Overview

Get started today
Replay past traffic, gain confidence in optimizations, and elevate performance.

The Atlanta Startup Podcast is the briefing room for the innovation ecosystem, featuring the investors, founders and activators creating the fastest emerging venture capital ecosystem in the country. Ken Ahrens, founder and CEO of Speedscale sits down with Valor Investor, William Leonard in this episode.

Podcast Source:

https://atlantastartuppodcast.com/ken-ahrens-founder-ceo-at-speedscale/

Ken founded Speedscale in 2020 with the ultimate ambition of developing a more efficient solution to prevent production incidents with automation. Hear Ken’s journey of starting Speedscale, his advice on developing a GTM strategy, and his take on industry/geographical trends that serve as tailwinds for Speedscale in this conversation with Valor Investor, William Leonard.

Transcript:

William Leonard

Hey, ladies and gents. Welcome back to the Atlanta Startup Podcast. I’m William Leonard, your host, and investor at Valor Ventures, a leading seed-stage VC firm here in Atlanta, Georgia. today, it’s my pleasure to welcome Ken Ahrens, founder and CEO of Speedscale, to the podcast. Ken, thanks for joining me today.

Ken Ahrens

Thanks a lot, William. I appreciate the invitation. Looking forward to our discussion.

William Leonard

Awesome, awesome. Well, let’s kick it off here. Excited to learn and equally have our listeners learn a little bit more about your path to entrepreneurship.

Ken Ahrens

Sure. My path, I think, is perhaps a little different from typical entrepreneurs. Nowadays, you hear folks saying early on at a young age, they realize there’s some problem and they drop out of college to start a company. That’s how you hear about the Facebooks and Microsofts of the world. I actually began working at startups early on even when I was an undergrad at Georgia Tech. I interned for many years at a startup here in Atlanta, and it got into my DNA. I really liked working at small companies and was able to be part of a couple of great companies like ILC here in Atlanta, a company called Wily Technology that was acquired by CA Technologies. Another company called iTKO was acquired by CA Technologies. I realized startups were a great place for me because I love the idea of having a lot of impacts and being able to really affect the company with some positive changes. In 2019, I knew it was time, everything was ready, the stars were all lining up. I pitched my co-founders on taking the plunge and in that we would finally start our own startup, which we ended up agreeing to and then officially kicked off the company in January 2020.

William Leonard

Awesome.. That’s a unique path there having worked at startups really right out of college. You went to Georgia Tech in the Atlanta area, talk to us about your experience at Georgia Tech as a student and how you think it helped shape you into the entrepreneur you are today.

Ken Ahrens

I grew up in the Bay Area, actually San Jose, California. My parents still live there, right down the street from Apple. Actually, it’s funny. When it came to Georgia Tech, my confidence was sky-high. Georgia Tech was so challenging and academically so difficult. It actually was this Co-Op program that made a huge difference for me, because I realized, although the studies were difficult, I was good at work. I may not understand electromagnetics but I could write some code that people would run, deploy, and use. For me, it was fantastic that I got a lot of opportunities through Georgia Tech and through the network, but also that I was able to apply it frankly, immediately to work. It was a huge boost. I ended up, when I graduated, working for that same startup where I interned and I stayed there for several years.

William Leonard

Got it. Can you give us the 90-second rundown of Speedscale. What is Speedscale doing? What problem are you solving? Talk to us a little bit more about that.

