How a GitHub Experiment Became One of the Most Popular Network Analytics Platforms

What started as a weekend experiment became one of GitHub’s top network analytics projects — and the foundation for Elastiflow. CEO Robert Cowart shares how community-first growth shaped their GTM, why network data is the next big source of enterprise insight, and how focus on context over complexity is winning customers.

Written By: supervisor

0

How a GitHub Experiment Became One of the Most Popular Network Analytics Platforms

The following interview is a conversation we had with Robert Cowart, CEO & Co-Founder of ElastiFlow, on our podcast Category Visionaries. You can view the full episode here: $8 Million Raised to Power the Future of Network Performance and Security Analytics

Brett
Welcome to Category Visionaries, the show dedicated to exploring exciting visions for the future from the founders who are on the front lines. Building it. In each episode, we’ll speak with a visionary Founder who’s building a new category or reimagining an existing one. We’ll learn about the problem they solve, how their technology works, and unpack their vision for the future. I’m your host, Brett Stapper, CEO of Front Lines Media. Now let’s dive right into today’s episode. Hey, everyone, and thanks for listening. Today I’m speaking with Robert Cowart, CEO and Co-Founder of Elastiflow, a network performance and security analytics platform that’s raised 8 million in funding. Robert, welcome to the show. 


Robert Cowart
Thanks, Brett. Glad to be here. 


Brett
Yeah, super excited. Let’s go ahead and jump right in. Talk to us about what you’re building today. 


Robert Cowart
So if you think about this world that we currently live in, the things we do on a daily basis, the things we rely on, right, commerce, healthcare, entertainment, even social interaction, all applications that tend to be run online, run over networks. And so really, this world we live in is incredibly dependent on the availability, the performance, and increasingly today, even the more important aspect, the security of network infrastructure. And Elastiflow is an organization we built that’s 100% dedicated to ensuring that the people that engineer deploy, operate these networks we all depend on, have the insights that they need to ensure that this world we live in as possible. 


Brett
Talk to us about the AHA that led to the founding of the company. 


Robert Cowart
Interesting question. So, you know, I always have to explain this, even when we talk to investors, it’s like, I don’t know that there was this aha moment. The, you know, any successful endeavor is a ton of hard work and a little bit of dumb luck mixed in. And that’s kind of more the genesis of Elastiflow. If you think back to the 20 teens, there was this emergence of all these new data platforms, search engines, time series, databases, document stores, et cetera. And I had been doing network observability pretty much my entire career. And I started to ask myself the question, like, I wonder how some of these new technologies could apply to or help improve the space that I work in. And so I just spent time over, you know, maybe a year and a half period experimenting with a lot of different things. 


Robert Cowart
And in late 2016, really early 2017, one of the experiments that I attempted was putting network traffic records in a search engine, which that was elasticsearch. And I was actually able in the first weekend to build something that from a user experience actually was better than the previous company I had worked for. And I’m like, this kind of interesting. So I happened to a few months later be talking a meetup, sharing about the results of this experimentation I had done. And someone said, wow, that’s really interesting. Is this available for download anywhere? And I said, no, let me see what I can do. And a couple of weeks later, I created a GitHub repository and cleaned up the code a bit and stuck it on there. And I named the repository Elastiflow. And that was the first humble step. 


Robert Cowart
To this day, I still wonder what would have ever happened if that person had never asked me if this was available for download anywhere. So anyway, it was available. People could find it if they did and try to use it, which they, turns out they did. In fact, even today, although the original GitHub repository is not maintained anymore, if you go to GitHub, search on the term Netflow, which is one of the core technologies behind these network traffic records, you will and you sort by number of stars, that original repository is still like number four out of like 900 different things, you know, solutions that are there. And that’s when you start. 


Robert Cowart
So this is, you know, very early 2020, you start to realize, wow, like, there are literally thousands and thousands of people around the world trying to use this thing at scale in production. But it was always a science project to that point. It was never really intended for proper use. It was a proof of concept. And so I got together with my Co-Founder, who, quite frankly, is a better coder than I am. I always say it this way, I have a good grasp on what the user wants. He has a good grasp on how to build things. And so we got together and started working on the new solution, incorporated the company in September of 2020, launched the beta in December of 2020, and GA, the product was in March of 2021. And fast forward today. 


Robert Cowart
We’re a thriving organization with a growing customer base of pretty large, well known logos among our customers and a growing team that will soon be about 40 people. 


