Cut Cloud Costs 80-90% with Kevin McGrath, CTO of Spot

(Intro Narrator at 00:00:03) Hello, my friends. Today, Joel is talking to Kevin, the CTO at Spot, and they discuss how Spot saves their clients 80 to 90% on their cloud compute bills, the concept of data gravity and how to conquer it, and how to design career paths that include growth without necessarily moving into management. All of this right here, right now, on the Modern CTO podcast.

(Joel Beasley at 00:00:30) Here we go.

(Joel Beasley at 00:00:31) This is the Modern CTO podcast.

(Kevin at 00:00:43) I was a young kid in middle school going into high school, and we started our first company, local company, basically hosting web pages. I mean, this is going back to the nineties, right? So we're talking, you know, early, early stages. Like, GeoCities didn't exist yet. And we were building, you know, some of the first web pages that companies ever even had just to get them on the web because they heard the web was cool, back when you didn't have to have your website up twenty-four hours either.

(Kevin at 00:01:16) It was like, you know, old TV days where it didn't matter if you had 24/7 support on your website. It was just there, you know, as like a vanity thing. So we started with that and we were building websites all over the DC area and brought that through college. And I actually exited that very, very small thing in college, selling it to another local area company so I could kind of get the college experience. But I only did the college experience for about a year until I started working at a company called USI, which was another startup in the DC area.

(Kevin at 00:01:48) And I started there basically my sophomore year. And since then, that would be running the application. They were called an ASP, application service provider. It's what we called it before it was SaaS, and before the first tech bubble burst. And I did that for a while, and then the tech bubble burst, and then we kind of meandered or went around for a while in the industry until everything came back.

(Kevin at 00:02:13) But yeah, I got started very early back when, you know, we built the data centers, racked everything. Seen the journey from there to cloud, you know, along the way, and then building applications as a first model for consumption, you know, very early on in the earliest days. Because back then you used to sell servers and you just used to sell racks. So even if you sold someone an application, you would sell them a rack. The rack came with a specific amount of servers. The application would get sold on those servers.

(Kevin at 00:02:40) You would cable it up. That was the delivery. And so USI was one of the first companies who just sold Oracle or just sold PeopleSoft or just sold Ariba. And then behind the scenes, it didn't matter. Like, we had a data center, data center without cages, as we like to say.

(Kevin at 00:02:55) And it was just a sea of compute that we would sell the applications on top of. So back in the late nineties and early two thousands, that was a novel idea. Now it's par for the course, as they say.

(Joel Beasley at 00:03:06) Yeah. I was looking at what Spot was doing. Obviously, being an engineer for all my life and looking at what you guys are doing is pretty fascinating. How do you describe it to people?

(Kevin at 00:03:17) Spot, we make, you know, the cloud as optimized and efficient as possible for you. We like to say that cloud's complicated, but we'll take care of that for you. Deploy applications, and then we'll take care of all the, you know, knobs and bells and whistles and everything that the cloud provides, and we'll just automate that for you at the end of the day. What we want to do is allow engineers and developers to, you know, build and package their application, deploy it to the cloud. And then the clouds are so vast with so many different options, so many different purchasing models, so many different ways to run that application that you don't really need to be concerned about that. You know, the whole serverless mindset, as they like to say, it's kind of an overloaded term now.

(Kevin at 00:03:59) But we still like to use it because that's what we really want at the end of the day. We don't want people to worry about what type of VMs they're using or what type of servers they're using, quote unquote, at the end of the day. Deploy your application. We'll figure out the most efficient way for it to run, whether that's in a container or whether it's on a VM, whether we have to manage the commitments with the cloud provider for you, whether we need to figure out how to launch the next most efficient VM for you. We want to take the whole infrastructure out from you ever thinking about it again and just allow you to deploy applications and know you're getting the best possible cost and the best possible utilization out of your resources.

(Joel Beasley at 00:04:34) So you do more than what I picked up on, because I want to share with you what I picked up on from the home page.

(Kevin at 00:04:38) Sure. Go for it.

(Joel Beasley at 00:04:39) You have a killer graphic. It's like a slider where I can slide over usage. You know what I'm talking about? And so it was basically, in my explanation of it, it was showing that as resources scale, you just plug in these new on-demand instances. And sometimes you'll put in an instance and you're not using all of it, and you guys shave it perfectly. So you're only using exactly what you need, and that's dollar savings. That's what I got from it.

(Kevin at 00:05:08) Yeah. That's where the company started, so I would say years ago. I mean, and that's still a big sell, right? When you talk marketing and getting out in front of people and selling them on the product, it's still very much about cost and showing you how you can use spot instances instead of on-demand instances or RIs instead of on-demand. Basically, we don't want our customers ever to pay on-demand. There's no reason to. So shave that instance and get the cost model down. And customers come to us. It's a great banner to put out at a conference, you know, how much you can save. But our customers stay with us because of the automation and the efficiency that our platform gives just to their everyday running applications. So it's kind of a weird setup where people come to us to save money, but they stay with us for the automation and the speed that their DevOps teams can run at just by using the platform. So there's a whole story that the platform uses. Like, with our Ocean product, Ocean not only does price efficiency, but it allows your engineers just to deploy containers.

