This site uses different types of cookies, including analytics and functional cookies (its own and from other sites). To change your cookie settings or find out more, click here. If you continue browsing our website, you accept these cookies.
In this episode, Alteryx legend Criston Schellenger guest hosts Software Engineering Manager Jarrod Pozarnsky and Alteryx ACE, Jesse Clark. They have a chat about ideas: where ideas come from, and how they come into fruition. As their "case study" when thinking about the anatomy of an idea, the group looks at the Alteryx Community ideas forum, and how those ideas make it - or don't make it - into development on the Alteryx platform.
Special thanks to @andyuttley, one of our winners of the 2020 theme music competition, for the awesome theme music track for this episode.
Welcome to Alter Everything, a podcast about data science and analytics culture. My name is Criston Schellenger and I'll be your host. I'm a member of the Alteryx community team and have been at the company for a long time. In fact, this week is my five-year anniversary. I wanted to chat with Alteryx ACE, Jesse Clark and Alteryx OG, Jarrod Pozarnsky about ideas. Where ideas come from and how they come to fruition. As our case study when thinking about the anatomy of an idea, we'll look at Alteryx community ideas forum and how those ideas make it or don't make it into development on the Alteryx platform. All right.
Sure, why not?
Let's get started. I'll have you guys introduce yourselves. Jarrod, why don't you start?
Sure. My name's Jarrod. I'm @JPoz on the community.
Yeah, you are.
I've been with Alteryx for seven years and I joined the Designer team, which was the third development team Alteryx had at the time.
After Engine and--
Six and a half years ago, the Designer team was created. And it was created to basically support the UI for the Designer desktop, not to be confused with the Engine side of it where the data processing happens, so.
Right on. Jesse?
I'm Jesse Clark. I am an Alteryx ACE. I work for a not-for-profit health insurance company in Massachusetts called Fallon Health.
Funnily enough, and I am very excited to share this, Jesse is actually a new Alteryx Associate. At the time of the recording, he was a member of our ACE Program but he couldn't get enough Alteryx and decided to join our ranks.
I have been using Alteryx since the tail end of 2015. I have been active in the community pretty much ever since then. And I was the first Alteryx Expert, now 1 of I think 15 or so in the world and I'm also the first, and as of time of this recording, only Alteryx Automation Master. So I've been doing things with Alteryx for a while. I kind of use Alteryx Server and Designer and I do our administration and our best practices and our deployment strategy for all of that stuff at Fallon. Really started with it at its infancy in the organization and have kind of brought it through to its current maturity.
Right on. What version did you start on?
10.0 is the first version that I personally had installed. And then I convinced the org to do a really quick upgrade to 10.1, I think, because it added browse anywhere as a concept and that was a key enough feature that I said, "It's worth it. We need to do this." And then we didn't upgrade for a while but--
Yeah. I started on 8.1, where everything was on the side and we all worked vertically, like we should and--
Oh, that's a strong opinion.
It is a strong opinion and I stand by it. All right. But today we're going to talk about ideas. So we have an ideas page on our customer community and according to the community team, the community platform ideas page allows Alteryx to see what people need, what is trending, and how analytics continues to evolve. The community ideas board brings the customer and Alteryx engineers closer together to continue to move the platform forward with the needs and imagination of the users. So, in other words, think of the apps and programs that you use every day. Have you ever thought of an idea to make your experience easier, faster, or better overall? Or maybe you've uploaded someone else's idea. That is exactly what our community ideas board is. To hear from the users who are in Alteryx day in and day out.
It's a great spot for the people that are day to day using our tool to say, "You know what? It would be really great if it did this," or, "This is really annoying. Why don't you fix this?" And then with the starring you can kind of get a sense of, "This is a feature that we really should have that we don't." And so yeah, I think it's pretty accurate.
