these relationships with other companies that had market share. And so originally we tried to put out our own products for consumers, and people loved them, but nobody ever was using them because we weren't out in front of people. We weren't promoting them. We didn't have the dollars behind that. So instead of going down that path-- our path was really around empowering our customers, and that worked really well for us, and certainly, at the point in time when we were all very focused on that business. We were the first people in the space. We stayed ahead of the competitors. We innovated a lot of products. Our customers trusted us, so that allowed us to get them to try new products like that lifestyle search engine where people would find like the neighborhoods in New York that best matched up with whatever they looking for. That was brand new, so when Coldwell Banker put that up on their website, they were taking a leap of faith in us that we had built a good product and that it was going to resonate with their customers. But that was a lot easier for us than, say, building a marketplace from scratch, if that makes sense. BRIAN 43:47 Yeah. Absolutely. Okay. Anything else you wanted to talk about today? Any other pearls of wisdom we should be sharing with the listeners? PETE 43:57 Well, maybe just another quick story about ideas coming out of frustration. And there's a guy I worked with who-- Joe Fernandez, is a great success story, really smart guy, worked with us as a product manager, but he had jaw surgery at some point. And he couldn't talk for like two months. Right? This is, again, back in the late 2000s. And he was trying to figure out things like what music should I listen to, what restaurant should I order food in from, or what movies should I go to? And this was sort of really before the heyday of a lot of social networks and things like that now. And he came up with the idea that ultimately evolved into the company Klout which I don't know if you heard of. And they're gone now also, but they were the first social influence measurement company. And he came to me and we sat down, and he was kind of scribbling on some paper because he couldn't talk. But he was going over this idea, is like how can we find-- how can you find out who-- not who like something but who could you trust their opinion and then make decisions for yourself about what they liked. And that was sort of the whole genesis of this concept of social influence in the marketplace and there it was all born out of frustration. Right? And he built a really big company with several hundred people and then eventually moved on. It was very successful from there. PETE 45:37 But I think the lesson from that is there is an idea there that he came up with because there was a very personal frustration and then the challenge, of course, was trying to figure out from a technology and analytics perspective is there a way to solve it. Right? And he went in and pitched folks like Twitter to open up their API more to give him access so that he could do these analytics on not just what people were posting but on their reach, their response rates people were getting, and all the other things that differentiate what I would call sort of fire hydrant influencer, who's just putting everything out there to somebody who maybe isn't as prolific but has a lot of engagement in what they're doing and ultimately a lot more influence in what other people might be thinking. So innovation, again, I think from frustration is a great place to start. And if you've been frustrated by 10 or 100 different things probably in there there's a couple of great ideas. BRIAN 46:41 I love that you just told Joe's story. I've heard that story a number of times. And part of the reason why is, I actually, briefly overlapped with Joe at my last company, Lithium Technologies, because they had acquired Klout at some point. So I'm very familiar with Joe and this story, and he's a great guy. And what he's doing today, for our listeners, because I think this is just another-- he identified yet another frustration. And I think it's a-- you'd have to ask him. But my assumption is that the data heavily told him this too. His new company is called Joymode. And what they're doing is they've picked up on this trend that people want access to things but don't necessarily want to make the investment to own them. And so what Joymode does-- they're based out of LA, here, just a little bit north from where I'm sitting right now. And what they do is you get the app or you go their website and you basically say, "I'd love to have a backyard barbecue tonight with a big movie screen and an inflatable flamingo and whatever else, right, but I don't want to go to the store and buy all this stuff, the oversized Jenga game and whatever." And what they'll do is they'll just rent it to you basically. So you say, "Hey, I'd like this package and that stuff," and they bring it to your house, you set it up, you have your beautiful event, and then it gets returned to them. So I think another great example of sort of reading the tea leaves, looking around the corner a little bit and building a