(Kevin at 00:06:10) Like, take containers, deploy them to the cloud, let Ocean figure out how to take those containers, what type of VMs they need, what type of special needs they need based on their labels. And it just takes the whole inventory of the cloud and then allows that to become your infrastructure without your DevOps team worrying about what VM they need or, you know, how to attach it to storage or anything else. Ocean takes care of that for you just by looking at what the scheduler needs in Kubernetes and then going out and automating the work after that. So we've definitely started at price, and you're definitely right. That's one of the main marketing messages.

(Kevin at 00:06:46) But we've definitely graduated from that in other areas where we're helping our customers not only save money, but also move fast.

(Joel Beasley at 00:06:52) That's exciting. I saw you have a bunch of products. Cloud Analyzer, Ocean, Eco, Elastigroup. What's Elastigroup? Is that the one we were first talking about?

(Kevin at 00:07:02) That is. That's where it all began, is Elastigroup. And Elastigroup is how we got customers comfortable with running production-grade applications on top of spot markets. And the difficulty there is that cloud providers have three different ways, normally, that they'll charge you. So there's list price, that's on-demand, that's, you know, they just list on their website. It's a dollar an hour. You go to them. You pay a dollar an hour. You get a four-nine SLA. When you realize you're running that instance all the time, cloud providers offer what they call a reserved capacity discount.

(Kevin at 00:07:35) So you can buy, like, a reserved instance. That means you're going to guarantee that you'll run that instance twenty-four hours a day, seven days a week, three-sixty-five days a year. So you get to pay less for it. Now whether that instance is running or not, you still got to pay for it, but you pay 40%, 50% off list. But then the real savings where you pay 10 cents on the dollar is spot.

(Kevin at 00:07:58) And the spot market is great because the cloud providers sell this extra capacity that's not being used for on-demand at spot prices. So when on-demand is not being used, they sell it on a market, and that price fluctuates. And if you understand how those markets work and you understand where the capacity is coming from, you can understand how to bring in instances at the right time to always keep an application highly available. And that's what we do. We can look at all of these different spot markets and then keep an application running on top of the spot markets as they change and keep that cost savings at that 80 to 90% level for you and keep the SLAs and SLOs of your application alive.

(Kevin at 00:08:37) So that's really where the magic started, and that's what Elastigroup does. It keeps a pool of VMs running on on-demand, RI, and spot, very, very heavily weighted towards spot, and it keeps that whole cluster running at the most efficient way possible for that application to deliver what it needs to do.

(Joel Beasley at 00:08:55) I just learned about spot markets. I did not understand what they were, how they worked. Thank you. What's, like, the name? Is there a name of a spot market? I just want to look up more later.

(Kevin at 00:09:06) Yeah. So on AWS, it's called Spot Instances. And then on Azure, they're called Spot VMs now. They actually just released—Azure kind of revamped their whole spot infrastructure, and they're calling it Spot VMs. And then Google, actually, we were just a launch partner with them. I think it was this week. My brain's going on me a little bit. It was either the end of last week or the beginning of this week. We just announced a partnership with them on launching Spot VMs over on the Google side. So on Google, it used to be called preemptible VMs, which only lasted twenty-four hours.

(Kevin at 00:09:39) But now Spot VMs on top of Google can last any amount of time based on the capacity of the market. So what you have is you have this spot construct on all three major cloud providers, and it works slightly different on them. But nowadays, they work more similar than they are different.

(Joel Beasley at 00:09:59) Oh, I thought—alright. So it's not a third-party exchange that, like, coordinates all the excess from all the different providers and you bid over there. It's just the spot instance inside of AWS.

(Kevin at 00:10:11) Right. Yeah. It's a spot instance inside AWS. It's a spot VM inside Azure. And the cloud providers control that market. But the thing is, if you don't have visibility into that market or you're not spread across thousands of customers, then you don't see what those markets are doing. You don't see when a capacity pool is getting full. You don't see when you can or can't launch into certain areas. And that's what our platform does. We can get in front of the markets. We know when markets are getting full or when they're most likely used, and then we can rotate the compute within your Elastigroup to compensate for that.

(Joel Beasley at 00:10:44) Can you rotate between providers?

(Kevin at 00:10:46) We don't rotate between providers today. We usually keep customers in the cloud that they are. I would say 9.9 times out of 10, our customers aren't running the same application in different clouds. They will use multi-cloud, but usually it's an application that's being used in a specific cloud for a specific reason, usually due to data gravity. But, yeah, we don't have a ton of use cases for rotating between the clouds today, but I'm always listening for someone to come to me with a great use case for it.

(Joel Beasley at 00:11:16) Yes. Yes. Alright. So Spot is just referencing—it's like when you're making these changes between on-demand and spot, you're doing that within Amazon.

(Kevin at 00:11:26) Absolutely. If they're an Amazon user, they have Amazon services, you're helping that transition between these on-demand instances and spot instances occur in an automated fashion within their account.

(Kevin at 00:11:38) Yep. Absolutely. Or even spot to spot. Most of the time, it's spot to spot. But, yes, if they need to go back to RI or on-demand, we do that also.

(Joel Beasley at 00:11:46) Were you early in this? Were you one of the first people doing it?