Brett
Jeff, from that point in September 2020 to the point that you had customers start paying you, how much time passed? 


Robert Cowart
Oh, not very much, actually. So one of the things about having that community is we had a number of people reach out of the community. Even before the code was done, I had been dropping hints like, oh, but I’m working on something new. Oh, there’s something new coming that’ll solve that problem or this problem with the original project. And so we literally had our first paying customers committed before we even had the beta out the door. So it was a valuable lesson in the value of community, because, quite frankly, without the community around the project, it would have been a completely different story getting started. 


Brett
Robert, what have you learned about community building? And the reason I ask, I feel like that’s the hot thing right now, right? Everywhere I look on LinkedIn, or every time I go into LinkedIn, everyone’s talking about community building. You need to have a community. I think community led growth was kind of a hot buzzword for a time there. What’s the key to building a thriving community? 


Robert Cowart
So I would have to say there’s a mindset to it. And I actually once heard Shy Bannon, who’s the Founder of Elastiflow, the company behind elasticsearch, which most of our customers use and users use to store their data. And he was talking about those early days of elasticsearch. And one of the things he had said was that essentially when you put a project out there, you put something up on GitHub, you’re essentially making a social contract with those people that you are going to be there for them, you’re going to maintain it, you’re going to answer questions, and essentially you have to treat it like it’s a full time job and it is your job, even if it’s nothing really your full time job. Right. 


Robert Cowart
And that resonated with me, and I feel like what really helped the original project that was on GitHub was essentially treating that community of users as if they were my paying customers, you know? Now, I probably failed in that in many regards, because I still had a full time job that I had to do. But at least the intention, and the intent, the attempt to do so was there. And I think that really helps in getting a loyal community around it, so that when we launch the new product, which, by the way, still has a community version, there’s still a free tier we always want to stick and not abandon the folks that really help make this thing possible. So still we have a community slack to interact with them. 


Robert Cowart
Launched community forums recently, and so I would just say that you’re never going to have the opportunity to treat your community 100% as you can a paying customer, but you should really thrive to do your best to get close to that with the resources you have. 


Brett
What about from a marketing perspective today, how would you summarize the marketing strategy and really just the go to market strategy. 


Robert Cowart
Gotcha. So again, we had that community to start from, which was great. We literally lived off of inbounds until we raised our initial seed funding, and that was enough to eventually get us to be a cash flow positive business, which was a really great position to be in. Actually, if youre going to raise money, its great if youre in that situation. So were really living mostly on inbounds at that point. We knew we could not do that forever. So we then raise some money to grow the team. We initially put some focus just on starting to produce a little bit more content and did our first few events with some of the partners that we had built up. Now, as we get here into 2024, we’ve made a much larger investment. 


Robert Cowart
And outbound sales team one side was fortunate enough to find a really strong sales leader who knows how to build that motion, and then also more recently a leader that was also has experience in our space, in the network observability space to start to build out our marketing team. Now, I will admit I knew almost nothing about like I’d worked trade shows before, but I otherwise knew almost nothing about the other activities and metrics and KPI’s of marketing. And so that was actually a place where RVC, our CVC venture guides in Boston, was really helpful because they have some folks on their team that are really experienced marketers and were able to help us with interviewing and kind of finding the right people. 


Robert Cowart
And so weve just started to do more things around paid advertisements, increasing a little bit more our social presence and other related activities that weve seen a big uptick actually over the last few months since that team started to get active. So I think we still have plenty to do and its an area that I would probably say we’re still, especially in the marketing side a little bit still in our infant phase. But I do feel really confident the team we put together, and it’s good to see some of the results of the initial activities. 


Brett
What have you learned about marketing team building? From my conversations and just work with a lot of founders, that’s something they really struggle with, is building out that marketing team. What have you learned from that process so far? And I know you’re still early in it, but what have you learned so far? 


Robert Cowart
You know, the first thing I would say about it is something I would say about anything when you’re building out a leadership function and any team you’re going to put together is as a startup. Right. I mean, you simply are not going to have time to hold everyone’s hand as we cross the street. Right. I actually think that’s a movie quote. And so you need to find people that have a plan. Like when you’re interviewing candidates, you really need to be 100% confident that on day one they have a game plan, they can come in and execute it, and that you have a strong belief and their willingness to do so. Hopefully they can demonstrate having done it before. 