I would generally agree with that as well. I think the other use case for it, which is kind of a weird one, is that it helps me as a customer to see what things other customers have encountered or experienced. And that not only lets me say, "Okay, that's something I need to plan for and work around," or, "Wow, that's a really good idea. I should kind of put my voice behind this and see what I can do." But it also helps to kind of foster some direct communication. There have a bunch of cases where someone posts an idea and I kind of go, "Well, hey, I have a workaround for you. Have you tried this?" And some of those have been really helpful. Or there have been times where I've instead kind of had a back and forth with some Alteryx representatives and discussed a particular feature and how to improve kind of the plan for implementing it. And so having that back and forth has been really valuable, I think, to me and to the Alteryx product.
And evidently you're a mind reader because you literally answered my next question about the kind of value it provides to be able to participate in that kind of feedback loop with engineering and PM.
It's incredibly valuable, not only because it means that the direction of the product is moving the way you expect it to but also because it really does help to bridge that communication gap. Obviously, with any kind of communication it's challenging to make sure both sides are hearing the exact same thing but in particular with online communication it's very easy to misinterpret a sentence or something like that. And so having that back and forth once an idea has started and really discussing it more is really helpful. There have even been some ideas where they were put into a status of not planned based on the initial request and through clarification and discussion they were changed to be on the road map or be things that were requiring for the review by Alteryx. And so I think that back and forth has shown its value from that space. And that's really motivating when it happens.
Ah, that's great. It's great that it's motivating because I feel like we get a lot of energy from that and I hope engineering get the same kind of energy. So it's good that we're all motivated by this kind of ideas exchange. Jarrod, do you have an example of a customer suggested feature that was implemented into the Alteryx platform? Project 27 notwithstanding.
Yeah, sure. I mean, all sorts of stuff-- I mean, one of the things people wanted for the longest time was the ability to sort within the browse tool. That one took quite a lot longer than you would think. But there's all sort of-- we've been really trying to put a lot more of those into the products in the last couple of releases and so we're actively-- I mean, I'm actively on the forum. I've read every idea on there. I've gone through that more than once and that's like a hundred pages of 10 ideas each. And so just some of the stuff from the recent release, the ability to convert a tool from one thing to another thing, there were about 4 or 5 ideas that were all related. And so for us, in engineering, it's like we have a piece of code that basically can convert a tool to anything else. And so it was easy to implement all of that very quickly. Let's see, I don't want to give any spoilers yet but-- I'll just look at the last release. It had, I believe-- 33 customer ideas were closed after that. And also it's little ideas we've implemented like seeing the taskbar status while you run a workflow, turning off the little ding. Those are all little ideas we implemented. But even a lot of the bigger features that we've implemented, sometimes you think you have a good idea for Alteryx and then you look in the community ideas forum and it's like, "Oh, well, here's somebody had suggested this three years ago." So I could list a lot of stuff. But, yeah.
So speaking of that, can you shed any light on why some of these things were implemented and why they weren't?
For the recent release?
No, not specifically with recent release but previously--
Just in general.
--in general, why--
What makes some ideas get picked and some ideas not necessarily?
On the scale of--
--difficulty or demand or--
Sure. I mean, there's a lot of different things that go into what features are going into Alteryx. So you have the product management department and they have kind of their list of priorities and things that they want to do. Then you have the customer ideas forum which, in an ideal world, we're pulling some portion of those. Product management is talking to customers regularly so they're hopefully aligned with what the customers want. But we also have-- engineering has, they're like hackathons. We call them Innovation Days. Or we have-- like on my team, I work with a PM that's very open to me saying, "Hey, I saw this on the community." And I come at it from an engineering perspective where I say, "This is actually really easy for us to implement." And he always says, "Oh, yeah. Cool. Let's do that." Or Innovation Days, the hackathon, a lot of customer ideas have been implemented in that and then they go, in a lot of cases, directly onto the road map. One example is the auto recover of files had been suggested in the past. That started as an Innovation Day project where somebody that works here made it so basically, whenever you run Alteryx it auto backs up your file in case something goes wrong. And from there, we started talking about it and it was just a conversation like, "Hey, on the community they've asked for an auto recover type functionality like Excel. Why don't we just take that and finish it out and make it a full-fledged feature?" And so there's a lot of different ways that an idea can come about.