(Kevin at 00:11:49) Oh, absolutely. Yeah. Talking five. I wasn't one of the founders of Spot. I came in a little later. But absolutely, when Spot was founded, they were one of the only people in the industry doing what they do today. Now, obviously, the spot market existed in AWS when they started. That was the first spot market, but they were a very early adopter of it and one of the first companies to start doing this kind of automation and this kind of scaling with spot instances.

(Joel Beasley at 00:12:16) How did you meet the founders?

(Kevin at 00:12:18) It's a good question. So I worked for a company called SunGard Availability Services for a while, and I worked for the CTO there. One of my jobs was to research, you know, go out, find who are the next biggest, and best—or maybe not biggest, but the best next generation of companies—and what could help SunGard AS get into the cloud at that point in time. So I just met a lot of people in a lot of different places, whether it was pertinent to what SunGard was doing or not. And it was during that job tenure that I ran into Amiram. Or actually, it was a webinar. I first met them because I knew some people at Rancher.

(Kevin at 00:12:54) And I joined one of the Rancher webinars. And if you don't know Rancher, they're a Kubernetes company. They just got acquired. I believe it was last year too. And they were a co-sponsor on that webinar. And when they did their pitch, I was like, This is brilliant. We had actually been thinking of something similar on our architecture team of using spot instances for this purpose. And we were like, Great. We don't have to build it. We can buy it. Awesome. We don't have the resources to build it. So I got in a conversation with them and we started using them there. And then over the course of the next year and a half or so, we eventually decided that my next step should be to jump over to the startup world and get out of the enterprise world.

(Kevin at 00:13:38) And here I am. Now, four and a half years later, still at Spot.

(Joel Beasley at 00:13:43) It's not really startup anymore. It's getting to scale, right?

(Kevin at 00:13:47) Yeah. I mean, we were acquired by NetApp last year, so we're in that weird—do you still consider us? We still consider us a startup within the NetApp umbrella. NetApp is very good. They let us execute on our roadmap and press forward, and it really doesn't feel like a different company today than it did a year and a half ago. So that's what's great is we still visualize ourselves as a startup even after being acquired. And we're still growing like a startup. Like, our goal is to two-x every year, and, you know, our goal for this year is to two-x over last year, and we did two-x over the past year before that. So we're still trying to grow like a startup. We're still trying to push the product and engineering and sales teams and everything like we did before acquisition.

(Kevin at 00:14:31) So I would still say we still very much feel like we're in the startup phase. But, yeah, I mean, we are at the larger entity, I would say, you know, strictly company speaking, right? We are under NetApp today.

(Joel Beasley at 00:14:45) Does NetApp have customers that you would want to sell to? Was that a useful part of the acquisition?

(Kevin at 00:14:52) Yeah, I'd say absolutely. NetApp, like every other company, is seeing the shift to the cloud. And whether customers are doing 100% shifts or even 50% shifts to the cloud, cloud is becoming—it's so funny. In the cloud space, we feel like cloud is so big.

(Kevin at 00:15:09) But in the total IT space, it's still actually kind of—I don't want to call it small because it's a big industry, but compared to on-prem spend, it's still pretty small compared to the total spend of all data centers and all IT spend that happens in the industry. So you have this big push of companies that didn't necessarily adopt cloud early or are moving to the cloud now or have teams that are moving. And so NetApp realized that, and NetApp realizes all their customers are moving to the cloud.

(Kevin at 00:15:39) And we can help. We talked about Elastigroup, but we have the other products too, Ocean and Eco. Eco specifically is a product that can help anybody moving to the cloud. It takes no engineering effort. It's all about managing those reserved instances that I talked about before, knowing when to buy and convert reserved instances from a cloud provider. That can be a complicated mess for a financial team and an engineering team to decipher between one another.

(Kevin at 00:16:04) Eco automatically just kind of takes that off of your hands. And especially for customers that are traditionally in data center and their first step is just to get to the cloud, we can meet them with Eco. And then when you're ready to become more cloud native or ready to become more scaling, we can introduce Elastigroup. We can introduce Ocean. And Ocean is that containers offering that I talked about in the beginning, which is built on top of Elastigroup, which just allows you to focus on containers instead of focusing on compute.

(Kevin at 00:16:33) So between those three products, we have a whole journey that we can sell to the NetApp customer base. And there are more and more cloud native people coming to NetApp too. So it's not only the customers that they've had traditionally for, you know, they've been around for what, thirty years or more now. It's not only those customers that are making the migration, but it's also the more cloud native people who are starting to use NetApp storage, which is natively integrated into the three cloud providers. So you can go to Google, and you can provision NetApp storage directly through Google APIs.

(Kevin at 00:17:06) You can go to Azure and directly provision NetApp storage directly through Azure APIs. You can go to AWS and provision NetApp storage directly through cloud native APIs through AWS. They're integrated into the cloud providers, and now we're adding compute and financial efficiency on top of that. So now we have storage and compute that we can add together in all three cloud providers.

(Joel Beasley at 00:17:30) You guys are growing. That's a lot. So what's happening? Is most of your new revenue from existing customers as you roll out these new products?

(Kevin at 00:17:39) Yeah. I mean, as you grow, there's a little bit of expansion and then there's a little bit of new logos. Now, we're always pushing for new logos. One of the great things about our platform is that customers grow with us. So we don't necessarily look to land these huge deals right off the bat.