Ken Ahrens

Sure, William. In a lot of my career, I’ve worked in the application monitoring world where modern apps, everyone knows what these are nowadays, but people depend on these apps running and running really well. I’ve had a lot of experience working with folks where they’re having problems. We’ve got some monitoring that tells them their app is broken. It’s slow, it’s offline, that kind of thing. It’s great to know that this is happening and to try to get it repaired and back online, but you’re having an outage. An outage is one of the most stressful things from a work standpoint. You just wrote all this code, you pushed it out, or maybe the outage is happening in the middle of the night. We knew we wanted to work on this problem and help companies prevent outages. What’s going on is, everyone wants to deliver code as fast as possible. You want to use new technologies like cloud technologies and assemble these applications. But it makes them very complex. They’re super complex, and you ship them all the time. What Speedscale does is we actually understand how your application normally performs, we capture the traffic, and let you play it back against your new version of your code. This is exactly what it will have when it goes to production but we do it in your test environment. If it blows up, actually, you hope it blows up, because that means you just prevented a problem from impacting your customer. It’s that same thing we’ve been working on from a monitoring perspective, but we’re bringing it into the pre-production environment in a safe way so that you know when you’re making mistakes and prevent them from impacting your customers.

William Leonard

That makes sense. How was this solution really being remedied prior to speeds going into the market? What were you seeing currently on the market that was actively solving this problem, or maybe not actively solving this problem, and you saw an opportunity here?

Ken Ahrens

There’s a couple of different alternatives to using something like Speedscale. One is, like I mentioned, from the application monitoring world, you can do things when you deploy your application to limit how many users will use the new version. There’s a lot of best practices around how to do this, actually, you still should do that. They call them canary deploys. 2% of the users will get the new version, and then you can closely monitor it and determine if it is bad, don’t release it to everyone. It’s still a good idea, you should do that, unfortunately, that 2% or that 20% of users may not give you a good signal. You may actually think your code is good but when you roll it out to everybody, it still breaks. The other sort of alternative is if you want to pull back on speed a little bit. If you say we’re not going to release quite as fast, we’re going to spend more time doing testing and building quality into our process, some companies do do that but there’s always this tension with the business, which is we’ve got to get the functionality out and testing is frequently dropped. What we’re trying to do is fix all these things by giving you something that’s super fast. You’re not gonna have that same tension with the business, you can still ship quickly, but also give you that really good signal of whether the code is good or not. We fit right in between those two if it makes sense.

William Leonard

It does make perfect sense. As you were thinking about who your ideal customer is with Speedscale, how would you define that profile? Has that ideal customer transitioned a little bit? Or has it evolved since starting a business a little over a year ago?

Ken Ahrens

It’s funny you asked this question because when you start the business, there are certain things you say, “Okay, these are our ideas.” Of course, the ideas are cheap. We’re going to go change them over time, test them and learn from them. But one of the first that we worked on was who’s the target market for this. We thought it was going to be larger enterprise companies, they have lots of engineering teams, they use a lot of these cloud services, and everyone wants to get in the environment at the same time. That has proven to be true. We’re getting great signals from these kinds of enterprise companies who say this is a problem. Actually, we have a lot of familiarity with these kinds of companies. If you go down kind of a smaller company, a mid-market type of company, there are some of these mid-market companies that have really big tech environments. Actually, you can think of a lot of the sort of Atlanta unicorn type profile company, they’re not huge yet, but they still have a big burden of delivering things fast. They also like our technology because they don’t have time to go hire a ton of people to do all of this testing work. Those were once we identified, then we ran some experiments around startups, startups love using cloud technology. Startups love testing things out and while we have some traction with startups, we’re finding the problems just not as big. If you’re a startup and you have an incident with your system, but you don’t have that many customers, there’s not a lot of pain. Also, there tend to not be as many engineers. It’s not as big of a problem of miscommunication. But when your team gets big enough, this miscommunication falls off a cliff, right? You go, “Oh, five people, we can all figure it out.” 10 people, suddenly, no one can get on the same page. These are some of the things we learned about our sort of target customer profile as we were getting going.

William Leonard

Absolutely. As you’re thinking about where Speedscale is fitting into the equation for your customer, can you maybe share with us a high-level customer success story of how you’ve helped improve decision-making within these testing environments?