I personally like the ideas that product management has said are not planned. I like to look at those because sometimes they're a lot easier or more important than I think-- it depends on the level of effort but there's been several of those recently that I've actively said, "Hey now, we need to do this. This is really important." And sometime the customers, after you tell them it's not planned, are still posting and saying, "We'd really like to see this." And I think that's really important too because, like I said, there's lots of different things you could do with Alteryx. It's a very vast product. But sometimes the decision to not do something could be revised later and if it's a good idea, it's a good idea, so.
I think it's interesting you bring up Innovation Days because-- for our listeners, that is a quarterly kind of a hackathon where the only rule is it has to be Alteryx related. And so a lot of really cool stuff has come out of Innovation Day. For example, Promote ostensibly was an Innovation Days project. One of our developers met with Yhat and started to build a connector and kind of all went uphill, downhill from there. And, I think, caching also came out an Innovation Days project?
It did. It did.
And so a lot of-- we do point to our ideas forum for Innovation Days ideas but we don't really want to say-- I mean, that's not really the point of Innovation is fixing defects or putting stuff into the product that people are asking for. But a lot of the really awesome features come out of that because it's something that, internally, we're like, "This is something customers would love and they don't even know it yet." Yeah.
One of my favorites was, before we called it Innovation Days, if you used to drag a tool onto the canvas and it was going to auto connect and the tool had multiple anchors, one of my Innovation Days back then was to put that little dot preview.
To show where it was going to [inaudible].
And I did that for the what we called it back then which wasn't Innovation Days. And then I remember--
And I got to tell you, I was there for that and those of us in the company that actually use the software, we lost our minds when you demoed that. We were just, "Oh, my God. Finally." Yeah.
That's an amazing feature [crosstalk].
It was fine. It was just like-- extend from using being like, "Well, which anchor is it going to connect to?" And then I remember Ned, one of the co-founders of Alteryx, came over and said, "Hey, I saw your project and I think that should have always been in there from the beginning." So that felt really good.
So you never really know when somebody might have an idea and it's a small level of effort compared to what you could get out of it. And you never know who's going to have an idea, so.
Yeah. And Jesse, you were saying something similar about that at dinner the other night, where we have all these shiny new features and all this cool stuff but it's kind of these little things that we implement that retain existing customers and make us want to recommend it to everybody.
Right. Absolutely. Things like that feature. I mean, I think that's been there since I started using the product but that's a really nice feature. It's really, really valuable. And there a lot of those little changes and little quality of life improvements that, when they get added to the product, they really revitalize people. It really does add a lot of extra excitement around Alteryx.
Definitely. I mean, that's one way that we internally form ideas. But externally, how do you guys-- I mean, as ACEs, you guys are in the product all the time. How do you guys form ideas?
Form ideas? So that's a great question. A lot of it comes from my own company, not even necessarily through the ACE space, in that I will have users-- at Fallon, I have users kind of scattered throughout the enterprise working on a number of different business cases and business problems. So sometimes someone will come to me with something I've never thought about using Alteryx for before and I'll go, "Okay, here's how we could maybe solve it." And I'll either find a gap or I'll say, "Wow, that was surprisingly difficult. I think there's a way to make this easier." And that's a very key point as far as an idea goes. Sometimes it's a mutual frustration point, pretty specifically where the ACEs in particular might be talking to one another and someone will say, "Hey, has anybody ever encountered this? How do we deal with it?" And then someone else will go, "Oh, I thought I was the only one. I didn't know other people had this problem." And that will spark a conversation and we'll all go, "Oh yeah, that's important." And someone will post an ideas thread and then we'll all kind of just hop in and put some feedback there. Or someone will say, "Here's an idea I had from a while back, a few months ago, whatever and I really think that it's important and I don't think anybody's looked at it yet." And we'll just kind of, as a group, look at it and say, "Oh yeah, that' resonates with me. That is exactly a problem that I'm having today and I think it's more important than people are realizing in the community." So we kind of use that space in order to promote ideas. Some of them aren't even our own. It's just people saying, "Oh, this is a good idea. Do you all agree?" And that's really powerful as a community.
