For most of drone delivery's short commercial history, the hard problem has not been getting a package from a store to a porch. It has been getting two competing operators to fly over the same suburb at the same time without anyone phoning an air traffic controller. On June 25, 2026, Flytrex offered the clearest data point yet that the problem is solvable at scale: in the Dallas-Fort Worth metroplex, its drones and Wing's drones — Wing being the Alphabet-owned delivery unit — have grown from a handful of overlapping flights to thousands per month in just under a year, with zero airspace conflicts and what the companies describe as a 100% strategic deconfliction rate.
That last phrase is the whole story. Two rival commercial operators, flying low over the same residential neighborhoods, coordinated their flight paths automatically, machine-to-machine, with no direct air traffic control involvement and no humans manually clearing each aircraft. It is the model the FAA has spent years building toward, and it is the model that any serious nationwide scale-up of drone delivery will have to lean on.
What the numbers actually say
The headline figures come from a structured evaluation Flytrex and Wing ran across January and February 2026 in DFW. During that window the two operators flew roughly 8,000 deliveries combined. They ran simultaneous operations on 30 of the 31 active days, with more than 10 hours of overlapping flight time each day — not a one-off demonstration, but a sustained, daily, dawn-to-dusk shared operation.
Volume climbed steeply over the period: combined operations grew 215% month-over-month. Across all of it, the companies report zero airspace conflicts and a 100% deconfliction rate. Shai Karassikov — a Flytrex product manager and co-chair of the U.S. UTM technical committee — described the operators as "scaling ... to thousands per month," an arc that, by the companies' account, has unfolded in just under a year. Separately, Flytrex says it has now passed 200,000 total U.S. deliveries since it began operating, a cumulative milestone that puts the DFW shared-airspace work in the context of a maturing commercial operation rather than a pilot project.
The geography matters here too. The two operators were not flying in loosely adjacent regions of a sprawling metro; their operating zones in Little Elm and Wylie, Texas sat as close as 1.36 miles apart. At drone speeds and ranges, that is genuinely shared airspace, the kind of proximity where uncoordinated flight planning would eventually produce a near-miss.
How the deconfliction actually works
The coordination did not happen through radio calls or a shared control room. It happened through a software standard: ASTM F3548-21, the specification for USS interoperability, using what the standard calls the Strategic Coordination service.
A short explainer is in order, because the acronyms hide a fairly elegant idea. "USS" stands for UAS Service Supplier — a third-party (or operator-run) service that manages the airspace intent of drone flights. Under the FAA's UAS Traffic Management (UTM) framework, the FAA does not personally separate every drone below 400 feet the way a controller separates airliners. Instead, it approves a network of USS providers that exchange flight intent with one another and perform strategic deconfliction — that is, they resolve potential conflicts ahead of time, before the aircraft ever leave the ground, by sharing where each operator intends to fly and adjusting plans so the volumes do not collide.
Flytrex and Wing each run inside that network. ASTM F3548-21 is the common language that lets two different companies' systems talk to each other, share intended operating volumes, and negotiate separation without either side seeing the other's proprietary internals. Strategic Coordination is the specific service within that standard that handles the pre-flight conflict resolution. The result, in DFW, was that two competitors flying 1.36 miles apart could each plan and execute thousands of flights while a machine handshake kept their airspace volumes from overlapping.
Why this is a test case, not just a press release
The DFW operation sits directly underneath one of the FAA's most consequential current efforts: the UTM Operational Evaluation. This is the agency's program for proving, in real airspace with real traffic, that the network-of-USS-providers model works before it becomes the regulatory backbone for large-scale beyond-visual-line-of-sight (BVLOS) flight. As of January 2026, the Operational Evaluation counted 17 approved providers and operators.
The FAA's own UTM program describes the architecture plainly: a network of FAA-approved UAS Service Suppliers that exchange intent and provide strategic deconfliction below 400 feet without direct ATC control. That is precisely what Flytrex and Wing demonstrated in the field. The companies' data is, in effect, evidence submitted to the question the Operational Evaluation exists to answer — can independent commercial operators safely share dense low-altitude airspace using interoperable software rather than a centralized controller?
Q&A: the questions this raises
Is "zero conflicts" the same as "zero risk"? No, and it is worth being precise. A 100% strategic deconfliction rate means the pre-flight coordination layer resolved every potential conflict it was designed to catch, and that no airspace conflicts occurred across the evaluation. It is a strong safety signal over a meaningful sample — roughly 8,000 deliveries with 10-plus hours of daily overlap. It is not a claim that the system can never fail, and strategic deconfliction is specifically the planning-stage layer, distinct from any tactical, in-flight avoidance.
Why do two competitors cooperate on this at all? Because neither can scale without it. If Flytrex and Wing each had to carve out exclusive airspace, the map of any given metro would run out of room fast, and the FAA would have little basis to approve dense multi-operator BVLOS. Shared, interoperable deconfliction is the thing that lets multiple companies operate over the same neighborhoods — which is to say, it is a precondition for the whole market, not a competitive nicety.
Does this make drone delivery legal everywhere now? No. It is a proof point feeding into the regulatory process. The broader unlock is Part 108, the FAA's anticipated rulemaking for routine BVLOS operations, which will depend heavily on exactly this kind of network-based, multi-operator deconfliction architecture being demonstrated to work. DFW is data for that case, not the case itself.
Why It Matters
Drone delivery's growth ceiling has never really been demand or even battery range — it has been airspace. A single operator can stitch together waivers and exclusive corridors for a while, but a nationwide, multi-company delivery network cannot exist if every operator needs its own private slice of sky. The Flytrex-Wing result in Dallas is significant because it shows the alternative working at non-trivial scale: rival operators, flying 1.36 miles apart over real suburbs, coordinating through an open interoperability standard rather than a controller or a carve-up.
That is the exact architecture the FAA's UTM Operational Evaluation — 17 providers strong as of January 2026 — is built to validate, and the exact capability that Part 108 BVLOS at scale will rest on. If thousands of deconflicted flights a month with zero conflicts holds up as the operation keeps growing, it moves the central question of drone delivery from "can two companies share the sky?" to "how fast can the rest of the network come online?"