Ken Ahrens

We’re working with a company, kind of one of these mid-market companies, that is fast-growing startups called Nylas. They do email synchronization. They can pull the emails down from Google Gmail, G Suite, or Office 365, from tons and tons of employees, think of Support people. Everyone in support, we have all their email synchronized. We know everyone who communicated with a given customer, that’s what their tech does. They are running into this problem that they want to re-platform their system for the next level of scale as they scale up but they needed to ensure that although they’re changing a lot of stuff inside their code, that it still all works correctly. We’re doing traffic replay with them, where we’ve collected all this data about, for example, how they call Office 365. We’ll also simulate these “what if scenarios”. What if you cranked it up to 10,000 concurrent emails that you’re trying to synchronize? They use our technology to simulate that environment. Actually, they’re even using our reports and showing them to their customers to prove their scalability.

William Leonard

Got it. Across how many customers would you say that you’ve seen a direct improvement right after implementing the Speedscale solution?

Ken Ahrens

We’re one of the things like I mentioned earlier if you want to move fast, and you want to use these kinds of cloud, native technologies, and Kubernetes, and that kind of thing, you tend to not know a ton about what’s going on in your non-production environment. That’s the first “Aha!” We show you. everything that’s going on in the environment, every detail about how you call this service, and how it responds. That’s always the first quick win. It’s kind of funny because we don’t make the data up, we collect it, and then show it back in there. There are always these questions like, “Oh, no, it doesn’t the application doesn’t work that way.” It really calls to this. We didn’t make it up, it was observed. That’s always sort of the first “Aha!” People don’t realize, this is calling that. “Oh, it’s not supposed to.” “Oh, I didn’t realize it works this way.” Because of your architecture diagram, you wrote it last week, and it’s out of date, right? That is always what we’re getting that kind of the first day when they install the software. Then there’s this sort of second push. After we have some of the traffic snapshots configured well, they begin improving the quality of their own application, which if you’re running tests, you’re running replays, and you’re getting this signal out of your code, you start finding things and you start improving things. That’s actually one of our things is we’re realizing it’s this two-part process of first, I may not be able to reduce the complexity. It’s very hard to reduce the complexity of these applications but we can increase the information. That’s what we do. That’s our first win. But then the second is, it’s hard to build quality into these large organizations. Ours is an easy button, you just hit it, and it’s and you get some quality procedures going in your own company. It’s cool.

William Leonard

No, I love that illustration you just gave. It’s an easy button that is going to immediately and quickly improve the quality. That’s great. Now transitioning here a little bit, in your introduction, you mentioned that you have several operational experiences after you graduated from university. I’m curious, across those experiences, looking retrospectively, was there a time or role where you felt most prepared for life as an entrepreneur and building Speedscale?

Ken Ahrens

I spent a lot of my career in customer-facing roles. Actually, almost my entire career has been spent in the field. But a funny thing in 2019, when I was thinking this will be the year to start the company, I hit a million miles on Delta Airlines. I pretty much just travel in North America. That’s a lot of customer-facing time. Those roles give you a lot of empathy to understand, what are the problems that a customer is encountering? Am I correct when I think I know what their pain is? Figuring out how to work with them so that you build the correct solution. It can be easy, especially for a tech startup to kind of fall in love with your own tech. It’s so cool, it does this, it does that, here are the features. But if you don’t have that connection with the customer, you can get out of sync. This is something we’re constantly trying to make sure we do. We’re going to build something new, let’s get the feedback first. Let’s make sure we understand that it’s something that they need and that they want. I would say we’re, it feels a little unnatural, because you want to kind of sit down and code for a long time and build the world’s greatest thing. But if you might build it, and it’s the wrong thing, customers won’t use it. I think all this time spent with customers, that’s a critical thing. How much of my time am I able to spend with customers? How many different perspectives? Can we get on this thing? Of course, we want to ship quickly and so on but it’s better to take the extra time and ship the right thing than ship fast and then realize that no one will use it.