Great. Speaking of being an ACE, in terms of that, can you tell us some of the perks you've had as far as discussing product ideas and providing feedback with engineers and with product management.
So kind of the key piece of all of that is that as an ACE, I'm under an NDA with Alteryx and so that gives a lot more freedom and flexibility with regards to what we can discuss and when.
For us, too. Right? We love being able to run this stuff by you and not have to worry about it.
Right. And so that means that I can ask about very specific features or challenges and have a very pointed and honest and open conversation about what's been planned, what the roadmap looks like for that, whether or not that's something that has been thought of before. And that adds a massive amount of value to that conversation because sometimes-- I mean, I can't give anything away, obviously, but sometimes I have a conversation with someone and they say, "Oh, yeah. I thought about that. And also, here's five steps further into that particular idea than you were thinking about. And we're trying to look at that right now." And I go, "Wow, that's incredible." And that is super rejuvenating, honestly, because it very much makes it so that I get more visibility into all of the great work that the product team is doing, which in a lot of cases is kind of invisible as an external customer. Because all I see are ideas that are out there that I want implemented that aren't necessarily getting implemented as fast as I want which is yesterday. And so it helps a lot to really show that there's this great staff that cares a lot about this, who are really open and interested in this feedback. And that open dialogue is just so important and so valuable.
Do you participate in ACE/DEV Day?
I have participated in the ACE/DEV Day, yes. It was a really cool experience. I don't think I can talk about much of what we worked on but we talked about a number of different things and kind of just--
To start, tell us what that is.
Sure. So ACE/DEV Day was started, I think, this year at Nashville for Inspire. And basically it was a coming together of the minds, almost, of all of the Alteryx ACEs who were in attendance as well as a number of Alteryx devs. And it was kind of just a group of us who came together and discussed how we're feeling about the product, what things we're interested in. What are we excited about? What are we worried about? And then just some general, good old fashioned Alteryx nerdiness as well. And I can't get much more specific than that but it was a very, very powerful and interesting day. And I know that there's another one planned for London and I'm really excited to get a chance to attend that as well.
Great. And I know you've answered this question probably 18 times this week, but for our listeners, if you could have any idea implemented on the Alteryx platform, what would it be?
So I actually have one that's reasonable and one that's absolutely, 100% not.
Oh, oh. I can't wait to hear both of them.
So the reasonable idea is-- caching is really cool and really powerful. It does a lot of good things and there have been a lot of improvements since the feature was initially implemented. But something that it still doesn't do in the current version of the product is allow me to cache further downstream and use the existing cache already. It kind of uncaches the workflow first, then runs through, caches at a certain breakpoint that I specified, and then keeps running. And then I can use my newly cached data. And so a feature that is out on the ideas page that had some amount of traffic, that I am really excited to see, and I think will happen at some point is to actually have it utilize that cache from upstream when it recaches.
Yeah, yeah. No, that one's actually on our radar. I can't give any timeline.
It's a little more difficult than what you think but that's probably the next caching enhancement that I think you'd probably see but I don't know when.
What that looks like. Absolutely. And yeah, and that's always also the interesting thing about that is that from my perspective as an outsider who doesn't know what the Alteryx source code looks like, I think about that and I go, "This is two lines of code that just need to be flipped around." And so it's also nice to know, to an extent, that it is a little bit more complicated than what I think it is.
