Subscribe
Share
Share
Embed
The Drone Network explores how drones are reshaping the world. Hosted by Bryce Bladon, the podcast documents the tech, economics and people piloting the world's largest standardized drone imagery network.
TDN213 - Ben Kovacs
===
Bryce Bladon: [00:00:00] Hello and welcome to The Drone Network, the only podcast in the air and on the airwaves. I'm your host, Bryce Bladon, and on this show we document the tech, economics and people piloting the world's largest standardized drone imagery network. Each episode we explore how drones are reshaping industries, creating new economic opportunities, and literally changing how we the world.
On today's episode, we explore how spatial imagery from drones is actually used and consumed, and how it differs from satellite imagery.
Today's guest is Ben Kovacs, senior product marketing Lead at Spexi Geospatial. Ben, thank you for joining me today.
Ben Kovacs: Thanks, Bryce.
[00:00:38] Ben's Background: From Space Tech to Spexi
---
Bryce Bladon: All right, Ben, just to kick us off, why don't you tell me a little bit about your background and how you ended up at S Spexi?
Ben Kovacs: Sure. So I've spent the beginning of my career actually working within space Technology and really starting as the industry starting to focus on developing for different commercial segments.
The space industry, generally speaking, has had quite [00:01:00] a government and institutional history, and at the time I joined, uh, there were a number of really key technological developments that actually allowed it to be much more commercial friendly. I came in at a really interesting time through that experience, started to understand where.
A number of different commercial applications were being developed for customers on earth using space data. And through that transition, started to see a number of things change within the landscape of aerial imagery specific to drone. And when I learned about Specia, it really changed my perspective on what could be possible with aerial based technology and how it could operate action in a very similar way to what I had seen and experienced with space-based technology.
And it was just a really exciting time too. Join a drone technology company and here we are today.
Bryce Bladon: Absolutely. So Sky Watch, one of your former employers started by commercializing space technology. What was it like working with something as relatively abstract as satellite data access and, and how has that shaped about these emerging tech [00:02:00] categories we've sort of brought up, whether that's drones or satellites or anything in between.
[00:02:03] Selling the Abstract: Satellite Data and the Expectation Gap
---
Ben Kovacs: I think abstract is a really good word to describe satellite technology, especially to many businesses that have physical operations on earth. The operations of satellites and other technology platforms and space is very abstract. We can't physically see or touch them once they're in operation. So a lot of it is like a mystery.
It's kind of behind a curtain. We don't exactly understand everything, except when we're working with the companies that are operating all of these assets in space. In addition to that, all of the data and all of the information that is collected by all of these systems when it's delivered back to earth, we don't always have a sense of how it might be able to bring value to our businesses until we're actually touching it and working with it within our computer systems.
And so one of the initial, um, interesting challenges that we actually faced, um, in the early days was even just working with. Setting and maintaining expectations with customers. Many customers had come to us using things like [00:03:00] Google Maps, Google Satellite, and they were exposed to really rich, really high detailed imagery of the earth.
And there was this expectation that if I use a digital platform like Sky Watch or any other platforms at the time, I might be able to get that image delivered to me within minutes, within hours, maybe within a day, et cetera. And at the time, most of the satellite imaging platforms. We're actually delivering data products that were still great, but but not nearly as good as maybe what you would be otherwise exposed to.
And so there was this big disconnect and this big gap between customer expectations and, and the reality of space infrastructure. And so there were a number of. Different applications that we were focused on building with customers. And you would think that maybe there'd be a long library of things you could do, but in reality it was actually only a handful of real commercial applications that were actually successful.
So big expectations management and, and certainly a lot of good lessons learned there in terms of. Ensuring you can work with customers to get what they need, but also delivering on promises that you can [00:04:00] hold up.
[00:04:00] Are Drones and Satellites Converging?
---
Bryce Bladon: There is an interesting through line between that abstractness you've mentioned, and even just space data and drone data, both are capturing the earth from above, just at very different altitudes and in very different economic scenarios.