William Leonard

Correct. Definitely, I think being in that customer-facing, hands-on role is really going to like you said, allow you to develop empathy for your customers and find out the true pain points which are going to help you really develop your Go to market strategy accurately and concisely. On that topic there, you’ve been building Speedscale for a minute now. How are you thinking about your Go to market strategy? Do you have any other advice for founders who are early stage, maybe pre-seed thinking about Go to market? How can they adequately find customers and choose the most ideal customer profile?

Ken Ahrens

I think Go to market is very challenging to get right. Kinda like I said earlier about you putting your ideas out there and all the ideas are open to being questioned. I think that’s a smart way. I try to always act that way. But when it comes to the Go to market side, one of our big debates that we were trying to have with ours is the first SaaS technology, like we have a B2B SaaS. Do you want to have a bottom-up type of sales momentum, there are many companies that use a bottom-up. As a matter of fact, it’s the most popular way right now to build a SaaS startup. That might mean you have some free component, you might have some open source component, you might offer trials, but it also means you have to build an enormous funnel. You’ve got to have a ton of people who are coming into your platform, and you might lose people as they progress through the funnel. You’ve got to make it as easy as possible to try. On the opposite side, of course, the top-down is a more traditional approach where you’re more careful about how you find your customers and ensuring that you’re able to get them into a trial and that kind of thing. We had a huge debate when we were starting the company. Are we bottom-up? Are we top-down? Are we both? We ran some experiments around this to try to understand and figured out we mostly had experience with top-down. Even though that’s not the sort of popular approach at the moment, our thinking was, “Let’s go with our strengths. We’re building a technology that solves a really big and expensive problem. It’s okay if it costs something for the customer, that it’s not a nominal fee, that it’s a more expensive type of solution.” The last key part of a top-down is there has to be clear ROI for your customer. If you come in and say, “We’re going to charge you an enormous amount, but the value is really small.”, you’re going to have a problem. This ends up getting mixed. Because if you’re not honest with yourself, you get confused. I would say we weren’t honest with ourselves early on. We said, “No, we can do this bottom-up, we’ll work like this.” Then we actually came back and said, “No, we’re doing it correctly, we’re doing top-down.” We’re solving a hugely expensive problem. That means for customers, it will cost a little bit more. It’s also expensive for us to deploy, it’s a big solution. As a matter of fact, we’ve had to kind of increase prices as we went along, because we started a little too low, but spending time and focusing on that, and making sure that you understand your motion is really important.

William Leonard

As you were having these internal debates with your team about the top-down, or the bottom-up approach, was there a point of validation within the company where you said, “Okay, we are going to stick with the top-down approach.” Because I know you were kind of on the fence a little bit about the bottom-up or top-down.

Ken Ahrens

One thing you might have heard me mentioned was an experiment, whenever you can design an experiment, it helps you test this. We designed a couple of experiments for the bottom-up, and we got people to try a different version of our software that would be much simpler to test, run, and deploy on their own. Then, of course, we were running experiments on the top-down, which was more familiar for us. Actually, the outputs of those experiments were the key finding for us that we didn’t have this kind of thing. Unfortunately, all the recent advice in all kinds of big successful companies, not all of them, but many of them. If you think of Atlanta companies like Calendly, one of the world-class bottom-up motions, you can go on their website and sign up. We use Calendly, it’s awesome. We don’t have a Calendly type of product, it’s where we’re not able to deliver that kind of thing. Ours is too complex. If it makes sense, going back to my customer point, we use customers’ experiments in data to help us answer these questions.

William Leonard

Awesome. Awesome. Building a startup requires so many difficult conversations. Like you say, Go to market approaches, customer discovery but I’m curious, fundraising is an element of the business that it’s mandatory. It’s daunting. It’s time-consuming. What advice do you have for other founders who are early stage pre-seed, seed, fundraising, and really how to navigate this process in an artful fashion?