business based on some frustrations of, "I want to have these nice things, but I don't necessarily want them forever sitting in my garage collecting dust." So I mean I'm glad you told that story. Joe's a great guy. And we'll put the links in the show notes to his story and his Twitter. He's great on Twitter as well, so. PETE 48:24 Yeah. I didn't realize that you had overlapped with him. I actually spent some time with him out at the Inspire show in Anaheim about a month ago or whatever that was. And so kind of an interesting thing about his new business, his old business was really a pure analytics business, the Klout. Right? His new business has a lot of parallels to what we do over at Zeel in the on-demand, in-home massage space, in that, it's really a logistics question or a logistics challenge. So if you think about it, he's delivering and then picking up product to thousands of people every day. Right? And he has to figure out what to store, how to store it, how to deliver it, all that sort of stuff, so there's some pretty similar problems there which I found interesting. But the original idea for Joymode, if I remember in my first conversation with him about it, was around what you just said. Someone's throwing a party. Nobody is going to buy and own a bounce house if you buy it and use it three times when their kids are little and then throw it away. That's too big an investment, so I think he was thinking about those things. PETE 49:34 And it turns out because he's so analytics minded and focused and has really concentrated his business, in part on the data collection side to understand how people are using the products, it turns out that what people really used his service for the most are things like waffle irons and vacuum cleaners, right, and tools and things that you probably have a whole bunch of sitting in your basement. And they don't seem like the type of thing that would work in this kind of sharing "I don't need to own it mode" but actually really do. How many times a year do you actually make waffles? How many times a year do you need a wet vac, or whatever else it is. So his business has evolved and pivoted from his original idea also which is-- I only bring up because that's super important is to understand when to listen to the information that you're gathering and make a change in direction especially with a small business. If you're a big business maybe you can split off a new warranty thing or do something else like that. But when you're a small business and you're trying to figure out your part to go down, analytics can really be your friend if you're collecting good information and you learn how to intuit direction from it, and I think Joe is a master at that. BRIAN 50:51 All right. So let's wrap up with our Community Picks. What has been interesting to you lately that you'd like to point people to? PETE 50:58 Well, a couple of things, so one is that I'm kind of an avid reader, and I really think it's important to read, not just things in your sort of space, but things outside of it. So for me, probably as with a lot of kind of data geeks out there, I like to read fantasies, science fiction sort of stuff. But there's a great service for finding new books and finding it at, I guess, a discounted rate called BookBub, and every day I get an email from them. It's got a list of books in it that the authors or the service that's selling it have discounted for that day. And sometimes they're even free. Right? And you pick your categories that you might be interested in. In addition to the science fiction side of things, entrepreneurship, small business, applications, things like that are in my list. And I don't always find something that I want, but it's a pretty good way to build up a library at a low investment. So that's a really cool service especially if you're selective in the topics that you let them send you stuff about. So you don't get too much and it stays on point for you. So that one's pretty cool. PETE 52:15 Another thing I listen to a bunch of podcasts. I'm one of those commuters on the train who's got the headphones on, and I'm not listening to music mostly. I'm listening to podcasts. One of the ones that I love and is actually a great way to get an introduction to some analytics and data science topics is Data Skeptic podcast. Kyle and Linh Da run it, and they do a couple of things. They do interviews kind of along the lines of this one on data science topics, but they also do a whole bunch of what they call mini-episodes. And they're a husband and wife team. And Linh Da does not have expertise in the analytics and data science space. So typically they'll bring up a topic like clustering or the curse of dimensionality or spatial analytics and Kyle will sort of introduce it as a problem to somebody who's not experienced in that area. So it's a great intro for people who don't have a background in things. And I've often gone and listened to those mini-episodes when I'm thinking about a new model or something I don't have experience with before, and it can be pretty helpful. And I also use that in my training. I do a lot of training of folks in the analytics