We Are the Picks and Axes — Selling the Infrastructure That Makes AI Actually Work

00:10 Warren Kucker: Hey there and welcome to another episode of Selling AI. Today we're diving into what it's like selling into infrastructure that makes AI actually work and doing it through one of the biggest transactions companies can go through. I have Joe Simpson on the show today and Joe is the director of enterprise sales at Alteryx, leading a team of field IEs across the entire West Coast. Here's what we dive into. Joe's been there for 11 years, which means he's gone through an entire evolution of starting out at get analysts out of Excel hell to where the AI data clearinghouse powering your agentic workflows. We talk about how enterprise sales has changed in these couple of sales motions as the product conversation has shifted and how buying criteria has changed. And most importantly too, how this team is using AI internally to help streamline the buying process. So great, Joe, thanks so much for being here today to talk about selling AI into the enterprise. Welcome to the show.

01:00 Joe Simpson: Hey, thanks, Warren. Awesome to be here. Great to see you. We've known each other for a long time. It's fun that we get to do something like this professionally.

01:06 Warren Kucker: I'm looking forward to it. I'm super pumped to dive in. So this is great. So first, give us a little bit about your journey. I know you've been in Alteryx for 11 years. You were at a couple of sales roles beforehand. And then let's talk about your journey within Alteryx. Let's dive in.

01:19 Joe Simpson: Yeah, absolutely. Man, I'm really excited to be able to tell my story here. So I've been in Alteryx for 11 years. That's a long time in the tech world. know, honestly, I kind of stumbled upon it so many years ago. I wasn't sure that I wanted to do sales as a profession back then. I was just looking for another job and I just fell in love with the people, the technology, the conversations that I was having with some of these customers that we were supporting. And yeah, it's kind of been a total life changer for me. When I started, we were less than 250 people, had less than 50 million in revenue and less than like 800 customers. And now we have over 8,000 customers and we've gone on the public, we went IPO, we went under private equity, now we're this giant company. It's really exciting, was a great journey and I'm ecstatic to still be around.

02:11 Warren Kucker: Yeah, it's a good point to like in terms of how you evolve through a career. So Joe and I have been friends for 20 years I don't think I really knew he was in sales until about five years ago. I like I trying to learn about sales You know about sales and like it just sales can come from anywhere kind of scenario So cool

02:26 Joe Simpson: My claim to fame is that my first sales job was as an ice cream truck driver on Long Beach Island. That's the fun fact

02:32 Warren Kucker: That's amazing.

02:33 Joe Simpson: about me. I learned the best sales, the best advice for sales back in those days. Shout out to Eric at Bayberry Ice Cream. He said, you're not selling ice cream, you're selling smiles. And that's true, right? As salespeople,

02:47 Warren Kucker: That's good to hear. Yeah.

02:50 Joe Simpson: all we want is for our customers to be happy. It doesn't matter what we're selling, as long as we can find out what makes them happy. That's great. That's the goal.

02:58 Warren Kucker: I mean, Choco Taco sell themselves. If you've built your sales habits off of what it is to sell ice cream, you might have some bad habits that have come into play.

03:06 Joe Simpson: I was always more of a Snickers ice cream bar myself guy, but you know.

03:09 Warren Kucker: scene. already added demerit.

03:11 Joe Simpson: Yeah.

03:12 Warren Kucker: Great stuff. Well, tell us about your journey within Alterix. How did you start out? Where did you start your career? And how have you kind of progressed through your career there?

03:19 Joe Simpson: Yeah, so I started off as what we called the commercial expand role. So basically we had patches of like 50 existing customer accounts in the SMB to mid market space, less than a billion in revenue, less than, you know, a hundred K and ACV spend a lot of times. And our goal was to find new buying centers essentially within the companies for Alteryx. We might dig into Alteryx a little later, but ultimately it's a data tool. like whether you're an HR or supply chain or finance or wherever IT, you can use our tool. So was our job to kind of expand outside of the user base and find new people that can benefit from it. So I did that for a while. Then I did full cycle SMB. I was focused on landing a new business, moved more into the global 2000 space, landing net new, know, 4,500 customers in the US. And then we, you know, kind of pivoted here and there and, you know, resegmented a few times. And then You know, I went through six bosses in 12 months, the last of which I kind of raised my hand and said, hey, I've been doing this for a while. I've been pretty successful. I've always wanted to jump into leadership. Like, let me give this a go. And, you know, the rest is history. I've led a team on the West coast here for the last four years. And it's been extremely, you know, heartwarming to see a lot of their successes and me helping and coaching them to, to, see some great successes. And, you know, we've had kids born and houses bought and Many president's clubs attended and it's just been an awesome ride from that perspective.