(Kevin at 00:17:54) We'll land any deal. And that's because we know customers love the platform, and they'll grow. A lot of our revenue comes from expansion because it's a journey. It's not we don't just make a sale and then, okay, great. You got a year deal, and we move on. We're a consumption-based product. You need to use our products. We're a success-based pricing. We charge based on the amount of savings we provide over on-demand.

(Kevin at 00:18:20) So our job is to every single day deliver savings because if we're not delivering savings to you, we don't make any money. And so our job is to produce that. And then as customers get comfortable with the platform, they scale more. Our cohorts are very good. All of our cohorts from day one all grow over the span of, you know, one to three years before customers even start getting, you know, what we would consider maxed out on the platform.

(Kevin at 00:18:47) So we put a lot of time into making sure customers are happy post-sale, and we get a lot of our revenue through that expansion. That's where we make most of our relationships. But, of course, to get customers to expand, you need customers coming in the front end. So we—my sales team is listening. I love new logos also.

(Kevin at 00:19:08) Don't get me wrong. But we love making those—more than new logos, we love making those new logos happy and expanding on the platform. So it's kind of a long answer, but yes, a lot of our money comes from expansion, but that's because of how the platform works. You start small and then you get bigger.

(Joel Beasley at 00:19:24) Now we got to talk about the dog. What's the dog's name?

(Kevin at 00:19:26) Yeah. So the dog is Teddy, and he's right now protecting us from the evils of somebody walking their dog.

(Joel Beasley at 00:19:35) Yeah.

(Kevin at 00:19:35) Across the street. So I apologize for that.

(Joel Beasley at 00:19:39) No worries. They shouldn't be walking so close to your house, right?

(Joel Beasley at 00:19:45) I have a dog named Teddy as well. It's like a little—

(Kevin at 00:19:48) Do you?

(Joel Beasley at 00:19:48) Okay. Yeah.

(Joel Beasley at 00:19:49) I mean, it's my wife's dog, but it's a little Pomeranian.

(Kevin at 00:19:53) Oh, okay. Yeah. Teddy's a pointer. He's about 75 pounds. We have an older dog too, but he's not barking because he's passed out somewhere.

(Kevin at 00:20:03) He's—oh, no—he's 16 years old now. So he just kind of roams around the house and does whatever the hell he wants.

(Joel Beasley at 00:20:08) Yeah. Teddy.

(Joel Beasley at 00:20:09) I call it Teddy. Teddy's like the purse type dog, you know, like small little fluffy thing. But then we have Bentley who's a Bernese mountain dog and he's a big dog.

(Kevin at 00:20:18) So I like him.

(Joel Beasley at 00:20:18) Yeah. That's a big dog. Yeah. Yeah. Big one.

(Joel Beasley at 00:20:20) Yeah. That's exciting. So what's the culture—tell me what the culture's like at Spot.

(Kevin at 00:20:27) You know, we're like most tech startups. We move fast. We like to work on exciting things, and we like to deliver—you know, customer obsessed is the biggest culture value that we push. If we're serving our customers, then everything else kind of works itself out. We're global. So we're remote.

(Kevin at 00:20:46) You know, and then with COVID, we're all remote anyway, but we have a big global team that works together across all time zones. So the company started in Tel Aviv, but we've expanded, and now we have teams in Tel Aviv. We have teams in Iceland. We have teams in the United States. India, we have—I hope I'm not forgetting anybody.

(Kevin at 00:21:10) Ukraine, we also have teams out of. So we're distributed all around the world, and we try not to—we don't necessarily pocket single things in areas, although sometimes that happens, but we have people working with each other on products from all over the place. So we're a very diverse culture now. We're a very diverse work group. You know, we're, quote, unquote, around the globe, around the clock.

(Kevin at 00:21:38) And, yeah, we're just—you know, we're customer obsessed. As long as, you know, everything we do from product down to engineering, everybody's focused on how we're serving the customer. And when we're serving the customer, we're doing well and everybody's happy about what we're doing. So we just kind of keep that, you know, keep that moving, and we keep expanding the team, right? We're growing, and we keep expanding to new areas.

(Kevin at 00:22:02) So it's always fun and interesting to add another country, another place on the map, finding talent wherever talent is. I've been a remote fan for years. I've actually been somewhat remote working for more than fifteen years now. And I just really enjoy the fact of finding great talent wherever talent is and then figuring out how to integrate them into the team and then watching magic happen. So, yeah, I mean, that's where we are today.

(Kevin at 00:22:32) It's just amazing how you can grow a company nowadays with, you know, Zoom—how we were talking about at the beginning—and the collaboration tools that we have today and the messaging platforms and the way that we can share. It's just amazing how you can grow a team all over the world. I think I'm amazed every day at how we get things done.

(Joel Beasley at 00:22:52) Yeah. And it's so interesting too, as you were naming all the areas, I've got relationships in each one of those and I was like, the cultures are so different.

(Kevin at 00:22:59) They are.

(Joel Beasley at 00:23:00) Yeah.