Do you see these worlds converging at all? You mentioned how, uh, there was an expectation from customers around certain data and how they had gotten it before. Um, I'd be curious if that informed their expectations, whether it's around satellite imagery or drone imagery or, or where there's any overlap between those two.
Ben Kovacs: Mm-hmm. Yeah, the worlds are definitely converging. I mean, satellite technology has certainly continued to improve in a commercial sense, just as drone technology has continued to improve. But beyond just the technological improvements, there's been a number of operational. Changes too that have actually enabled a bit of a different commercial landscape and different business models that are evolving.
Where I certainly see them converging is in relation to how customers today and moving forward will have more opportunities to benefit from the flexibility like the. [00:05:00] The flexible nature of working with these systems, where in the past, getting access to data was, was really like supply constrained. If you needed an image, if you needed data tomorrow, it was very difficult to access.
But moving forward there will be so many more opportunities and it's, and it's because of all of these changes in systems and. Suddenly what happens is the shift that takes place, really the narrative is instead of having to rely on getting data based on operator timelines, so those operating all of these satellite aerial systems, et cetera, you are in more control of getting access to the data where and when you need it.
That's a really big shift.
Bryce Bladon: What are the systemic or operational changes that that sort of enabled this, this convergence or the shift as you've framed it?
Ben Kovacs: I'll start first with the satellite side. Certainly with the change and growth of the commercial satellite industry, more technology, more hardware and space.
Cheaper access to space. It just enables platforms and more platforms to [00:06:00] go to space that are more orientated towards commercial markets in terms of their operations and their capabilities and what they can deliver. So that certainly helps. And then from the aerial and the drone side, just what we're seeing in terms of technical innovation and to speak to Spexi directly, the ability to operate a.
Large drone pilot network with so many nodes in the network to be able to collect data when and where needed. Those systems actually kind of operate in a very similar fashion. So historically, drone and aerial systems haven't always been associated with the level of scale as satellite systems, but under this.
New operational model that Spexi is certainly bringing to the market. Actually, they're, they're very similar.
Bryce Bladon: Alright, so help me, help me understand the different types of, of data out there. There's a lot of spatial data for mapping the real world. There's satellite, there's lidar, there's street level, there's drones, there's biplanes.
[00:06:53] What Makes Drone Data Different (and Better — and Worse)
---
Bryce Bladon: What, what does drone data do differently? And, and let's be candid, what does it do better? And, and maybe even, what does it do worse? [00:07:00] That's what I'm almost most interested in.
Ben Kovacs: So certainly drone will provide a level of detail that is fairly unmatched as it relates to other platforms. And this is just truly due to its altitude that it's able to operate at.
So in many cases, we're talking about an operating uh, ceiling, kind of between 40 to 80 meters in elevation above the ground, compared to say, hundreds of kilometers away. When we're talking about satellites or. Planes, right where we're talking about thousands of feet. So you're much closer, closer to your subject, you get better image resolution, and that just means that you're gonna get access to better data, which is great for customers.
Historically, one of the challenges of this has just been the fact that a drone can only collect imagery and data, uh, within a smaller footprint, which is highly relevant to its immediate site or project that it's being used, but very difficult to suddenly cover a much larger area. This is of course where platforms like satellite and fixed wing planes have been able to provide large coverage.
And of [00:08:00] course, this is where we are starting to see changes now that one can rely on a distributed pilot network to give them that coverage. They get that high resolution, and then they also get the coverage that they ultimately need. To monitor their operations across a state level, country level, whatever their needs might be.
Bryce Bladon: Well said. It really does seem like it's a, it was traditionally, let's say, a balance of scale versus detail, and also maybe a degree of speed. The ability to get a satellite to do something is probably relative to how centralized your control over that satellite is. I need to. Image something from a satellite.
I imagine it'll take me a lot longer to get some sort of imagery than maybe someone who directly controls it or, or has some sort of preference to it, and that comes with caveats. But I know a lot less about satellites than you do, so I'm gonna stop talking about them.