04:51 Warren Kucker: Awesome. So had six bosses in 12 months and then you raised your hands to step into what seemed to be like a fire where you get churned out every two months. What do you think what changed? What was the inflection point that made it so you weren't seven?

04:57 Joe Simpson: It was a heavily... No, it was just a major period of change. We had our co-founder and CEO step down. We had a new CEO. There was new CROs that were being hired. And just like a lot of the old guard was looking to make their exits voluntarily. And so, yeah, I joined a team with three people that had been around for a while. and a boss who immediately put in her notice as soon as I pivoted over to her team. I reported to her for all of like two days. She went over to Databricks and is doing very well over there, along with some of the other people from my team. But like for a while it was just like me and two other people running an entire territory meant for six AEs by ourselves for like six months. It was interesting. It a fun ride, but...

05:46 Warren Kucker: Awesome. Cool, cool, cool. Let's talk about the evolution of Alteryx a bit. When you started, it was all about getting analysts out of Excel, easy access to data, these types of things. Obviously, every company is repositioning itself towards an AI play, you guys especially. How has that positioning changed to your sales process? What have you seen from that evolution?

06:04 Joe Simpson: You know, it's funny because like, you know, I mean, this business is so cyclical. We go through these, you know, kind of pendulum swings back and forth. I'll probably reference that a few times, but like the Excel hell problem still exists. So for a lot of the outreach and outbound work that we do, like we are still solving that problem. There are still way too many enterprise companies who run their business on Excel. And that's partially because the Excel models have been around for 10, 15, 20 years and you can't operate without them. So how do you kind of... how do companies kind of move out of that spreadsheet world and into a more procedural world, right? A pipeline world that actually has, know, auditability from, you know, the source system to the report that eventually gets, or the filing that ends up happening. So we still solve that problem. But I'd say that the interesting thing in the market is that there are more points throughout that process where AI is coming to the forefront. So our product positioning isn't really changing in the age of AI. We're still solving the same problem that always existed, which is largely data workers, whether you're an HR analyst or a supply chain analyst or an accountant, right? You're getting data from a system. You need to do something with that data and then you need to tell somebody or put together a report with your findings. Now a lot of that middle part is being used by AI. And so our pivot has been to, we've always kind of done that middle thing really well in a way that allows the data to be traceable. All of the changes that you make to be auditable and certified by the governance that IT is putting in place and all those protocols. Now there's just a spot for LLMs that are in the middle of that process. AI is awesome for contextualizing data, but like you still need to feed it good data in order to get the results out of it that you need. It's been the same problem for forever, right? It's the same thing before LLMs were around as just predictive models, right? Forest clusters and linear regressions. It's the same thing. So as long as we can continue to portray that we can help with the everyday data worker, somebody who doesn't know how to code, maybe is an advanced Excel user, but doesn't have like the data science background, we can help that person get data into a system that is governed, certified. know, approved by their IT or data science teams to use to then contextualize that data in a better, smarter way to make their job more efficient.

08:33 Warren Kucker: It's interesting that maybe some of the marketing terminology has shifted, but the actual problem being solved and how it's being solved has remained the same. Has the objection changed? So I imagine when you were selling Alteryx for most of your time there, if not all your time, Excel is great in that it's flexible, right? I can just open up a spreadsheet and I can do whatever I want with it. You wanna add some inflexibility into the process, right? There's data governance, there's pipe, there's like, wanna keep you flexible, but I wanna also... make it traceable that inherently adds a layer of rigidity. Does that has that changed as like adding AI into the layer and made people more receptive to like not doing it in Excel anymore or is it the same old ballgame for you?

09:17 Joe Simpson: It's an interesting question because you got to figure the accountant or the tax analyst who's just trying to get their job done, like they're just trying to do it as quickly and efficiently as possible, right? So a lot of times they don't follow the right governance protocols. And this is where you come up with what we call shadow IT, right? It's basically technology that's being used without any visibility from the IT side of the house. So a lot of the times when we're having these conversations with our, you know, we're a line of business tool. IT does use us. have use cases within the office of know, technology and data science teams, et cetera, but our bread and butter is the average business user. We help the business user explain why a tool like Alteryx is needed to the folks that only care about governance and making sure that there's no changes being made to the system of record, right? So like over the course of the last 12 years, the pendulum is swinging, whether IT has the power or business has the power. What we've noticed in the last few years is that IT is definitely more a part of the conversation now than probably ever in my time at Alteryx. I think AI has a lot to do with that. There's a lot of security questions and every technology out there has some sort of a co-pilot element to it. We're guilty of that as well. It's making it super easy, a lot easier to use these types of technologies. But then that raises all kinds of security questions, data privacy questions, how are you using our data? What type of data are you tracking? And so a lot of the... the budget conversation and the do we actually need this tool conversation falls solely within the IT. Business doesn't really have as much authority as it did five years ago to make those types of decisions.