(Joel Beasley at 00:23:01) I'm always a fan for some reason to stereotype. I like the Israeli people because they are so direct. I've never met an Israeli that's not incredibly direct and tells you exactly what they're thinking or how they're feeling. They're just like, they put it out there. And for me, I am so relaxed around those people because you always know what's going on with them.

(Kevin at 00:23:23) Tel Aviv is a great time. I recommend it to anybody who can go. It is a wonderful city. And yes, you ask a question, you get an answer. It's very refreshing.

(Kevin at 00:23:34) And the food in Tel Aviv is amazing. So if you ever get to go to Tel Aviv, I highly recommend it. And there's a few excellent kebab places that you can go. I'm actually missing—I used to go there about once a month before COVID, and now I don't get to go anymore.

(Kevin at 00:23:49) And so there's this place called Jasmino. And if you're Israeli and you're listening, I'm jealous. I need to get back.

(Joel Beasley at 00:23:58) I love it. I love it. So tell me about that. Is Israel locked down? Is the—why can't you go over there? Is it you don't want to, or is it locked down? I'm curious about what's going on.

(Kevin at 00:24:08) I think things—you know, all the rules are a little crazy right now. So if you're an Israeli citizen, obviously, or you have dual citizenship or something, you can get in and out. If you have family in Israel, you can get in and out. If you're not one of those, you have to be in a tour group.

(Kevin at 00:24:23) And what they do is they're keeping the tour groups together. So if you're going to Jerusalem or the—

(Joel Beasley at 00:24:29) Like a birthright trip.

(Joel Beasley at 00:24:30) Yeah. Yeah.

(Kevin at 00:24:31) Right? They'll keep you in a group, and they'll keep that group kind of isolated. But if you're just going for work, you need special permission. You got to go through all this stuff through the government. Once they open it up so I can fly in and fly out without the fear of being quarantined for two weeks, I have a wife and four kids at home.

(Kevin at 00:24:48) I don't need to end up being stuck in a hotel room for two weeks somewhere, you know, just in quarantine. So, you know, I'm all the things I need to be to travel, like vaccinated and all that stuff. So whenever it does open up again, you know, I'll get out there and I'll do it. I just need to make sure that I can get out and, more importantly, get back. But right now, they're not really letting in—I think it's particular countries.

(Kevin at 00:25:14) I think they're going to open it up for the US soon. It's hard to keep track of all the countries and all the rules, but since the US just opened up more travel for countries into the US, I think those countries are going to turn around now and allow travel back the other way, right? So I think it should be soon.

(Joel Beasley at 00:25:34) I would be a little bit nervous. So my kids, they have little friends that they play with. So this one couple is originally from England, so all their family's over there. And it's been really difficult. For a year, the grandparents couldn't even come over and see them.

(Joel Beasley at 00:25:51) And then they were able to go over and see them, but had to quarantine. At one point they went over there and then the rules changed while they were over there.

(Joel Beasley at 00:25:58) So for me, when I'm talking about traveling internationally, it's like, I'm okay with the health because I've got a great immune system. I'm okay with that. But what I'm not okay with is being stuck in a country that changes the rules right when I land. That would not be fun.

(Kevin at 00:26:12) Yep. Yeah. And that's exactly where I'm at. I just want the confidence that if I go in and I have all the right paperwork, I can get in and I can get out. And getting out is kind of the main part.

(Kevin at 00:26:22) We did an acquisition. You can look it up. We did an acquisition of Data Mechanics. I forgot to list this as a country. They'll kill me if they—but Paris and France is also on our list now.

(Kevin at 00:26:31) So I was supposed to go into Paris, but there was no—no one really knew if you could get out of Paris. They knew you could get in, but getting out was a question mark. And so we did everything virtually, but I need to get out there and see that team as well. I need to get out to Paris. So I have a lot of places on my list.

(Kevin at 00:26:48) I got a big checklist that I'm keeping. And then maybe 2022 or something like that, I can actually get out and start checking things off the list. I was at KubeCon last week, so I did get on a within the US flight, which was neat. The conference was a little light attended in the Expo Hall and stuff, but it was really good to get out and just kind of communicate with people. Again, I got to meet a lot of people I hadn't seen in two years.

(Kevin at 00:27:16) And, of course, we had masks on and everything else in the convention center, so abiding by the rules. But it was really good to just see people again. It really was. It was so nice just to see people. Yeah.

(Joel Beasley at 00:27:27) Yeah. Yeah. It's good to see people. It's almost like we're social animals and we need each other. It's weird.

(Joel Beasley at 00:27:32) Right. Yeah. I know. We are.

(Joel Beasley at 00:27:36) So let's bring it back. I'm curious to know, typically, you've got a couple people involved in a decision-making process, someone who's going to buy Spot, right? But then you've got the people that are involved in the decision-making process. And then, typically, the person that approves the budget or is leading it, and then you've got the users of it who are actually using your interfaces and things of that nature. Tell me a little bit about who those people are that are actually using—what role are the actual people that are in Spot configuring it?

(Joel Beasley at 00:28:06) Where do they work at your customers?

(Kevin at 00:28:08) Yep. So for Ocean and Elastigroup, it's a DevOps engineer. We make DevOps teams' lives easier. That's where we're focused. We sell to the DevOps engineer and then we help the DevOps engineer sell up within their organization.

