One of the best aspects of being part of the Evolv leadership team is the chance to work closely with our founders, Mike Ellenbogen and Anil Chitkara. In light of our recent listing on the NASDAQ exchange, I felt it’s an appropriate moment to sit down with Mike and Anil to get their perspective on their road to now.
Dana: Why did you start Evolv back in 2013?
Mike: Evolv is my third startup in the physical security space, so I am deeply aware of the challenges and technical limitations associated with preventing active shooter and terrorist attacks. Many of our original Evolv team members have been together now through three startups in physical security. The 2012 Sandy Hook school shooting and the 2013 Boston marathon bombing directly affected people close to both Anil and me. Those events really crystalized for us that the world needs a fundamentally better way to prevent these types of attacks, and that there was currently no good solution. We looked at the situation and said, “We know how to solve these types of problems and we’re in a unique position to make a real impact. If not us, then who?” It just felt like it was time to get the band back together.
Anil: It’s personal to me. I have been close — painfully close — to multiple terrorist events, and I decided that I needed to turn my energy to making the world safer from future attacks. My close friend and college roommate, Steve, was on the 101st floor of the North Tower on 9/11. He had just gotten married and had a son. Twelve years later I was on Arlington Street in Boston with my three young children waiting for my wife to cross the finish line of the 117th Boston Marathon. She finished, we drove home and found out that 45 minutes after we left that the first explosive device had detonated. A close friend wasn’t as lucky. He was seriously injured and still has shrapnel in his neck. So, as Mike said, we started Evolv to stop these types of senseless acts. We saw that they were happening in more types of locations in more towns and cities, harming more and more people. We just knew there had to be a better way to prevent them from happening.
Dana: What problem did you set out solve?
Mike: There are plenty of technologies that help minimize the response time after an event has already started: video analytics, gunshot detection, etc. But this after-the-fact type of solution doesn’t address the real problem. The world needs a way to prevent the bullets from flying in the first place. At the time, the only available solutions to try to stop attacks like the Boston Marathon bombing were old school metal detectors. This technology is over 100 years old and was never designed for today’s visitors and spectators, with all of the items we normally carry — like smart phones and tablets — or for today’s venues. If you’ve ever stood frustrated in a security line, cursing the slow security screening process, you understand the problem. We saw the need for a frictionless process that can identify threats without slowing down visitor flow, ideally without even breaking stride. Our goal was to help any venue, with or without government mandates, to create a safer environment for their visitors without negatively affecting the visitor experience.
Anil: Old metal detectors and manual security checks were widely used after 9/11. These approaches treated everyone as a threat, forcing them to stop, empty their pockets and bags, and submit to a search. We wanted to pivot the paradigm. The vast majority of people are not a threat – so why not let them pass through without ever stopping and only stop those few who need a closer look? Why can’t most people be screened as they walk through at the normal pace of life without ever stopping? That’s what we wanted to deliver.
Dana: Why did you think you were the right people to solve it?
Mike: This isn’t a problem space that you just decide to get into and a couple of weeks later you fully understand it. A couple of kids in a dorm room aren’t going to figure it out. The physics is very, very challenging. The math is hard. There are all kinds of subtle environmental issues that cause huge problems in the real world but don’t exist in the lab. We have a unique team of very talented people with the depth of experience to anticipate many problems and the context to cleverly solve new problems as they come up. We’ve also been able to leverage the latest advances in sensors and machine learning that hadn’t been available or applied to this problem space before. Our prior success in this market also gave us excellent access to capital from really smart, deeply connected? committed? investors. Even with all these advantages, we had to work the problem really hard from many angles for a long time, but in the end, we cracked it.
Anil: Building on what Mike said, we really benefited from having a multidisciplinary technical team with an intimate understanding of different venues and their operational requirements. I don’t know of any team that has collectively spent more time on the front lines, shoulder-to-shoulder with security professionals as they conduct screening operations. We knew the challenges the staff were struggling with, and we knew what they wanted and needed. There are just as many subtle process issues as subtle technical issues of the kind Mike mentioned. We combined our knowledge of all these issues with our background in user experience design to solve for both the visitors being screened and the security staff operating the system. We’ve got hundreds and hundreds of lessons from operating in so many different environments in the real world. And we continue to incorporate these experiences into our product and our team’s approach with our customers.
Dana: Why wasn’t it solved before?