10:59 Warren Kucker: That's fascinating. When you position this at Auth0x, what I find fascinating is like, you may be, but you're not really selling AI features directly, but your pitch is like, We've always been doing this data piece really well. Now it's especially valuable that you have LLMs that open up this box of possibility with the information that you have. So still use us. you're pitching the infrastructure. Have you found it like? an easier sale or how is the buyer ingesting that type of like change and or that that type of like similarity in the message.

11:35 Joe Simpson: Yeah, yeah, it's another really good question. And it's an interesting one. I don't know about you, Warren. I know that you guys have a really great tool, by the way. Check out topic later. But I don't know what your conversations are, but I'm just getting a lot of tire kickers. A lot of companies, especially enterprise companies, they don't actually know what they want out of AI yet. It's still very new. Chat GPT exploded on the scene 18 months ago. And nobody really understood the power or where it was going to go. I think we're closer to that point where companies understand the power, the value, and where it can be most beneficial. But ultimately, in a lot of the conversations that we're in today, it's a lot of discovery. It's like, well, what can you do? I wonder if that aligns to our corporate initiatives around efficiency gains that we think we can get from AI. But also, You know, another interesting thing is, kind of going back to the budget conversation. Like companies are investing in AI tools. Like a lot of the budget is being pulled from, you know, SAS systems to, you know, AI toolkits, whether they are, you know, in what are they're utilized and deployed and in the right places or used in the right ways or not. It's just like, it seems like they are implementing AI for the sake of just testing it out. And that's okay. Like new technology, you want to do POCs, you want to test stuff out. But it does make it harder for, know, frankly, companies like us to validate and take some of that budget or, you know, earn and keep some of the budget that we have because we're not a true AI platform, right? We are, our CEO described it when he first joined the company as we are the, you know, in the gold rush era, right? Where you, did the miners make a lot of money or were the people selling the picks and axes making a lot of the money? Like we are the picks and axes, right? Like we're not doing the AI work. We are helping, create efficiencies for a new AI world. So without being an AI tool, can we get the AI budget? We're still trying to solve that problem, answer that question.

13:40 Warren Kucker: Now, it's interesting, this is a common theme and I think one thing that's been common in lot of these conversations has been what's happened with this new tech is that, specifically boards of directors have come out and said, you must find ways to use AI and I want to see a 5 % improvement in EBITDA based on an AI tool usage, right? And the reason they're doing this is that they think either one or two things is happening. They see it as an existential threat, right? They worry that a brand new company is gonna do exactly what this company does really fast because they can just rebuild the software from scratch. or that all their competitors are just doing this and they have to move ahead. So there's like an existential threat response that's happening. What then happens on the other side of the fence is like, it's kind of interesting. There's a mismatch in like roadmap development, right? Like where like the customers don't know where to use it. And so they're looking to like their vendors or their potential suppliers as like the true consultative sale that we always wanted to be. But we really don't have the answers yet. Right? So it's like a great, like, cool. It sounds like, no, no, you tell me, like, what part of the process would be better with AI? Right? Like, like, no, no, no, you tell me, what are the other people using AI for? Well, they just kind of bought this platform. They don't really know what to use it for. Right? So there's like a weird roadmap mismatch that I think is really actually started to come around to being, let's do it the old way. Like, let's move a little bit slower. Let's be methodical. Let's figure out this solves a problem. So that's really kind of fascinating that you're getting more tire kickers. People are coming to you asking how they're using it. and they're running these more POC models, they're saying they have AI initiatives without committing the funds to it. So whether you get a piece of AI spend is really a good way to think about this from a go-to-market standpoint at all terms.