[00:08:47] Explaining Data Freshness Without the Jargon
---
Bryce Bladon: But what I do want you to do is sort of explain these data products to me without getting too deep in the weeds.
Like the, the number one thing I found with drone data is explaining the concept of data freshness, and that is something that [00:09:00] when I try to explain it to my mom, I see her eyes glazing over. How do you explain this stuff? How do you even arguably sell this stuff without getting too deep in the weeds?
Ben Kovacs: Certainly when it comes to imagery itself and what customers and consumers might think of imagery today, and to use the the Google Maps or Google satellite view example. Being able to see imagery as is like we would see the world with the naked eye. Um, that's certainly the default or like defacto image product that is certainly created by many of these systems compared to, say, other systems like lidar, which of course uses light detection and ranging technology.
So you can do different layers of measurements, et cetera. I mean, each of these systems are going to operate in and function slightly differently in terms of when and where and how the data is collected. And so when we talk about things like data freshness or, um, when an image is collected, like we're really just talking about when and.
How frequent the [00:10:00] data might be collected. And we know that the world is always changing, and many of these systems are just designed to capture, collect, and record that change in a way that can be used for decision making by teams and by digital systems at a later point in time. So the freshness piece.
What is certainly important there is to customers who need to record that change. If change is happening frequently in a particular location. Having access to systems that can capture that change on a regular basis, on a regular cadence that is relevant to their operations is incredibly important so that that change can be measured and monitored, and they can then use that information to make better decisions within their business.
[00:10:40] The Temporal Layer: Maps, Change, and Prediction
---
Bryce Bladon: I've always thought about it in terms of how maps have evolved. We originally just had one horizon, then we had latitude and longitude. Then at a certain point we started to see, well first of all, we mapped the entire world and that's incredible. Um, but then there was a sort of historical context layer to our maps, a temporal layer as it were, and you could [00:11:00] see how things used to be.
And the thing about data freshness is the more of those temporal layers you have. The more upgraded that map is, and you start to get to a place where you can not only see minute changes over time, you know, cracks appearing in roads as it were, um, but ideally with historical context and accurate spatial data, you can start to predict changes.
You can start to see things before they happen. And that's one of the, the catch 20 twos of, of good spatial data is once you have a bit, you start to realize the best version would've been starting this 18 months ago or, or 18 years ago, ideally. And, and let's maybe even talk about that for a second. What does good spatial data look like?
Like what makes this imagery valuable versus worthless?
Ben Kovacs: That's certainly an interesting question because historically within imaging as an industry, old or stale imagery, imagery that might've been collected months or years ago in some cases may not be valuable to some customers. But in many cases it could be very [00:12:00] valuable to others, and it really depends on the application at hand.
And so I think this question gets answered based on whatever the needs are for the customers, and those needs are quite diverse and being able to provide flexible options for. Customers to meet a variety of needs, I believe is the way in which all of these imaging platforms can continue to remain relevant.
So whether they have a large library of historical data to use, or their operating capabilities are able to provide frequent updated data for what customers need, those will give them the opportunities to continue to support all of these. Particular applications and use cases that, that customers truly need.
[00:12:39] Standardization: Why It's the Key to Drone Data at Scale
---
Bryce Bladon: That sounds like it's the ubiquity or, or the standardization of the data. The ability for somebody to take this data and basically implement it into their system or workflow and, and have a complete operational view versus having to whip up something bespoke is, is that accurate?
Ben Kovacs: Yeah. Let's even just chat on the standardization piece for a [00:13:00] second too.
Standardization as, as a concept, has certainly been implemented across many different imaging platforms already. So, so satellite, for example, all satellites, imaging satellites, um, that, that are using imaging technology, that we would be able to take a look at an image and say, Hey, I can see a, a, a tree in a road, right?