Mike: Most companies in our space wait for a clear market with a well-defined specification, usually from some government agency, before they’ll consider committing the time, resources and capital to develop a new product. There is no established “firearm detection standard” or “IED detection standard” out there to start from. We recognized that there are thousands of venues, from schools to stadiums to workplaces, that want to a create an environment that’s safe from threats to the crowd. Most of them aren’t really worried about objects that might be considered threats in a prison environment, or even an aviation environment. They are primarily concerned about firearms in the US and other similar threats to the crowd outside the US. These concerns have been well-known and almost universal for decades, but the venue operators had broadly rejected security metal detectors because they are awful. There are few products in the world so universally hated as walk-through metal detectors. Users made a value judgement and decided it wasn’t worth creating a line that trails around the block in order to screen visitors for weapons. Most just went with the lowest common denominator – guards looking in visitors’ bags or perhaps a cursory hand-wanding. We were willing to create the detection system we thought people were really looking for, even though there wasn’t a specification available to reference. It was definitely the harder path, but we believe it was ultimately the right path.
Anil: I think we looked at the problem very differently than others in this space. We didn’t want to find metal, we wanted to find weapons. In fact, we wanted to ignore personal items such as cell phones, keys and belt buckles. Once you look at the problem from a different lens, you start to think about the technology direction differently. The hardest part of the problem was to build a robust, resilient system that was adaptable to operate in a multitude of environments with different types of visitors carrying a wide range of personal items. A family going to a theme park is carrying different personal items than a worker going into a warehouse, a kid going to school, or a couple going to the opera. We focused on understanding three key factors: stream of commerce coming through, the environmental factors at the site, and operational variables for different security approaches. We then built a system that would be robust, resilient and flexible to meet these varied situations.
Dana: What were the major challenges?
Mike: One of the biggest challenges was being able to identify threats without slowing down the visitor flow, ideally without even breaking stride and with people walking together, even side-by-side. This requires being able to isolate individuals and find threats almost instantly, which is hard when you also need detection to be both more precise and more accurate than existing systems. Another related problem was providing a welcoming, non-threatening visitor experience while also creating a clear visual deterrent. Having an industrial design that unsettles threat actors by conveying that there is some serious tech under the hood without making it scary to harmless visitors was a tricky balancing act. And finally, there is the simple fact that we had to raise tens of millions of dollars of capital to adequately fund R&D and production. We had to innovate at the edge of the possible in both bits and atoms, and that’s just harder and more expensive than developing other types of products. That said, there’s nothing I enjoy more than being together with a group of smart people tackling tough problems like these.
Anil: It’s hard. It’s just really, really hard. Because the system is detecting and preventing weapons from entering facilities, it needs to operate at extremely high-performance levels. It can’t be right just some of the time. Additionally, we look at the security system as a combination of technology plus people (security staff) and process. These elements all need to work hand-in-hand. And people are fallible and inconsistent. It’s insanely difficult to maintain the same level of vigilance for every person coming through over a two-hour shift. With lines forming, anxious visitors, under hot sun or in driving rain — it’s just hard. We used advanced technology to automate the mundane, repetitive tasks so the security staff can focus on the most important tasks that require human attention. They need to address those few people who may be a threat with focused attention and follow their prescribed protocols. It took lots of iteration to get that balance right.
Dana: So, is this what you’d call a deep tech problem?
Mike: This is absolutely a deep tech problem. First, you have to understand the physics and develop the sensors that enable the system to discriminate between innocuous everyday items and real security threats. Then you need to design a hardware and software architecture that can work consistently, anytime and anywhere, while screening up to 3,600 people per hour, or one person per second. That’s essentially as fast as people can stream through a set of double doors. And then once you have the data and can keep up with the flow, you need to process the information and make a decision while visitors are still within a stride of the threshold. This requires a combination of advanced embedded software and machine learning. Anyone with a titanium hip or knee will appreciate the system’s ability to ignore these implants and other everyday items while automatically detecting actual threats.
Anil: What Mike said. It’s deep tech that requires a cross functional, highly integrated approach. I don‘t even understand the math on our whiteboards or the signal chain through the system. But it works, and it works really well.
Dana: Where do we stand relative to accomplishing the Evolv mission?
Mike: When we started Evolv, we envisioned a world where people were safe in all the places we live, work and play. We’ve taken a big step toward that vision, but it feels like we’ve only just scratched the surface so far. There are plenty of venues that want to create a safer environment for their visitors, fans, employees, students and guests but are just now starting to learn that Evolv exists. We need to do more to get our story out there. We’re also thinking deeply about other ways to apply our core technology to prevent gun violence, active shooter and terrorist events in different types of applications and spaces. There are plenty more technology and business problems to solve on the road to fully realizing our vision. We know we have a long way to go, but we’ll get there eventually.
Anil: Evolv has taken a major step toward making the world a safer place. Many of our customers were not using any security screening technology before we deployed our products at their locations. We’ve kept thousands and thousands of weapons out of places where they aren’t welcome. We’re now screening over 11 million visitors a month, and that number continues to grow rapidly. But the fact is, there are still shootings and bombings. There are fatalities and injuries that can be prevented. We’ve got to accelerate and scale everything we do to match the scope of the problem. Our story has just begun.