15:16 Joe Simpson: Yeah, I've seen a lot of articles and different thought pieces on this exact thing. And everybody is saying that exact thing, which is the budgets out there are growing for AI and reducing for other areas. But getting that AI budget is so challenging because you have to prove that you can solve a specific problem. And the other problem here is, like me as the AI team, right? told by my board that I need to go find these AI incremental gains. Like, who am I talking to? There's dozens, if not hundreds of business units out there that all have very specific needs. So do I do an LLM approach and create a bunch of GPTs that kind of serve each of them? Is there a niche tool that gets the most bang for the buck for this particular business unit? Or, you know, like, how do I, how am I going to go about spending this new budget that I have, which isn't really new budgets, just moved from one place to another. It is making it challenging for sellers. And, you know, we have to do a better job at understanding kind of what the modus operandi is for the person with the AI budget. And how do we get it that our piece of that piece.

16:32 Warren Kucker: What you describe, again, often come to the same point. It's new technology that's making old things new again. What you just described is what sales has always wanted to be, right? And that's what we always should have been, right? So it's like, great, how do we figure out how to solve a problem? And I've actually found, when I'm talking to sellers, I think we're suddenly starting to inch away from even talking about having an AI solution and talking about problems to solve that are actually new problems that were kind of unsolvable before. So like, and that's all that matters, right? Like no one needs AI validation. I actually don't talk about AI and sales processes myself nearly as much anymore. I try to almost openly avoid it. I have a new way of computing information that makes this process better. Is this process better for you? What problems does it solve? It's the exact same paradigm in sales. It's just kind of actually like fresh. Now we have new problems to go after, right? It's kind of what we're always been hunting.

17:24 Joe Simpson: I mean, the key remains is all in the discovery. Just figure out what, to your point, what problems are we actually solving? What are you actually trying to accomplish? And it doesn't matter how you get there. If there's an AI element, great, you can go get the AI budget. But at the end of the day, if it's a problem worth solving, they'll solve it. Or they will pay to solve it, is my point.

17:43 Warren Kucker: and they'll pay for it. Right, exactly right. Cool. All let's pivot a little bit into like sales leadership. How are you using it, any sort of AI tools to manage your team better? How is anybody on your team using AI to create a better buying experience for your buyers?

17:58 Joe Simpson: Yeah, you know, probably compared to some of your other guests that have been on this podcast, like my 2026 goal is to do a better job at this. I don't use AI as much as I nearly should. We have an open AI chat GPT license with a bunch of custom built GPTs that have been super incredibly helpful for a lot of my AEs. We've built a bunch of our own, you know, talk about the AI data clearinghouse. Like we kind of demo our GPTs in this way because we... grab data from our internal Salesforce, from our contracts, our licensing. Flexera was our licensing provider. We get all of this stuff and we fed it into an LLM using the Alteryx AI data clearing house methodology and then created this GPT that now allows us to do deep learning and a lot of the common things. What is an email that I can send to the CFO at this company who has this many users and so on and so forth. We feed in a lot of use cases and things like that. Hey, I'm talking to a chief accounting officer tomorrow. Like what are the five things that I should care about? And they you know will the GPT will spit out basically a PowerPoint ready to go that we can that we can leverage And then we have other tools like converted into our corporate templates and stuff like that So that part has been amazing for account research my eight users seeing a I don't know a 20x lift on what they were doing before ripping through 10ks and all that good stuff to try and find those initiatives that we're talking about, right? To be able to allow for more pointed discovery in those initial calls because we have more data on what potentially people are looking to accomplish. From a sales leadership perspective, yeah, I'm bad at this. The one thing that I started doing, I do a monthly pipeline review which is looking at quarter plus one and beyond for my reps. Just basically like how healthy is next quarter and beyond gonna be? And so I started, it would take me so long to like, first of all, I was multitasking while I'm on the call, cause I'm taking notes, trying to remember everything that we're talking about. And then to do the followup and say, Hey, this is what we talked about. This is what you committed to doing. This is the next steps. And this is what you said about your, what you're going to use your SDR for, et cetera. And that took so much time. But man, I created a project in chat GPT that basically gave me the exact output of that message. I would just throw in the transcript and be able to send those things immediately. And that was a huge game changer for me last year once I started doing it. But there's a lot of, we just started using, you know this, we just started using chorus and zoom info co-pilot. Half of our team has been implemented that now it's still being rolled out, but starting to dabble into that and kind of getting, know, being able to coach without being there because you have the, you know, the conversational intelligence and things like that. It's very brand new. We're still trying to. wrap our hands around what that's gonna look like. So anyway, but

20:51 Warren Kucker: That's great.