They're, they're gonna operate in what is called a sun synchronous orbit. So they're always gonna be passing the earth at like a regular cadence. And, and, and that is actually just a standardized operating procedure. When it comes to other systems and, and how they might operate and they have their own operating playbook you could call it, and the ability to standardize data for each of these platforms has been important for ensuring consistent operations.
When we think about drone imaging and what standardization looks like for drone imaging, like the next step in the next um, level for this is being able to standardize what has otherwise been like a very challenging imaging system to standardize because of all the different variations and variables.
In order to [00:14:00] provide not just, uh, fresh data, but also usable data that can be used for digital systems in an automated fashion, or it can be used to take a look at historical data and do good time series, change detection, comparisons, et cetera. Standardization is a huge important part of that.
Bryce Bladon: I had Dylan Gorman on recently, and one thing he brought up is probably one of the.
Biggest mistakes drone pilots make is when they sell their drone services, they build themselves as drone operators versus positioning themselves as a solution through the use of their drone, um, that they are engineering for the client. So to put this another way, instead of selling yourself as a drone operator to.
A real estate agency, you sell yourself as a real estate photographer who happens to use drones to do it efficiently or in a way that would matter to that agency. What I'm building up to is, as we've noted, the concept of spatial data, uh, can be a little abstract.
[00:14:54] Who's Underserved by Spatial Data? (Cities & Utilities)
---
Bryce Bladon: So who's currently underserved by spatial data?
Like what's an [00:15:00] industry or a use case? That's flying under the radar that if they knew about spatial data, if they implemented it into their systems, it would be transformative or catalyze something really interesting.
Ben Kovacs: Yeah. This is interesting because so much of this is just based on access, getting access to usable data, and I think the first example that comes to my mind is even just.
Operators within our own cities. I mean, cities are such complex spaces and oftentimes the organizations that are keeping the lights on and all of our systems functioning and all of the benefits that we experience living day-to-day within a city. I mean, these, these organizations in these teams could benefit from many applications with getting access to the spatial data.
And in many cases, they already have access to sensors and other spatial data, uh, formats within their. Operations already. Um, but getting access to, to, to new data, for example, um, highly relevant, high quality, high resolution imagery data is certainly [00:16:00] another opportunity to help improve their operations and to make their jobs easier.
So, so that's certainly one category that stands out. I'll name another one as well, just as another layer on top of that, and that would be the different utility operators that operate within cities. I mean, utilities are just so complex to service thousands and, and, and millions of households within cities.
These organizations are already using lots of spatial data today. But again, the access to highly relevant, high quality, high resolution imagery data is just the next step that will help make their operations more efficient. And that's certainly something that, uh, I'm personally excited to, to see evolve and I know.
Specia is as well.
Bryce Bladon: Interesting. So we've talked a lot about like the, the ambiguous nature of some of these, these products and, and rather how like the edges are changing or operations or systems are shifting. Um, and so I'm curious, you, you do product marketing and, and the product you are working on functionally is still actively being shaped.
[00:16:57] Marketing a Product That's Still Being Built
---
Bryce Bladon: What is the hardest part of doing that? How do [00:17:00] you market a product that is in some ways in flux?
Ben Kovacs: There are certainly a number of interesting themes. To a product and to the market right now that are continuing to evolve. Historically, companies providing imagery products were, were really, truly seen as imagery providers moving forward with the rise in better spatial resolution, better operating coverage, et cetera.
The ability for imagery providers to. Collect and maintain. A real up-to-date representation of the real world is now opening doors for a number of automated physical AI applications. And the positioning of these companies that are providing this data actually shifts from being not just an imagery provider, but effectively an essential data layer.
To many of these systems and to become a data layer for the physical world, I believe is the hallmark positioning. And this is all happening rapidly. And to be able to [00:18:00] keep up to the developments, not just with the technology, but also what customers are expecting is certainly a interesting challenge, but, but a fascinating one to be involved in.