Ken Ahrens

We were very fortunate that we were able to raise some funds. The world changed, obviously, in the last year with things being much more open and distributed. You don’t necessarily have to be co-located where all the investors are. When we decided we wanted to do fundraising, we said, “Let’s figure out our value. Let’s create a pitch deck. Let’s pay attention.” You have to treat this like a top-class project that you’re running along with everything else. This is what makes fundraising very difficult. We have three founders of the company so we were able to dedicate pretty much one full-time person. It wasn’t always the same person, actually. To this process, your first pitch deck won’t be very good. You need to test it with people where you can get real feedback, you need to take the feedback, improve your pitch, you need to identify the things that are important. To your point about Go to market, it’s one of the sections you should have in your pitch deck. It was the weakest section in our pitch deck in full transparency because we were on the fence about the bottom-up and the top-down. I’ll tell you that was the story we told it was both and we got feedback. That wasn’t correct. We underappreciated our own experience and background. The three founders, Nate and Matt, and I each have over 15 years of experience, and it’s directly applicable to the problem that we’re solving. We ended up actually showcasing that a little bit more; our backgrounds, the companies we worked at, we’re all Georgia Tech grads here in the Atlanta area. Of course, we supplemented the problem with solutions and these things. One is, you’ve got to have that iteration. You’ve got to dedicate the time to it. But the other is you have to get a lot of at-bats. You don’t know what’s going on with all of the VCs, they could be raising a new fund and not ready to make more investments. We’re the first investment in a new fund from one of our investors. Those things have nothing to do with you. You had to make up for that with volume. We treated it like a sales process, and identified a lot of investors, and had a lot of costs. Yeah, that part is daunting. But I would say my advice is, you’re probably going to have to have double the number of calls you’re thinking, if not five times but you can probably get the VCs to want to talk to you, too. Be prepared, allocate the time correctly.

William Leonard

Definitely. I love that approach of really dividing up the fundraising responsibilities among the team. I think that gives an opportunity for investors to learn more about the dynamic and in each role that one of the founders plays. That’s great advice there, especially around really harping on the Go to market strategy in the deck. I see a lot of decks. oftentimes that’s a very light talking point. it should be far more emphasized, I believe. Great advice there, Ken. You were in the Bay Area for a point in time in your career, what really influenced you to come back to the east coast to the Atlanta area, and build a startup here?

Ken Ahrens

It’s funny to say that. As I was graduating college, I was getting ready to go back to the Bay Area. I sort of had a picture in my own mind. I had interned with a company in the San Jose area. And of course, at this time, things were really on a huge growth curve until they weren’t. The the.com bust was happening, as I was graduating, and I had this sort of decision to make. I had interned with this company here in Atlanta, I could go back and sort of take a shot while in the Bay Area, it was a tough time. I thought, let me take this path with this current startup that I’m at and go into full-time work and that kind of thing. It was a great decision for me. As I mentioned, that company transitioned into this sort of heavy customer-facing role, which I was able to leverage for my entire career. Atlanta has a huge airport where you can travel most in North America with a single leg, as well as Atlanta’s geolocation. It’s a little further west than people realize. You can get to a lot of places. I remember I was working for a company at one point, I could fly to Dallas and arrive at the office at the same time as people who were driving from home. Sounds simple, but having Atlanta as a hub location actually works. I know at the moment, there’s not a ton of travel, but this will open back up. Atlanta has been tremendous for me. I leveraged that in multiple companies covering the East Coast, covering the Mid Atlantic area, the Midwest, and of course, the southeast. Atlanta was a great place for me to be stationed from.

William Leonard

Definitely. Atlanta is growing as a city. Microsoft is building a hub here, Airbnb, Google just announced they’re building a tower in Midtown. I think Atlanta and really the southeast ecosystem is going to solidify itself as a tech hub on the east coast. We’re seeing a lot of companies from California, Texas, high tax areas relocating here and bringing talent along with them as well.