20:52 Joe Simpson: 2026 for me is the big like, need to double down on this because hopefully I'm at Alteryx for another 11 years, but if I happen to find myself on the street a year from now, if I don't have these AI capabilities, like I'm not gonna be able to find a job. That's the reality of life now.

21:07 Warren Kucker: whenever I ask this question, it's really a good way to illuminate. Everyone should take pause and realize that when salespeople start talking about how they're using AI, this is exactly how our buyers feel when they're talking about buying products. So you just said to me, great, you know what, like, I'm not doing anything, right? Like this year, I really have to focus on this. 2025, lost year when it comes to like AI implementation, unlike all the other people you're talking to. I get that exact answer at the same time. And then what happens right after is you just listed off 12 incredibly complex processes that you're leveraging custom GPTs for that you weren't before. And then the second piece of really, again, that salespeople should pay attention to that happened here is you barely mentioned until the very end software that you bought. What you mentioned was, great, we whipped together this GPTs and we fed information to it. And obviously, they've been phenomenally valuable. And this other thing that my reps can do just with chat GPT on their own, right? They do really well on a 20x uplift, but I'm not doing anything, right? Like, and we really need to focus on this here, right? That's how our buyers feel when we're having these conversations. That exact response is

22:11 Joe Simpson: That's hilarious.

22:11 Warren Kucker: how everyone is going through it, right? So we should remember that as salespeople. It always kills me. Let me ask, these are really good systems to be built, right? Who built them? Who built these custom GPTs and got the data through? Is it your engineering team? Because it's your AI data clearinghouse. Is it rev ops? Is it your sales people just like rallied around and got the work done? How did you build those processes internally?

22:32 Joe Simpson: We had a dedicated Tiger team, mostly rev ops and product marketing, engineering did play a role, sales engineering played a role, honestly, so that we could get a better handle on what we were building so that we could demo it later. It was definitely a multi-touch effort. We have dedicated subject matter experts that focus on different areas of the business. So like I said, we can... You know, everybody from your average tax analyst to an HR person to a supply chain inventory specialist can use our software. So we have custom GPTs that specialize and focus on those particular business areas and the areas that those, you know, personas are targeting. So we have an office of finance one, we have a supply chain one, inventory, you know, you name it. We have a GPT that kind of helps us understand what a day in the life of those types of people are. That's one of the things that I love about Alteryx is, I talk to so many different types of people. I learn so much about these different departments, but also like I'm only this deep. And buyers want to feel, I always say, it's like they're all, their own unique little snowflake. Everything in their world is unique to them. And so if we're only this level deep, we're not actually giving them the right buying experience because we don't know enough about them. So these GPTs have been really good at kind of helping us live a day in their shoes. in a way, and use the right verbiage and the right language when we're talking to them. And getting customized GPTs in those ways has been really good going back to Discovery, like allowing us to ask the right questions to the right people.

24:04 Warren Kucker: That is a really good segue because what you're describing is a buy versus build scenario internally when you're talking about this, right? You didn't buy a piece of software to build everything out you just described. Again, other than Lightsight Chorus and Zoom Info, but again, mentioned it last. How are you handling that in the sales process when people are like, listen, I have Excel, I have Copilot, like I have Power BI, AI, Copilot and Power BI, I don't need Alteryx, right? Like what's that sales conversation like right now for you?

24:30 Joe Simpson: It's a good one because the build versus buy thing always comes down to do you have the technical skills at the business level to actually use the system that you plan on building? And most of the time the answer, most of the time they'll say yes and then they'll find out that they're wrong later. like I just go back to, Power BI is a great example. Power BI is an awesome tool if you're just using it for visualization. It's similar to Tableau, it's similar to Click, it's similar to a lot of these other. know, visualization interactive dashboarding type solutions. Can Power BI do data pipelining? Yeah, it's actually really powerful at doing it. If you know DAX, DAX is a specific coding language that allows for connections within the Microsoft data universe. Your average accountant is never going to learn DAX. So as much as you say, hey, yeah, we can do that in Power BI, great. You're gonna end up with the same problem. You're gonna create a system. where a business user will then go to you to say, hey, I need these tweaks because I can't do this myself, because you've created a system that is too technical for me to actually change things. And so you're moving from one ticket system to another ticket system, whatever that looked like before, okay, I'm building this new thing back into the ticket system. The power of all tricks and what we're hoping is, you the vision of AI ready data, which is we are still giving the power to the business to do their jobs just in a governed way that IT has certified and made successful. They don't need to build anything new. They can continue to focus on making sure that the source systems are accurate and up to date, again, certifying the right data assets for use, and then let the business cook at the end of the day. Let them use the certified, the pre-approved LLMs and the different things that you've given access to. and let Alteryx be the glue that keeps everything working together as opposed to building a brand new system that just is adding another thing on top of your tech stack. So that's how we address that type of question or response from a buyer. But honestly, the buyers are a lot of IT folks now and they push back a lot on that.