[00:18:10] Hype vs. Real Use Cases — How to Tell the Difference
---
Bryce Bladon: There's a lot of, or rather there has been a lot of noise and, and some hype around drones and, and even drone data. And having seen Spexi navigate a, a very interesting position where basically the product introduced, uh, it moved to base main a, a layer two blockchain. Um, and that's in, uh, basically the crypto industry, which is.
An industry that can be utterly fraught with hype and all these, all the speculation and, and all these other things. At the time of recording, uh, this move to main net was, uh, over a month ago. And as far as I can tell, everything's going great. What I'm building up to is the question of not necessarily Spexi, but whether it's other drone operators, other drone data suppliers, or spatial data even.
Like where's the hype and, and like. What, what separates a real [00:19:00] use case from that hype?
Ben Kovacs: Hmm. Certainly lots of moving parts right now within the industry. What, what ultimately separates the hype from a real use case, of course, will be a paying customer. Hmm. And that is such an anchoring theme, and I believe.
We will see many prototypes and many pilots for many emerging use cases continue to develop at a really rapid pace. And I believe many of them actually have legs for commercialization, and it might just take a little bit longer for those to come to fruition. But if someone was to look and to try to detect hype from, from a real use case, um, definitely look to see which customers are involved and look for really clear problem statements and really clear solution statements and just see if those line up and, and I think what we'll see.
Is, uh, a number of rapid developments of new use cases that will come up and they'll meet that requirement and it'll be really interesting to see them
Bryce Bladon: evolve and customers is a real good answer. I think that's also part of why, uh, what I talked about with spec's move to main net actually worked. Is it functionally [00:20:00] improved?
The payment experience for pilots versus just being a move to blockchain for X, Y, or Z for venture capital reasons, even whereas real customers, and in this case real suppliers too, uh, are, are real good canaries in the coal mine as it were. Um, I wanna talk about, uh, what's coming up, um, whether it's.
[00:20:17] What's Next: Parametric Tasking and Machine-Requested Data
---
Bryce Bladon: Within the Spexi product space or within the drone industry as a whole, um, what's on the horizon that's got you excited? Whether it's a product feature or whether it's something in the drone industry as a whole? Like is there anything coming up that you are pumped about that you think will be transformative or interesting?
Ben Kovacs: Yes, certainly, uh, levels and layers of automation as it relates to using these systems. Today to request really high quality aerial data. In many cases, it is quite a manual process and there is certainly some automation that's being introduced. But I'm really excited for a scenario where this drone network can be tasked not just by humans, but by other digital systems.
And this is a [00:21:00] term referred to as parametric tasking or parametric ordering. And you can imagine a world where. Sensors embedded in the real physical environment when certain thresholds are reached. For example, if a river, uh, stream discharge sensor reaches a certain level for flow rate, it might indicate that a flood might be coming.
And so when that sensor reaches that level, if it has connection to. To the internet, it can send a task and request to an imaging network to preemptively prepare that network to collect imagery where and when it's relevant, either before, during, or after a flood event. And the ability to have a system requests data parametrically rather than a human in the loop might be the difference between being able to take action for emergency response or other.
Timely critical related applications. And that is something I'm really excited to see roll [00:22:00] out. And it's not far away from where we are today.
Bryce Bladon: Interesting. That's very interesting.
[00:22:04] What Feels Different About Geospatial Right Now
---
Bryce Bladon: It, it does feel like geospatial has been about to take off for years. So with someone with as much experience as you, like, what does feel different about this space right now than three, four, or five years ago?
Like how have things changed?
Ben Kovacs: The feeling of change feels. Like this, geospatial technology is becoming more and more critical to real world operations. Not to say it wasn't before it, it always has been. Geospatial has, has always been relevant. Um, but, but just the, the level of criticality and, uh, the integrations and the interconnectedness between.
Real world operations and the technology, I think has just continued to increase. And as more systems are continuing to operate in a more, in a more like remote first basis, our reliance on this technology has just continued to increase. So that's, to me, I think what, what really just stands up the most, and I think we're seeing that in the conversations we're having [00:23:00] with customers on a day-to-day basis.