(Kevin at 00:28:20) So if they need approvals for the budget or they need us to help them pitch it to their boss for whatever reason or whatever approvals. But we sell to the DevOps engineer. It's a bottom-up sale. It's a very technical product that helps a specific group of individuals support their business at a much larger level than they're supporting it today, right?

(Kevin at 00:28:42) For Eco, we sell into the cloud director or your financial, whoever's in charge of the financial purse strings of the cloud, because they're the ones who are usually in charge of reserve commitments. So it's a little bit different sales cycle over there where we're selling more into management, and management already has the decision-making power. So yeah, we got two sales motions. We have that management with Eco, and then we have the bottom-up sale directly to the DevOps engineer with Ocean and Elastigroup.

(Joel Beasley at 00:29:12) All right. Let's say I've got a setup. Maybe I'm getting a couple million people coming to my site a day. Uptime, super important. And I would love to save money, but I feel like a salesman right now, but I'm running a hypothetical situation.

(Kevin at 00:29:27) Right, right. Okay. Let's do it.

(Kevin at 00:29:30) Yeah.

(Joel Beasley at 00:29:30) So I want to do it. I think it would be great to save money, but I gotta be honest with you, the whole trust factor is huge. You guys are going to be automating my entire infrastructure. You could take me down. It could be chaos. How do you get over that hump with clients?

(Kevin at 00:29:49) Yeah. In the beginning, it was harder. Now that we have a lot of referenceable customers, it's a little bit easier. When we give a reference like Ticketmaster, and when you go buy tickets, Ocean's behind that somewhere, right? And we can actually tell people, you know, there's these big customers out there that you can that will get on stage with us and tell us how well the platform works at scale. In the beginning, it was definitely, let's get in the dev environment. Let's show you how we work in your dev environment, your staging environment. Let's do a smaller application in production and then eventually grow. So we're perfectly fine with stepping in and taking over a smaller section of your cloud.

(Kevin at 00:30:24) In fact, sometimes your dev environments, depending on what you're programming and what you're doing, some of these dev environments can cost a ton of money in the cloud. So let's get into the dev environment first. Let's get into the staging environment. When you become comfortable there, let's go right to production. Now I will say because, you know, we got five years behind us and we have a lot of great case studies and a lot of great use cases, we can get into production environments a lot faster today based on our references and based on who we can bring to the table to ease those concerns.

(Kevin at 00:30:52) You know, we stand by our SLA. We have excellent support. If you reach out, there's a chat button on our website. If you hit that chat button, in thirty seconds, someone's gonna respond to you. An actual human will respond to you instantly, and you'll start getting help on whatever it is.

(Kevin at 00:31:08) So if anything in the platform even seems just a tad off to you or something in your group or configuration or anything like that, you are always thirty seconds away from getting help from someone who can actually help you. And that's a big, big part to helping customers grow on our platform is having this excellent technical support team and an excellent NOC team that can sit there and monitor everything, and I mean every second of every day, and respond instantaneously to customers coming through the platform. That also helps ease concerns. And it also helps customers even more with just Spot. If you have a question about anything, we answer literally any question on that chat button.

(Kevin at 00:31:47) You can ask us about your cloud setup. You can ask us about another service that doesn't even run through us, and we'll work with you on the answer and get it configured. So the power of those things, now that we have the years behind us, those references, the power of our technical support, the proof of what our platform's been able to do over five years, all of that goes into helping, you know, run in production. And then if you're still not there, let's start small. We'll start small and grow.

(Kevin at 00:32:15) I guarantee we'll grow once you start using it.

(Joel Beasley at 00:32:18) I like that you have people all over the world because that means you've got talent everywhere that can—you don't want to wait for advanced support to come online, right? You have people everywhere. That's pretty cool.

(Kevin at 00:32:31) Yeah. Every now and then, there's a team somewhere, but we make our—again, part of being customer success-based is always having someone available. So if something needs to get escalated at a certain time, there's always someone whose job it is to make sure that that gets answered. And we're very, very stringent about that. We always make sure that the customer will get an answer in the right amount of time.

(Kevin at 00:32:54) So really, no matter where that person is in the world, and we're bigger now, so we're not single-threaded in most places. Multiple people can answer those questions. And we have a really, really good policy in place to make sure those questions get answered by the right person.

(Joel Beasley at 00:33:09) That's exciting. Are you loving what you do, Kevin?

(Kevin at 00:33:13) Yeah. Yeah. I mean, this has been a really, really great run, especially, you know, under Spot, starting with a team that was, I don't know, not even 20 people at a time, and now growing to a team with the most recent announcement of the acquisition of Data Mechanics and now with the intent to acquire CloudCheckr. You know, it looks like we're gonna be more than 300 people very soon. So being on that journey and not only being on that journey, but having a product that customers just love to use, that's the most exciting part.

(Kevin at 00:33:47) Right? I don't say this as a sales pitch. I really don't. Customers come on and they start using a small amount of your product, and then all of a sudden they're using a huge amount of your product. And you watch those cohorts grow quarter after quarter, and you see people investing in you as much as you're investing in them, and you become partners with these companies over the years.