26:44 Warren Kucker: It's Now you can tell the difference. Right now that question should have been how can we figure out to sell what 11 years at a company looks like compared to an average 10 year rep, which is about 2.3 years right now. I never heard of DAX and I'm sold. So good pitch.

26:57 Joe Simpson: You

26:58 Warren Kucker: So nice work. Great. So I mean, let's say last question. You've been selling until let's say the global 2000 for five, six, seven years at this point. Has anything changed in their buying behavior over the past 12, 18 months? I know you talked about a shift more towards IT, but overall just buying process, research, what they expect from you to come to the table with. Has anything changed or is it like everything else kind of remains the same?

27:21 Joe Simpson: No, it's definitely changed, especially post pandemic, I'd say. There are a lot more approvers in the cycle. So

27:31 Warren Kucker: Interesting.

27:31 Joe Simpson: I'll give an example of a customer that we sold to, this is back in 2021. So this is when it started. This was the first time I noticed this. So we had done three deals with this customer, very large company. But we knew their buying process to the T. We knew that they met every Tuesday for any purchase over $10,000 and they approved that purchase, yes or no, on every Tuesday. And if you put in a request on a Friday, it gets on the agenda for that Tuesday. And we had a deal. It was a good size deal. It wasn't a huge deal. was like $100K, $150K, something like that. And we knew everybody in the approver chain. And something like something happened and all of a sudden like we missed it one Tuesday, then we missed it the next Tuesday and then it was the end of the quarter and the deal slipped and we were like trying to figure out what is going wrong. And we found out that all of a sudden their process just changed overnight with, you know, everything needed to go to the CFO now. Didn't matter how big or how small, everything needed to go to the CFO. So we were at like, you know, the VP of Finance level. We had to jump three levels to get approval for our deal. And that took takes more time, right? Because executives are busy, like to do that type of analysis on everything, it just takes longer. From then on, like we tried to go more in depth with mutual action plans, making sure we know the process, know, MedPic, I know you're a big MedPic fan, like making sure that we had all of the right elements in place for that paper process, because it was just changing dramatically. I do think some of that had to do with what I talked about earlier with IT now also getting involved. but also the executives. Like you see so many layoffs in the last five years. We were in a growth economy, right? Like growth, growth, growth. Wall Street kind of pivoted and then it became profitability. So with profitability became stricter visibility into what we are buying because that affects the profitability. Those were the big changes that I've seen. So there's just more people in, know, hands in the pot when it comes to making decisions. And that does make it harder on sellers because now we have to think of not just our champion and what they care about, but what is the potential point of view and perspective from the IT group that's going to be oversight? What is the perspective from the finance group that's going to ultimately have to pay the bill? Like you need to be making sure that you're putting together a plan to appease all of those different hands that are now in the pot. How have you seen it? Have you seen anything different in that world from

30:00 Warren Kucker: I don't know that I've seen like an increase in the approval for call, but I don't think I would have noticed at the scale that you had up. So that's um, it's really interesting you think about this process. The only reason to add more approvers to the chain is because they like enterprises feeling an increase in risk, right? Like that's if there's if there's a decrease in risk, then you people at the front line should be able to make decisions more individually than like every $50,000 approval has to be sent to the CFO. And part of that may actually be really structural. Interest rates are not zero anymore. So the ROI calculations are greater, right? And so those cost pressures are there in a different perspective. A financial modeling might be different. But interesting, data governance risk, IT risk, there may just be an uptick in the feeling that we need to put additional layers in the process. It doesn't seem like that would be by, it seems like it would be subconscious, not a conscious decision, but also doesn't seem like it would be by default or frivolous, right? There's definitely like a probably a perception that there's more risk in buying new services or new processes. It also could be, I'll throw this out there and see if the cat picks it up here. There's also more options than there used to be. So like the optionality may increase. If you think about this space for a long time, you're competing with Excel and maybe like Tableau. right, like for a long time, right? Now there might be, like for this type of budget, there may be like five X the competition in the space. like, each with

31:34 Joe Simpson: percent.