Oh, actually, a funny little story about that. A month ago - we have a story for that - and I went to my junior developer and I was like, "Here's one that should be easy for you." And three days later, he came back and he's like, "I'm really confused." So then I had my most senior developer look at it. Anyway, we want to get that one. We really want to get that one. I don't want you to lose hope. Apparently it's more complicated than--
Sure. And that's always where there's that interesting piece from a customer perspective versus what a dev is seeing in terms of what's easy or what's hard, even from a product perspective that you had mentioned earlier where something sounds really difficult and then a dev might say, "That's easy. Let's just do that." So my very grandiose, probably never going to happen idea-- so I like Alteryx a lot. I live in Alteryx. I breathe it. I think I understand what the engine does pretty well most of the time. The feature that I individually, very, very selfishly want more than anything on the planet is more control of the engine. I want to be able to tell it exactly what to do in every case. I want to be able to override the engine and say things like, "No, really, I know you're trying to optimize and do a good job with that and I understand. But I want you to run this stream before anything else. I want you to ignore everything you're trying to optimize and just do what I'm telling you to do." I've also kind of jokingly but not so jokingly asked for/pestered a dev once to just let me put my own custom sort algorithm into Alteryx for fun just to see if I can beat it. It's a very specific and very nerdy request but I really would like to just play around with some sorting algorithms in Alteryx.
Let's get him a build, right? Can we just get him a build?
He's on an NDA, right?
Yeah. All right. Jesse and Jarrod, question for both of you. What are your favorite features of the Alteryx platform?
I think being able to build a tool without writing code and then the fact that some of the tools are actually macros. Now, that feature could definitely use an overhaul but it's pretty cool. It's pretty cool.
Yeah. I think that a use case that I actually only recently really found the true power of is I've started to use Alteryx for very basic application mock-ups because as it turns out, you don't need a workflow behind any of the analytic app tools. So I can build an entire interface without anything behind it and show that mock-up to a user, have them click around on it, make sure it makes sense to them, all without anything behind it. And then later on, I can just build a workflow and sort of come behind and create that.
That is super interesting.
Yeah. And I did it recently and it was super helpful. It saved me-- because I would have otherwise had to spend a few hours building at least a mock-up of what that workflow was doing. And instead, I spent 10 minutes and I built an app interface and got the user to say, "Yes, this is the direction I want. Here's where I need changes." So it really increased my agility. So I like that a lot. I love macros. To your point, there are some challenges with them but the ability to extend the Alteryx platform really effectively is super valuable. And to give kind of shared pieces of code that my enterprise benefits from or that Alteryx at large benefits from, depending on the scope of that particular piece.
Right on. What are your favorite features on the community? I'll let Jesse start on this one. Jarrod, take a minute to [crosstalk].
I am a very competitive person so my favorite feature on the community is having all of these solution metrics and gamification. It was a very big motivator for me in kind of my journey to where I am today because there were badges associated.
In the Alteryx community, you can earn badges in your profile for the activities you participate in. Provide an accepted solution for other users? Earn a badge. Write a blog? Earn a badge. Interestingly, Jesse and I are tied with 55 badges on our profiles. So I know what I'm doing this weekend to finally pull ahead of Jesse.
I got points. People liked my things.
I'm addicted to badges now. Totally.
Yeah. It's amazing. And so I kind of just started from that. And then the leaderboard for solutions was also a really big motivator because I would-- every once in a while without thinking about it, I'd be answering more posts in a given month and all of a sudden I'd my name in the top five and I'd go, "Oh, I bet I can get up past those people. I bet. I'm only within five solutions. I can do that. I can solve five more things this month." And so that really drove a lot of my engagement and helped my personal journey with increasing my skill set with Alteryx and also increasing my presence on community. So I think that's probably my absolute favorite. As far as other ones, I mean, I know we've already talked about it and it's the theme of the whole thing but I very much do like the ideas threads and the ability to kind of directly see what other members of the community are thinking about and what they see as important for Alteryx. Because it also helps me to define my own development path, what things I need to worry about when I'm talking to my users. And so even from that perspective, it's super valuable.
I would say the-- when you go to search for something in Designer, the fact that it searches the community. That's, to me, a very simple but a very brilliant idea. And I don't know how many times I've been at Inspire and the Solutions Center and somebody asked me something and I'm just like, "Well, here you go." And there's just so much. The content is constantly growing, But I just think having it integrated right in that search is really cool and it's something, to me, I see it and I'm like, "I wonder why other products don't do this by default?" It's just really helpful, so.