space at the company I work with, and this is a great resource for them. So those are two. PETE 53:40 And I guess the third one that I had mentioned which is kind of a shout out right back at you, Brian, is the Alteryx Community and the training area have really improved over the years. Right? And right now I think there's a lot of great resources out there. So one of the things I'm working on now at Zeel is distributing kind of analytics know-how and tool sets out into some of the other departments. And so we've got a group of folks from marketing and product and even the sales team who are starting to use some of the tool set. And not only can they self-serve a lot of resource there, but I've actually found, from the perspective of holding kind of training and information sharing sessions, that having them do some of the weekly challenges that are up there-- and there's like 120 of them now, so you're not going to run out any time soon, having them do that and then talking through how there are different approaches to solving the same problem and why one might have chosen one approach over another is really useful for learning. Right? You often get an introduction to a new tool because there's several hundred on the tool palette these days, and I don't use them all. Right? So sometimes someone uses something I've never used, and I get to learn from it too. So I think that's another great resource that if people aren't using, I definitely recommend it. BRIAN 55:05 Yeah. Those are great picks. Thanks for the shout out to Alteryx Academy there. Like you, I'm also a podcast listener, so I guess I'll follow your lead here and give a couple that I think might be interesting to our audience. One is Podcast Your Data, and that's our friends over at InterWorks. They're actually a partner of Alteryx. But they have a great show where they put it out twice a months and they talk about different analytical challenges and software and platforms and things like that. So that's a good one. I also get a lot of value out of Freakonomics Radio. I don't know if you've ever listened to that one, but Stephen Dubner who was actually one of our keynote speakers at Inspire a couple years back, they sort of gloss over analytics specifically as a topic. But you can tell that a lot of the things that they say and talk about are rooted in that, and how they're using different data sets and data points to look at the sort of economics of any given topic on any given week. And they're pretty prolific. It was just like show, after show, after show that gets published. So that's another one where you'll never run out of content. And it's so widely varied what they talk about that I think you can probably cruise through the list and find something that's immediately important or interesting to you. So those are two good ones. And then also, like I said, we'll throw Joymode in there because I think it's a fantastic service, and Joe's a great guy, and it would be great for everybody to go check that one out. PETE 56:31 Yeah. They're great, and look, there's so much great stuff out there. It's too bad that there's not unlimited time to go and listen to those things. The Freakonomics Radio is a great one also. And I think that Joe and Joymode are getting ready to do some expansion. So hopefully some of your listeners outside of LA will be able to check out the service some time soon too. BRIAN 56:55 Awesome. All right, Pete. Well, hey, this has been fantastic. You are a crazy, insightful guy, and I'm so, just, trilled to have the opportunity to talk to you. So thanks for your time. Thanks for being on. Thanks for sharing with us. Where do people go to get more of you? Are you a Twitter guy? Are you blowing up the Twittersphere? Where do they go to find more Pete? PETE 57:17 Well, first, thanks. And it's great to be part of the community here. So you can find me in the Alteryx Community, of course. I am on Twitter. I'm definitely not prolific there, though I would absolutely respond to folks, and I should probably be doing more. My consulting company is called WitLytic, and I'll send you the link to it because it's a little hard to spell. So you can find me there also, but my handle almost everywhere is Peter.Goldey, so I'm not that hard to track down. And I love starting up conversations with folks especially if you're thinking about some new idea. You're stuck. You're trying to validate something. Talking through a challenge is really the most fun for me. Execution is the most important thing to getting your business going but brainstorming it is sometimes the fun part. BRIAN 58:11 Yes. And, as always, we will put all of the links to everything we've talks about and ways to get a hold of Pete in our show notes at Community.Alteryx.com/podcast. So thanks so much, Pete, for being on. Again, I just can't think you enough. This has been great, and we will talk to you again soon. PETE 58:30 Awesome. Thanks so much, Brian. [music]

This episode of Alter Everything was produced by Maddie Johannsen (@MaddieJ).

Alter Everything Podcast

Podcast Guide

13: Innovation comes from frustration

Guests

Topics

Community Picks

Transcript