31:35 Warren Kucker: pretty heavy nuance, 20, 30, 40 % differences where a choice in software is a choice in strategy that may not exist. Do you think there's something there? I like, I hadn't really thought of the problem before other than just riffing at the moment. Is there something there that might be interesting? Are you facing more competition in your sales process?

31:51 Joe Simpson: There certainly is. One of the challenges that Alteryx has had was we are natively an on-premise product. And so over the past five years, we came out with a cloud-based edition of our flagship products. But there are other competitors in our space that have been cloud-native from day one. And so as a lot of IT departments, again, are taking ownership and moving to a more cloud-centric operating model, right? Less on-prem servers and boxes, more, you know, hyperscalers. That was a challenge for us to scale. And there are so many, to your point, like technology out there that do the niche things that, I mean, you how many startups, tech startups do you see like a day basically? Like they're always solving a very unique problem. And we are more kind of... We're not really big box in this, for the world, but we are like a Swiss army knife. You can use this for a thousand different things, which is kind of a benefit, but also sometimes a hurdle because then you go in there and the technology buyer says, well, great, I'm glad that you do all these other things, but we're really only looking for this one thing. And guess what? There's a niche player that does this better than you. And so even though we are trying to consolidate the toolbox in a lot of cases, we see that, right? Reducing the numbers of technologies, like if it's... If it's a big enough problem, they're gonna go with that niche player who only solves this thing but probably does it better. Yeah, there's a ton of competition out there.

33:20 Warren Kucker: And Dairy, like, Alteryx is like that three to seven year decision from a strategic standpoint on top of it, right? Like, it's just not like a rip and run kind of scenario. So it's just like...

33:27 Joe Simpson: It's it's so we we've always operated in the land and expand model So we go in with a small a small group, right? Like we sell five

33:31 Warren Kucker: Okay. Okay, good.

33:34 Joe Simpson: licenses to the tax team for example And they you know do their thing for six months and then all of a sudden they start to see a ton of ROI and then accounting jumps on board and then it's 15 licenses, right and then like that's awesome like audits now trying to wrangle this thing and this is an unbelievable and then it's 25 and then you know before you know it we're three years later fully embedded across the whole office of finance so It's a little bit of both. like, we, yeah. And so that's the thing

33:58 Warren Kucker: It's not high commitment then, right? It can be low commitment, right?

34:01 Joe Simpson: is like, we don't have like a true from the beginning enterprise sale. is, we have cut customers that have chosen to go that route, but like usually it's

34:09 Warren Kucker: Right.

34:10 Joe Simpson: we go in in one place and word of mouth spreads because everybody's seeing a ton of success. But, you know, again, like when you get that first sale, you need to beat out those niche players. And sometimes it's just Excel, right? Cost of doing nothing. But other times it's, hey, yeah, we're using a specific tax software that does this exact thing that we're looking to use you for. So whatever.

34:33 Warren Kucker: That's a call out. Cool, man. All right, well, last question. Just like, do you have any opinions either in sales process or just overall tech and how things may shift at Alteryx? Anything like two years from now, the way we're using LLM, is anything come to mind that might be drastically different than we're just kind of really thinking about the problem right now?

34:52 Joe Simpson: Yeah, you know what? I will make one comment. This wasn't really on your preview, but one thing I want to address is just like, everybody just assumes that in two years we're all going to be out of a job because AI is going to take over. I think that's total BS. The thing that AI right now that LLMs are really good at is, I talked about earlier, taking data or taking a prompt and contextualizing and making the question that you're asking more understandable for the end user. There still needs to be human interaction to feed in the right data and the right question and then interpret and tell the story of the outcome later on. I think over the next two years, the skill sets that we all need to learn in sales, I can talk about this earlier, but like in sales and also in all other areas of the world is like we need to really make sure that we know how to work with LLMs. That stuff's not going away. It's only gonna make our minds way easier and way more efficient and we're gonna be more productive. And I think the people that are going to benefit the most are the ones that are spending the time to learn how to get the most out of LLMs for the job that they have today. I think that's we're going to see. And the technologies that are using LLMs to do really cool things that have never been done before, to go back to your earlier comment, like that's who's going to win this AI fight. Whether that LLM is something homegrown, whether it's Claude or Gemini or ChatGPT, Whatever like it doesn't matter what LLM you're using. It's what you do with it that matters So those are the skills that I think we all need to do and that's what I think is gonna You know, the cream is gonna rise to the crop two years from now