(Kevin at 00:34:09) And I just think it's a really exciting product, and it's not something that, you know, we built this really technical thing, and then we got a sales team to sell it. And, you know, I've been in other enterprises in the past where it's like, you built a product and then sales sells it, and there's always this tension between, you know, sales and engineering, what's gonna be built next and things like that. We don't have that here. We don't have that yet here. Everyone still feels like we're just running fast and building things that customers want.

(Kevin at 00:34:40) And, you know, it's just a lot of fun to be there. Right? It's a lot of fun that sales is still very close to engineering and marketing, and everybody's still working together. And we have these sessions where, you know, a customer comes in and asks us something, and we realize that there's a whole new feature or a whole new thing that we can build, and everyone just bands together and does it. So it's still a very exciting place to work for me, and I really, really enjoy every time these teams grow, to see how they grow and to see how people have grown with the company, to see the people that we started with at 20 people growing into managers and managers of managers, like you were talking about before.

(Kevin at 00:35:17) Right? You start with managing, and then you become a manager of managers, and like, that career progression and watching people on that career progression as well. You kind of get a little bit of everything here. And it has been really exciting. I'm really happy to be here. I'm really happy to be working with the people that we're working with.

(Kevin at 00:35:35) I think we have a great team.

(Joel Beasley at 00:35:37) It's spot.io/careers. One of the things you do that I love, I recommend to people all the time, is putting your company values on your about page or your careers page, because they really say—I think the stigma of them being corporate things that are just sent down has kind of been eroded a little bit. I think now people are trusting that I can go look at these values these people hold, and they really do hold these values, which I like.

(Kevin at 00:36:06) Yep. And we adhere to them too, right? Customer obsession, getting things done, right? These are things that I talk about from a management perspective, and these are things that we build out in our interview process. And these are not just words on the screen, but things that we hold true, you know, day to day. Now we don't necessarily put them up on the whiteboard behind us or on the virtual screen behind us every day, but they are things that we talk about. And there are things—like, when we have a conflict, we try and resolve it with the value, right?

(Kevin at 00:36:40) Like, what is the value of Spot? Were we being customer success? Were we trying to get things done? Were we being as professional as we think that we should be? Right? If you relate it to that and you have a North Star of why you're doing something, it's a lot easier to make a decision at the end of the day because I can point to a thing and go, you know, we're doing this because of this, right? And so it makes it very clear to the organization of what we're doing and why we're doing it.

(Joel Beasley at 00:37:06) You said something earlier called data gravity, and I've never heard that before. Can you explain that to me?

(Kevin at 00:37:13) Ah, yeah. Data gravity is one of the reasons why multi-cloud is still kind of a hard thing, right? And so where's your data, and how does that data move, right? So if you move data from AWS to Google or vice versa from Google to AWS, that costs a lot, right? So there's a—it's always—you can always put data into a cloud. That's not the problem. It's getting data back out of the cloud.

(Kevin at 00:37:37) And if you operate within a cloud, it's not a big deal. So if your application needs to churn on a bunch of data and it's within a cloud, you're really not paying for it. But once the egress cost happens and you start moving data between clouds, you start paying a lot of money for that data just moving around the internet, right? So data gravity is where is your data? Does the application need that data to operate? And then how much does it cost the further away from that data that you get? So it's data gravity because the closer you are to the data, the more efficient and the lower cost the application runs at. But the further away you get from it, the more expensive and more latent it is, right?

(Joel Beasley at 00:38:15) That makes sense.

(Kevin at 00:38:16) A lot of times, if you have a lot of data in one cloud, you're operating in that cloud for that application, and then you might have a certain subset of that data in another cloud, or maybe you have a completely different dataset in another cloud, and you run that application in that cloud. Now there are applications that aren't data-centric, kind of run everywhere. That's applications that run on CDNs and distribution networks and things like that. A little bit different architecture in what you're doing there. But think about it.

(Kevin at 00:38:44) If you have a database and that database—let's take a MySQL database—and you have that MySQL database planted in one of the clouds, if you wanna replicate that database, you gotta pay the egress cost to get it out to the other cloud. Is that as efficient as just using another region in the cloud provider that you're already in? These are the questions you have to answer for yourself.

(Joel Beasley at 00:39:04) I love it. Before—I know we're close on time here, right? So I think we have about ten minutes left, but I wanna get some leadership insight. You said you're a dad. You've got a family. You've done very well in your career. What sort of—let's say you've got some new fathers out there that are listening to this. What sort of insight or sage wisdom do you have for them?

(Kevin at 00:39:28) All right. Not to sound too—I ask my kid—I one thing I do is I tell my kids that they're required to learn one thing every day. And, of course, you always learn things. But at the dinner table, I ask them, what's the one thing you learned today, right? And I just want them to articulate that they learned something or they went out of their way to learn something. One of the most important things that I understood was that I had to learn how to learn when I was younger, when I was very opinionated when I was much younger. But then I eventually, you know, realized that it doesn't matter what you're trying to do. You're always gonna have to learn how to do something you don't wanna do, right?

(Kevin at 00:40:10) That's just part of life. So just the constant learning and, like, driving home to your kids that learning is part of life, and every day you should learn something no matter how old you are or where you are. That's what I try and drive home with that statement. So I ask my kids all the time, like, what's one thing you learned today? I don't even care if it's about school. They don't have to tell me that it was math or something else. I accept from my oldest son that, you know, it was—I can't even—some Nintendo Switch game that he was able to find something or zombie or something and figure out, you know, where it was unlocked. But if he can bring me through it and he can bring me through what he learned at that point in time and it's different, you know, he picks something up that he didn't know before, that's fine. So that's my sage, my old man advice. That's what my four kids have to hear from me on a daily basis.