Robert Cowart
And so that was something, especially in marketing we looked at, because while there are certain roles in the company, I can step in and fill that gap, or my Co-Founder could step in and fill that gap temporarily. This was not one of those roles. And so while I had some good initial conversations, were fortunate enough that someone I knew from my past actually recommended the lady we hired to run the marketing team. So that helped, of course. But even then, I had a good feeling when went through that initial process, but actually got someone, a real expert in marketing that RVC helped us with, to actually also interview those folks. 


Robert Cowart
But in general, id say the most important thing is have a plan, because they’re going to need to hit the ground running and really understand what they think their 1st 90 days are going to look like. In our case, that 1st 90 days was a big campaign around brand awareness. Our ultimate goal. Like I mentioned, we built some outbound sales motion, you know, have a team of sdrs that are, you know, reaching out to potential candidates. And as I explained it, I said, here’s what my goal is. If twelve months from now, one out of three, and I have no idea if that number, by the way, is a good ratio or not, but one out of three times that, our sdrs make a phone call and say, hey, this is Joe from Elastiflow. 


Robert Cowart
I want the person on the other side of the phone to go, oh, I know who yall are. I just read this blog, saw this video. I learned something about XYZ from some piece of content you created, right? And if we can have that kind of recognition, then our sales motion, our pipeline building, is going to be much more effective and efficient and the company’s going to be in a much better place, right? So that we really focused on what are they going to do in the first 90 days in our interview process. And yeah, went with the person who had the best plan.

Brett

When it comes to your market category, whats your thinking there? What is that market category that youre part of? And maybe a better way to ask that whats the line item here? 


Robert Cowart
So the solution that we sell and we collect various types of network data, but the primary thing, what our real focus is on network traffic. If you look historically in the area was network management. You’d hear it called back in the day. Nowadays we use the term observability a lot more. It tended to focus around network devices. There’s actually a parallel to what we do in the server application side and that 20 years ago everyone was focused on managing the server itself. Nowadays the focus is really on managing applications and the interactions that pieces of a distributed application have with each other. 


Robert Cowart
And very much the same way we’ve seen a shift in the networking market where there’s less of a focus on measuring the health or availability of the individual devices and much greater interest in understanding the traffic that is traversing the network. Because quite frankly, when you know who’s talking to who and the various attributes of those conversations, there’s a lot of interesting use cases you can pull out of that you can understand when the network is performing well or not. You know, someone says hey, today the network is slow and the it guys go well, yeah, that’s because we’re using 100% of our pipe to the Internet. And the next obvious question is well, what’s driving that utilization? That’s something that network traffic records can help answer. 


Robert Cowart
And what we kind of add on to that and where we believe demand from the market is coming in this area of observability is providing context. So if I was to tell you there’s a PC that’s talking to a particular server with a few aspects of a protocol that it’s using to do so, I mean that may be an interesting piece of information, but if I can put some additional context around that and tell you something like someone in your warehouse is trying to access your accounting system, now the whole meaning of that conversation across your network just changed because you might have a possible security incident going on. There could be a number of different situations you now need to pay attention to that you would not otherwise have any knowledge of. So the space we play is one, the network observability. 


Robert Cowart
But again, as that example I gave points out, if you know who’s talking to who and in the various attributes or context of those conversations, then there’s a lot you can do in the security space as well. There’s kind of two subdivid, I’ll call them kind of sub trends that we see in that space. One is there’s continues to be this migration to the cloud. I know it feels like we’ve been talking about that for the last ten to 15 years, but there’s still lots of organizations that are making their initial migrations to the cloud. And so when you’re in the cloud, there are no network devices, at least not any that you have access to. 


Robert Cowart
So there’s other methods in the cloud providers and network traffic records from the cloud providers, which we also support, is the way that you can then again have visibility into the network traffic and derive that value that I was mentioning earlier. But the other thing we see is this coming together of network teams and security teams and a lot of organizations. Those have historically been very siloed teams, and we are increasingly seeing across our user and our customer base where they are bringing those organizations together for the first time. And the way we’ve built our solution, the way we leverage that underlying network traffic record data, we end up being a good pivot point for that because we support those multiple use cases. 