Let's keep our [crosstalk]. Yeah.
You can cut that if you want.
Right on. And then finally, 2019.3. The name of that release is You Heard? We Listened? We should cut this part out when I don't remember the name of what the release is called.
They called it You.Release.
You.Release. 2019.3. And we are touting that as 27 customer ideas put in. Jarrod, can you tell us a little bit more about that?
So it started as an Innovation Day project and I wasn't really that active on the community prior to that. But I went through the ideas page and I just saw that there was a lot of stuff on there that had been on there for a long time, that was-- some of it was really, really easy. And I realized that product management probably didn't realize it was easy. So for our Innovation Day project, we kind of assembled a little team and Jeremy and I - and we had a QA - we just attempted to knock out as many of them as possible. If we got stuck on one, we would go on to another one. So we did 23 of them. And then some of them covered multiple ideas and then we kind of followed on, since then, with more of a focus on finishing out customer ideas that we see. And so for the upcoming 2019.4 release, there's more ideas in that but they're actually-- I'd say we knocked out the low hanging fruit. There's still some but they're bigger, a little bit bigger. I don't want to get you too excited but we have-- so in 2019.3, there was-- I counted it and there was 300 total stars among the ideas we implemented. We're looking to meet or beat that in this, 2019.4. Although, it's with a lot less ideas.
So anyway, I actually try to watch more than the ideas forum. I try to watch all of the discussions now because I'm always looking for things that are easy to fix, that I can have my team work on. I've heard the feedback from ACEs and I really want it to be as high of quality of product as possible because it's a really cool product. It's really fun. And just those little annoyances, I want to fix them. And if I don't know about them, I'm just more actively looking for those things or soliciting feedback. Actually, after this last Inspire, there's probably six defects that we put in our backlog and fix just from things we saw customers experiencing. So quality is really important to me and my team, so [crosstalk].
Ideas come from everywhere.
Absolutely. You never know where a good idea comes from.
Right on. Hopefully today we shed a little bit more light on where those awesome ideas come from.
Yeah. So this brings us to our community picks segment. So this segment, I'd like to call on each of you to just highlight something loosely related to STEM, analytics, tech, leadership, TED Talk or a book you recommend or-- for me, it would be the web comic “xkcd.” So if you guys aren't reading that web comic, you absolutely should. Love, mathematics, tech, it's a good weekly read.
You've half stolen the one I was going to suggest--
--which is Randall Monroe's new book, which is-- I think it's How to Do Everything or something like that. I might have that title wrong but essentially it's a book on taking wild ideas and treating them seriously and kind of what would it take in order to fill a moat with lava? How effective is it to have a professional tennis player defend you against a drone strike? All of these kinds of ideas. That one's actually fascinating because he got Serena Williams to test it for him and try hitting a drone out of the air with a tennis ball. And there's an excerpt of that online, I think, on the “xkcd” page that is absolutely fantastic. So I'm a big fan of that. I think it just came out recently.
Outstanding. Well, thank you guys both so much for being here. Jesse Clark, our ACE extraordinaire and Jarrod Pozarnsky, our favorite designer-developer. So tune in next time, etc. etc.
Thank you. Thanks for listening to Alter Everything. For behind the scenes tidbits and photos, check out the show notes at community.alteryx.com/podcast. We're also going to be at our Inspire Conference in Sydney, Australia, February 25th through 26th. So come hang out with us there and use offer code AYXANZPODCAST, all one word, for 15% off your registration. [music]
Testing one, two, three. Hello? I'm a karaoke guy. I don't know.
You really are. We need to do that soon. It's been a really long time since we've had our last karaoke session, so.
It's good to know. I'm learning lots of important things about the Alteryx team.
This stuff matters, Jesse.
Oh, it does.
It truly matters.
I'm not being sarcastic. That's sounds very [crosstalk].
I believe you. [music]
This episode of Alter Everything was produced by Maddie Johannsen (@MaddieJ).