Ken Ahrens

My previous company before I started Speedscale was a company called New Relic. A San Francisco-based company. As I said, I played the role of helping them to open the East Coast. They opened an office here in Atlanta. It was the second headquarters. I also saw PagerDuty similarly did this. These are both publicly traded companies but not the scale of Microsoft and Google, you actually see it all through from the largest companies down to these kinds of publicly traded but bigger companies. Of course, you know, the startup ecosystem has really kicked into gear in the past five years here in Atlanta.

William Leonard

Yeah, definitely. Playing into a little bit of that point, what are some of the industry or geographical trends that you think could potentially serve as a tailwind for Speedscale?

Ken Ahrens

Definitely, this one you just mentioned, is one that we’ve been paying attention to. Because you see, people are building engineering centers in particular here in Atlanta. We have a ton of engineering talent from Georgia Tech, Emory, and UGA. The schools are really great and turning out top talent. It’s been a little bit of a secret in the gym, although I think that it’s not a secret as much anymore, that’s not just Atlanta actually. Just down the road in Charlotte, have spent an enormous amount of time in Charlotte, of the banking center there or you go over to Raleigh, and all the engineers that they have in Raleigh in the top schools there, where there you can find lots of high-quality engineers everywhere. Actually, even Chattanooga sort of going the other direction. They wired up the whole city with broadband Internet, and people joke about that, that actually has enabled some tech startups there who can say I wouldn’t get an office. We can jumpstart by having this really fast internet connection. I think that one trend is that geographic trend. You’re also seeing some companies are moving their headquarters to Atlanta, or you see companies like NCR moved their building down to the Midtown area. Not to mention, by the way, we’ve got Home Depot, we’ve got Coca Cola, we’ve got Delta Airlines, we’ve got top big businesses, just up the road, you’ve got Bank of America, you’ve got Lowe’s of Charlotte. These are where the headquarters are. On the technology side, there’s a huge tech trend that we’re riding, which is this cloud-native approach. What companies are trying to do is to say it’s a version of the build versus buy decision. How much can I get from Amazon or Google, or Salesforce, one of these kinds of big tech vendors, can I use part of their technology to make my development go really fast? It’s fantastic for these companies because they don’t have to build such a big thing. You can sort of pay by the drink a little bit. It lets you get started smaller. That’s great. For day one, getting the project out the door. Day two, you start to encounter some of the problems that we work on and help with. It’s a huge enabling technology for us as cloud-native, Kubernetes, APIs. The growth of these is skyrocketing. That’s great for us because we help you with those.

William Leonard

Yeah, that’s awesome. I’m excited to really watch how Speedscale progresses and really solidifies itself as a growing startup here in the southeast. This is great, Ken. This was a very insightful conversation, I think our listeners will get a lot of value out of the advice and the perspective that you shared here. We’d love to have you back on the podcast, maybe in a few months here again.

Ken Ahrens

That would be great. I really enjoyed this, William. I appreciate the time today.

William Leonard

Of course, Ken. Take care.

Ken Ahrens

All right. Thank you. You, too.

Lisa

Thank you for listening to the Atlanta Startup Podcast. You know, we’re not just a podcast, we’re a community, and we’d love to see you at one of our digital or physical events, go to valor.VC and sign up for an event that makes sense for you. We have events for founders and the investors who back them. Another event you might enjoy is Startup Runway. The Startup Runway Foundation is a Valor organization that provides $10,000 grants to founders who are women or people of color building next-generation software products. Applications are free and we’d love to hear from you at startuprunway.org. And as always, thank you so much to the organizations that make this podcast possible. Not only Valor Ventures, but also Write2Market, a tech marketing and PR agency in Atlanta, Georgia, and the Startup Runway Foundation and Atlanta Tech Park Valley’s headquarters, and also headquarters for over 100 local entrepreneurs, building global businesses. See you next week. Please bookmark the podcast and join us.

Ensure performance of your Kubernetes apps at scale

Auto generate load tests, environments, and data with sanitized user traffic—and reduce manual effort by 80%
Start your free 30-day trial today