(Kevin at 00:40:59) So they have to be able to answer what they learned any given day.

(Joel Beasley at 00:41:04) I just have one more question for you because we always like to ask this question. It helps our audience. We'll put it in newsletters and things like that. But if you were to give advice to a technical leader who wants to take on some more responsibility, right? They're listening to the show, so they can listen to people like you who've gone very far in their career, and they're ready to take that next step. What sort of insight or advice do you have for them?

(Kevin at 00:41:28) That's tough. I think the toughest thing to become a technical leader is to start becoming more hands-off. That was the toughest thing for me is to realize that you set direction, but not necessarily implementation. You need to be—some implementation you wanna get involved in, but you have to start—if you wanna be a leader, you need to start setting the direction of where you want the team to go and not how the team needs to get there and allow them to do that. And I think that's one of the hardest things to learn as you go higher up the ladder, so to speak, and realize maybe your way is not necessarily the best way.

(Kevin at 00:42:05) There are more ways to get it done. But pick and choose your battles. Every now and then, you need to reach down and say, no, we're gonna do it this way. But pick. Don't be that person all the time is what I'm saying. Let the team decide and come in for the things that you truly believe matter and then put your hand down on that. And, yeah, I mean, the hardest thing for me becoming the leader of where I am now is I'm not a good engineer anymore. I don't code well anymore. I don't sit down in VIM, which is my day. I don't sit down in Visual Studio Code.

(Kevin at 00:42:38) I'm just—I'm not there, but I'm a soundboard for everybody else who is, and I can still add a lot of value there. And that's what I do.

(Joel Beasley at 00:42:45) And some of the principles transcend the language that you'll remember forever, right?

(Kevin at 00:42:50) Oh, yeah. The principles transcend—like, language is an implementation tool, and don't get me wrong, you need to be a master at your craft if that's what you're doing every day. But as you move up the ladder, in most scenarios, you need to become more of a directional technology leader and not an implementation leader. So that's a tough one. That's a tough one.

(Joel Beasley at 00:43:09) It's hard, dude. It was the hardest for me because I was trained to do, do, do, do, do, you know, grab the ticket, do it, get it done, ship it, get it out the door. And then to explain something was really inefficient when I could just do it myself faster. And then you realized, if you can get five people to go in a direction, they can accomplish more in a day than you can in a week.

(Kevin at 00:43:32) Exactly. Yeah.

(Joel Beasley at 00:43:33) And then that's when you start getting the economies of scale and then you sort of build trust with yourself to let go a little bit more. And then you realize how much work it is to actually do the strategy and bringing people together to achieve a goal. And you're like, wow, this is a full-time job.

(Joel Beasley at 00:43:48) Yeah.

(Kevin at 00:43:50) Yeah. And I think you also need to be real honest with yourself if that's something you want to do. I hate saying one thing that we try and do at Spot is not always have the next progression level be management. There are some people who are very good, and they want to be individual contributors. And I think we, in the industry, we need to embrace that more.

(Kevin at 00:44:10) Not everyone has to be a manager. You don't have to be a manager to always get to the next step. I think there's really great individual contributor jobs, whether it's an architect or whether it's an influencer or whether it's a mentor. And you can have people still graduate to larger and larger responsibilities within an organization without managing. I think we miss that sometimes, and I think that's very important to still have in an organization.

(Joel Beasley at 00:44:36) I fully agree. It's partly due to our culture too. Aside from just the technology industry, it's a bonus to go into management. That's the theme in just the atmosphere is that's the next step. So to explain to your parents who aren't engineers that I'm going to potentially even make more money staying on this individual contributor path, and I'm going to write books and be an expert in this, maybe Kubernetes or Python or whatever it is, and speak at conferences around the world.

(Joel Beasley at 00:45:02) To explain that is kind of a little bit difficult, because they want to see you make progression in their sense of what progression is versus what's reality for your role.

(Kevin at 00:45:12) I'll say management, it is a big broad statement, but management is always there. Once you take your fingers away from the keyboard, it's hard to get back. I was in IC for more than a decade to start my career, over a decade to start my career. And then as you start getting there, your fingers start getting slower and slower away as you manage people and you manage business.

(Kevin at 00:45:35) And, you know, the CTO role that I do today is more about how do we interface technology with sales and marketing and set direction and then put the right leaders in place and things like that. And it's not about that code anymore, so you've got to be really comfortable with that. Because I truly believe you go a long way with keeping your fingers on the keyboard, so to speak, and then deciding that you want to take that next step. Because once you take that next step, it's really hard to go back the other way. I think it's super hard to go back.

(Joel Beasley at 00:46:08) Thank you so much for listening. And if you found this episode useful, please share it with a friend or colleague who you think would get value from it. And if you have topics that you'd like to hear discussed on the podcast, either add me on LinkedIn or send me an email, [email protected]. Every time I get an email or LinkedIn message, it absolutely makes my day and inspires me to keep going.