Just weeks after Google Wing launched drone deliveries in Texas, Alphabet Inc. CEO Sundar Pichai revealed on an earnings call that the company tripled its drone deliveries in the first quarter of 2022 — completing more than 50,000 commercial deliveries. That’s up more than three times year over year, Pichai said in late April.

One might think Amazon.com Inc., with its sprawling investments in logistics technology, is a core competitor.

Instead, the e-commerce giant’s drone program has reportedly suffered from nearly a decade of crashes, management turnover, unrealistic timelines and a raft of engineering challenges that have proven remarkably intractable, considering that Amazon spends more on R&D than any other company in the world — $43 billion in 2020 alone, or 11% of sales.

Anyone who has walked down a major city sidewalk since the beginning of the pandemic — and perhaps before — understands the appeal of having their groceries or phone charger glide out of the sky. Streets, walkways and bike lanes are crammed with stressed-out armies of delivery folks, many of them temporary or “gig” workers, sporting reflective vests with their attention glued to their cell phones, often to the exclusion of everything around them. A few years ago, these “last-mile” workers consisted of UPS, FedEx and Amazon drivers, plus a smattering of pizza delivery guys.

What was a trickle is now a roaring, rushing river of humanity thanks to platforms like Uber Eats, Instacart, and DoorDash. But here, economies of scale work in reverse. The more deliveries are demanded by customers, the slower to navigate delivery routes become, while costs grow higher. Couple this with the tightest labor market in decades, and the race for a workable drone delivery network has taken on the urgency of the Berlin airlift.

Except, the drones themselves are simply not working. Amazon, despite its monopolistic aura of invincibility, has yet to crack the code, nor has anyone else. Self-driving semis are racing through the Arizona desert and driverless taxis have taken to the streets of San Francisco, but drones bearing your dinner or prescription have yet to take to the skies over any major city or even suburb. Why is that?

Read more: Amazon Drone Crashes, Delays Dim Jeff Bezos' Delivery Dreams

Physics is an obvious culprit. Ever since the Wright Brothers lifted their aircraft above Kitty Hawk, the flight industry has wrestled with the tradeoffs between weight and lift. You might think that if a double-decker Airbus A380 filled with hundreds of people, their luggage, and enough food and alcohol to service several restaurants can stay skybound for 20 hours, delivering a small package one or two miles in the air would be child’s play. Drones, however, use electricity, and electricity is stored in batteries, which are very heavy. The more weight you need to lift, the more batteries you need, which adds weight — in other words, it’s a vicious cycle.

Another issue is where to land. While delivery drivers have addresses, drones have GPS coordinates that (hopefully) lead them to a suitable landing space. In between, the drone must avoid birds, power lines, wind, other drones, anything that’s happened to fall onto the landing pad before it, and, assumedly, pranksters. Weather, especially, has been an issue for Amazon. “Neither snow nor rain nor heat nor gloom of night stays these couriers from the swift completion of their appointed rounds” is still the Postal Service’s motto, but so far, drones have proven far more susceptible to the elements than trucks and boots.

Finally, there’s the regulatory challenge. Tesla has faced increasing criticism from lawmakers over its “Full-Self Driving” (FSD) feature, and understandably so. FSD has been linked to more than several accidents, though none in which a heavy object falls from the sky in the midst of a densely populated urban area before catching fire. Regulators are well aware that accidents happen, especially with new technologies; no one wants to be blamed for the first drone fatality.

These are the challenges Amazon faces, and ones which, so far at least, it has been unable to conquer. The company acknowledges these setbacks, but says that, like its competitors, it’s plowing on. The last mile has long been a holy grail of the logistics and supply chain industry, and air supremacy would be a supreme competitive advantage over terrestrial slog.

Only very large companies with access to lots and lots of money can even aspire to owning a drone airfleet. Last month, Amazon sold nearly $13 billion in bonds, on top of a $19 billion sale roughly one year ago. Small businesses will have to piggyback onto Federal Express or UPS, both of which are developing their own drone delivery services, but at what cost? The economics favor Amazon, whose vertically integrated structure means that it can make up some of the additional cost of drone delivery in the price of the items it sells on its ever more popular website. Otherwise, it could offset these costs to its third-party merchants, raise the price of its annual Prime membership, or both.

One thing is certain: Amazon will not give up. According to Bezos, the world wants its stuff in 30 minutes or less, like a Domino’s Pizza. The world also wants to feel safe. The thought of dozens (if not hundreds) of drones flying overhead, buzzing like leaf blowers, is not a particularly pleasant one for most people, and probably not one that keeps Bezos up at night.

Annoyance does not trump survival. There is little doubt that Amazon sees drone delivery much as it did the Kindle — a company-saving innovation and moon shot rolled into one. In its all-important holiday quarter last year, Amazon actually lost money on its core e-commerce business, both within the U.S. and overseas, despite record sales. This year, an Amazon warehouse voted to unionize for the first time ever, just as a series of unflattering reports about delivery subcontractors going bankrupt started appearing in the media — the same media that’s now making hay of Amazon’s travails in drone delivery.

Amazon contested the unionization vote, but its business model clearly needs to be disrupted once again. Getting rid of all of those drivers and trucks is the only conceivable path to sustainable profits, especially as other costs increase and record inflation begins to slow retail sales.

Amazon has proved remarkably adept at disrupting itself in the past. Still, the important question is whether or not drone delivery will ever be a practicable last-mile solution. So far, the answer to that question is a loud and resounding no.

Shaun Passley is founder of ZenaTech.