In 2016, Michael Mazur and Adam Wiśniewski helped author one of the drone industry’s most widely cited forecasts, putting the addressable market for drone-powered solutions at $127.3 billion. A decade later, the AI Clearing co-founders revisit that prediction, examining where the industry developed as expected, where it surprised them, and why the next wave of value may have less to do with the aircraft than with the data they collect.
In this guest post, the authors offer a rare retrospective on a forecast that helped shape expectations for the commercial drone industry, and consider what the lessons of the last decade suggest about the next one. DRONELIFE does not make or accept payment for guest posts.
We Wrote the $127 Billion Drone Forecast Ten Years Ago. Here’s the Scorecard.
By Michael Mazur and Adam Wiśniewski
In May 2016, our team at PwC published Clarity from Above. It was the first serious attempt to put a number on what commercial drones were actually worth across industries, and the number we landed on — $127.3 billion in addressable market value — traveled further than anything either of us has written before or since. It became the most widely quoted research the firm had produced at the time, picked up more than 70,000 times in global media.
That reception told us something we did not fully appreciate at the time. The market was hungry for a number to anchor on, and it was going to use ours whether or not it read the footnotes.
Ten years is long enough to grade the work. Here is our own report card.
What we got right
The biggest call in the report was that infrastructure would be the largest commercial application of drone technology. We put it at $45.2 billion, ahead of agriculture at $32.4 billion and transport at $13 billion.
That one held. According to the FAA’s own survey of Part 107 operators, construction and real estate are now the most common commercial drone activity in the United States, with utilities and energy production close behind. Analysts tracking the commercial segment consistently report infrastructure as the largest end-user category by revenue. And the specific applications we flagged in 2016 — investment monitoring, maintenance, asset inventory — are now routine on large capital projects rather than pilot programs.
We believed that call was enough to act on it. Both of us left the firm to build a company in exactly that segment, applying computer vision to progress monitoring on large infrastructure builds. That was not a hedge on the forecast. It was the most direct bet we knew how to make.
The second thing we got right was the language. We deliberately did not title the report “the drone market.” We called the category drone-powered solutions, and we argued in the opening pages that this was a story about applications rather than machines.
That distinction looked pedantic at the time. It does not now. Services account for roughly 78% of the civil drone market today; hardware accounts for something closer to 17%. The aircraft became the smaller half of the business, and it happened faster than most hardware companies expected.
What we got wrong
It is worth remembering where the market actually was when we published. Commercial drone activity in 2016 was worth somewhere in the neighborhood of $1 billion globally. Almost nobody we briefed believed the category could move as quickly as we said it would.
It then grew roughly fortyfold in a decade, which is a rate of expansion very few technology categories ever achieve. So the forecast was not wrong about direction, and we would argue it was not wrong about magnitude either. It was wrong about time.
The $127.3 billion figure was an addressable market number — our estimate of the value of existing labor and services that drone-powered solutions could displace. It was never a revenue forecast, and we said so in the report. But a number that large travels without its footnotes, and it set an expectation the industry did not meet on anything like the implied timeline. Credible estimates now put the commercial drone market around $40 billion. The ceiling we identified still looks roughly right. The path to it was far longer than the report suggested, and we did not do enough to say so at the time.
The miss we would flag hardest is transport. We valued drone applications in transport — parcel delivery, medical logistics — at $13 billion. In 2026, the entire drone delivery market is somewhere between $1 billion and $1.5 billion. Zipline passed two million cumulative commercial deliveries in January, a genuine engineering achievement and a useful reminder of the distance between technical success and market scale. Delivery was the application that captured the public imagination. It has been the slowest to matter commercially.
Agriculture was our second-largest call at $32.4 billion. It has grown. It has not grown to that.
What we did not see coming
Two things, and the first is geopolitics.
In 2016, when we wrote about regulation, we meant airspace. We highlighted that Poland had established the world’s first complete legal framework for commercial drone operations in 2013, and we argued that regulatory clarity was the precondition for commercial adoption. That was correct as far as it went. What we did not model was that the hardware supply chain would itself become a policy question. The FCC’s December 2025 order restricting new foreign-manufactured UAS — arriving after years of debate about a single manufacturer’s roughly 70% to 80% global share — is a category of risk that simply was not in our framework.
The second thing we underestimated was what would do the actual work.
We wrote about “data analytics” as the mechanism that turned drone imagery into value. We were directionally right and technically vague. The unlock was not analytics in the 2016 sense of the word. It was deep learning — convolutional and later transformer-based models that can look at an image of a construction site and tell you what is in it, at a scale and consistency no human review process can match. More than half of new drone deployments now involve AI or autonomy in some form. That is where the margin lives, and in 2016 we could only describe it in outline.
What the next ten years look like
Here is what we find most interesting about the decade just finished. The industry spent it arguing about the aircraft. Airframe specifications, flight endurance, BVLOS waivers, Remote ID, and now supply chain sovereignty. Those arguments weren’t unimportant — several were existential for particular companies. But while they were happening, the commercial value quietly migrated to the layer above.
In most industrial applications, the drone is now a sensor delivery mechanism, and an increasingly commoditized one. That is not a knock on the hardware companies. It is what happens to every hardware category that matures. The defensible value moved to what happens to the data after it lands: the reconstruction, the classification, the reconciliation against what was supposed to be there, and the decision that comes out the other end.
What that means for the next decade is a much larger population of companies than most people currently expect. Construction progress monitoring was not a meaningful category in 2016, and now it supports real businesses with real customers. The same thing is going to happen in utility inspection, mining, forestry, insurance adjustment, port operations, precision agriculture, and a dozen segments we would not think to list. Each will produce its own set of companies whose actual product is interpretation rather than imagery. We think there will eventually be thousands of them, and the large majority have not been founded yet.
For drone service providers operating today, that migration is the strategic question of the next three years. A business built on fly, capture, deliver is competing on price against a cost base that keeps shrinking. A business built on what the data means is competing on something considerably harder to replicate.
If we were writing Clarity from Above today, we would spend far less of it on the aircraft and far more on the pipeline behind it. The total market number would probably land in a similar place. The distribution of who captures it would look very different.
Michael Mazur and Adam Wiśniewski are the co-founders of AI Clearing, an AI-powered construction progress monitoring platform deployed on major infrastructure projects globally, including roughly one-third of U.S. utility-scale solar construction. Mazur, the CEO, was previously a partner at PwC and global head of its Drone Powered Solutions practice. Wiśniewski, the CTO, was a director in the same practice. They co-authored the 2016 report Clarity from Above.
Miriam McNabb is the Editor-in-Chief of DRONELIFE and CEO of JobForDrones, a professional drone services marketplace, and a fascinated observer of the emerging drone industry and the regulatory environment for drones. Miriam has penned over 3,000 articles focused on the commercial drone space and is an international speaker and recognized figure in the industry. Miriam has a degree from the University of Chicago and over 20 years of experience in high tech sales and marketing for new technologies.
For drone industry consulting or writing, Email Miriam.
TWITTER:@spaldingbarker
Subscribe to DroneLife here.