36:32 Warren Kucker: That's a great way to put it. It's simultaneously like a hot take and really shouldn't be, right? So like on one hand, as these conversations have evolved, where this technology in particular is probably the most concerned we should worried about, like being a job destroyer compared to a job creator. So it's worthy of the discussion, right? And it should be thought of, right? Like in a thoughtful manner. On the other hand, it's still relatively unlikely. And if you look at it like, If you look at the evolution of tech, you went from like just analog, everything's on paper, right? And then we got computers on everyone's desk, which happened like mid 90s, late 90s. But your computer, your desk just turned your notepads into ones and zeros, right? So we have like, we got connected, we started sharing information at a scale that never existed before, but you still had to type everything and read everything, right? This is the first shift in this type of like digitization where like, Now you have a new way to compute that actually takes human language and generates new data sets. The new data sets, could be a problem where we like, great, these jobs that we used to have to do things, we don't have to do them anymore and we need to find the solution to that. But it also could open up a lot of new areas. This is really like, say picture a world where this is the way we set up transistors was just the way we set up LLMs and this is what computers did off the rip. Like it would have been, it probably would have been okay, right? Like it's not unseen, foreseen that I could have done that, right? So I actually, I thought about this last night and maybe I'll throw it one more time. Part of me thinks we shouldn't have branded it as AI. We should just say, great, like we got these new chips, man. They're lightning fast, right? Like, and

38:15 Joe Simpson: Yeah

38:15 Warren Kucker: they like, we processed everything that ever happened on Twitter and Reddit and like figured out what humans say next with probabilities, which is how your brain works. And it's not AI, it's just new compute. Like we should use it. Like I wonder, wonder what that... alternate path looks like. But anyway, I'll throw that out there.

38:34 Joe Simpson: I love it. And I'm going to make a funny pivot here into our personal lives. Warren and I have played in the same fantasy football league for 15 plus years now or whatever. It's the analytics conversation in football. It's like, why do we ever use the word analytics? Now it's the analytics fault. Dude, analytics has been around forever, right? Since you have information and you make a decision based on the most probable outcome. Like that's just, what do I want to eat tonight? Like I'm going to rely on the analytics of what I like to eat and I'm going to make a decision based on it. To your point, like

39:00 Warren Kucker: Yeah. Yeah, right.

39:02 Joe Simpson: that's kind of just AI. Like we're feeding it con, we're feeding it information and we are giving it context and asking it a question. It's just kind of spinning out and regurgitating the information that we've given it. Like, yeah, it's a chip. The other day it's a chip. I love that,

39:15 Warren Kucker: Right. Yeah. Now it's like well I'm glad you pulled it into a nice, relatable

39:18 Joe Simpson: it's just like it's just making us all smarter.

39:20 Warren Kucker: analogy. Yeah, exactly right. And by the way, it's doing a lot of work that I don't really like doing anymore, so I would cool write that persona up. Man, I don't really care what a CMO cares about, just like figure it out and feed it into the machine. Great stuff, Joe. Really appreciate you taking the time. It's been really fascinating to see, like, again, things change, things remain the same is like a good theme and maybe what we even title this episode. If you travel a lot, do you go to conferences? Where should people find you if they wanted to like hit you up?

39:48 Joe Simpson: Yeah, thanks for the plug. mean, I do travel a lot to customers and mostly on the West Coast. We do have our annual customer conference this year called Inspire. I think we'll be in our 15th or so year. This year it's in Orlando in May. You'll see me there. You'll see us there. We're at a lot of conferences. We partner with Google and Databricks and all the other major players in the data space. So you'll see all tricks around, but yeah, fun stuff.

40:14 Warren Kucker: If need your data to be AI ready, call Joe or show up in Orlando and we'll go from there. So I appreciate you the time,

40:19 Joe Simpson: You can call Warren too, he'll put us in touch.

40:22 Warren Kucker: I go for that. I'll take the phone call for you. I'll be with BBR. I appreciate it, Joe. have a... If I'm right,

40:25 Joe Simpson: Now that you know what Dax is, you're ready to give the pitch.

40:29 Warren Kucker: I so. I just regurgitate your pitch. I'm just like an artificial intelligence

40:31 Joe Simpson: You

40:32 Warren Kucker: for you at this point. Gave me all this context. I could just regurgitate it. So

40:35 Joe Simpson: Perfect.

40:36 Warren Kucker: great stuff, Joe. I appreciate you taking the time.

40:39 Joe Simpson: Yeah. Thanks everyone.