And from those that are excited about finding ways to truly integrate and rely on this technology on a day-to-day basis.
Bryce Bladon: Excellent. Alright, two more questions for you, Ben. First one, who in this space, in the industry as a whole, do you think is doing a good job of communicating the value of spatial data?
This. It could be you. This could be Spexi, this can be X, Y, Z. But I'm curious because it's probably one of the, one of the biggest open questions I've had since diving into this space myself.
Ben Kovacs: That is a good question. Generally speaking, I would say any organization that's able to visually represent their data products in action, I think is.
Definitely doing the right thing to communicate the value. And, and I say this is something we certainly strive to do at Spexi. Uh, there are certainly some other organizations, I, I would actually say in the utility space. Um, so like utility innovations specific to electrical utilities, power lines, and vegetation interference that have done some fantastic [00:24:00] work to provide highly relevant.
Product demos to really showcase and communicate the value of their technology. And I would certainly advise people that are looking for inspiration to actually look to that industry for a few of those um, organizations.
Bryce Bladon: All right. Good answer. Did not expect it to be quite so nuanced or interesting. I was gonna say Niantic spatial.
If you didn't have anything, they have a video I sent people. When I try to explain what spatial data can do, which I think really backs up your point of, as long as you can show the product somehow remove the abstraction as it were, I do think you're miles ahead of. Most of the competition. But this brings me to my last and arguably most important question.
[00:24:35] Drone or Don't?
---
Bryce Bladon: Ben, do you have two minutes to join me for our hit segment drone, or don't?
Ben Kovacs: Absolutely. Let's do it.
Bryce Bladon: Fantastic. So how this works is pretty simple. I give you three quote unquote drone facts. You tell me which two are true and which one is false, or if you wanna play it real fast, just tell me the lie. Uh, alright. So you ready?
Ben Kovacs: Yep. Let's do it.
Bryce Bladon: All right. Drone fact number one. In 2023, a drone [00:25:00] operated by the Canyon Red Cross completed over 1000 blood delivery missions to remote clinics cutting average delivery time from several hours by road to under 40 minutes.
By air drone fact number two, the FAA now processes more drone registrations annually than manned aircraft registrations a milestone first crossed in 2017 and one that has widened every year since. Drone fact number three. DJI holds more drone related patents than any other company in the world, including Boeing, Lockheed Martin and Northrop Grumman combined.
Which two are true and which one is the lie?
Ben Kovacs: I believe the first two are true, and the third is a lie.
Bryce Bladon: That is impressive. You are correct. So. The combined claim, I think is what gives this one away. Those companies hold enormous patent portfolios, uh, and, uh, b's. Uh, patents are numerous, but my God, uh, Boeing and Lockheed Martin alone have.[00:26:00]
Real competition in that regard. Ben, that was really good. I, I put a lot of spins on all of these facts. I've, I've had Graham do this a few times and he always blows it outta the water. So I tried to, uh, turn it up to 10 and you crushed it. Thanks. Absolutely. And thank you, Ben. I really appreciate your time and, uh, for joining me on this show.
Uh, is there anything you'd like to direct people? Towards, uh, at the end of this call.
Ben Kovacs: Yeah, absolutely. I mean, check out Spexi, take a look at what Spexi ISS doing, and of course Lair Drone as well continue to stay involved in ongoing innovation within the space. It's an exciting time.
Bryce Bladon: Alright, thank you again Ben, and thank you all for listening.
[00:26:35] Thanks for listening!
---
Bryce Bladon: Thanks for being a part of the Drone Network. Subscribe wherever fine podcasts are served. To get a new episode every week and remember to leave us a five star review on your podcast app of choice, it helps a lot. Today's show was sponsored by Spexi Geospatial and LayerDrone. Learn more about standardized drone imagery built for global scale at Spexi.com.
That'ss SPEXI.com and Layerdrone.org. Thanks again for [00:27:00] listening.