Robert Cowart
So this is truly a data set that both the network engineering team and the security team are going to care about and can get value from. So there’s a term that some will call it like Devsecops or Netsec ops, but something that we’ve started using more recently is this term, Snock. So there used to be network operations center was the NOC and the security operations center was the security team and what they were doing. And we see more organizations doing what we would call a SNOC, which is a security and network operations center. And because among, at least from our viewpoint, among our competitors, we have the best combination of use cases for both performance and availability and security. Because most do one or the other, we’re pretty well positioned for that segment as it continues to emerge. 


Brett
What about lessons learned from fundraising? As I mentioned there in the intro, 11 million to date. What’s that journey been like and what have you learned? 


Robert Cowart
Actually, 8 million to date, I would say so kind of interesting question because I’ll first say this. What I hear a lot of people say about fundraising was not our experience at all. And this kind of colors a bit. My advice to other founders that I have given and will continue to give, which is of course we’re in the software space, I certainly limit any opinions I have to that space because they might not be applicable in other domains. But once you get out and start winning a few customers and getting that first business closed, we always had VC’s that were reaching out for us. Now typically this is some associate level investor that just reaching out to say hello. But as we continued to grow the business fairly successfully, there was interest there from the various VC’s. 


Robert Cowart
And so were able to get into some pretty advanced conversations with a number of investors fairly quickly. In fact, I don’t even know if I ever presented a pitch deck, like a proper pitch deck more than once or twice. That’s probably the other thing I would say, don’t invest too much time in the pitch deck. It’s just going to be about conversations. At some point they might want to see something. So getting back to like, the main thing I would learn is instead of spending a lot of time on pitch decks and trying to chase down lists of investors and writing 900 emails or whatever, spend more of that time building your product and winning your first customers. 


Robert Cowart
Because nothing is going to impress investors more or get you to that next conversation than having paying customers that validate that what you built provides value and that there are organizations out there willing to give you money for it. 


Brett
Final question for you, let’s zoom out three to five years into the future. What’s the big picture vision here? 


Robert Cowart
Well, we are certainly going to stay in our space and network. Actually, before I agreed to join your podcast, I listened to a number of episodes and now I’ve totally forgot the name of the gentleman that you spoke with. It was like a AI company in the UK and he had spoke about this really resonated with me. He said they needed to identify a segment of the market that was large enough to matter but small enough to dominate. Clay. And like I said, that really resonated with me. And when I looked at our space and the success we had to date, I realized we’ve kind of have that kind of niche and that there’s a lot of organizations out there that are chasing what I say I call applications and servers, some of which have grown really large and been really successful. 


Robert Cowart
However, all of those applications and servers are attached to a network and that is a slice of this market that we see is largely ignored. So over the next three to five years, we’re going to continue to build out our position in network observability. Network traffic records will stay at our core, but there’s a lot we’re going to be working on to one side, provide even more context around the traffic crossing our customers networks. We just released a product called Net intel. It is like a companion to what had been our core product, which we recently rebranded Netobserve. And its whole thing is around adding more context to the network traffic records. But theres other things were going to do as well to ensure as the world continues to move toward what we will believe to be an increasingly hybrid it environment. 


Robert Cowart
That means using cloud resources for some things, using on prem resources for other things, likely using though like cloud like interfaces to orchestrate and otherwise interact with those environments. As that move happens, we’re going to continue to build out our capabilities to ensure that the same thing we do now for our customers in their on prem networks and to some degree in the cloud, that we have a story across all of those environments that really serves as that source of truth and insight for our customers. 


Brett
Amazing. Love the vision. I’ve really loved this conversation. We’re up on time, so we’ll have to wrap here before we do. For any founders that are listening in that want to follow along, where should they go? 


Robert Cowart
Well, our website is elastiflow.com and of course we do have active presence on LinkedIn as well as on X, although I still want to call that Twitter all the time. So, you know, I would say follow along on either of those. I’d say LinkedIn is definitely where a lot of our content and where I share more thoughts myself. So that’d be a good place to follow either the company, again, the page Elastiflow on LinkedIn or Robert Coward on LinkedIn. 


Brett
Amazing. Robert, thanks so much. Really appreciate it. This episode of Category Visionaries is brought to you by Front Lines Media, Silicon Valley’s leading podcast production studio. If you’re a B2B Founder looking for help launching and growing your own podcast, visit frontlines.io podcast. And for the latest episode, search for Academy category visioners on your podcast platform of choice. Thanks for listening and we’ll catch you on the next episode.