It's just wildly inefficient, and will never see serious usage outside of tiny niches. Not until we get science fiction antigravity or tiny fusion reactors or something.
I'm not sure the laws of physics are on the side of the status quo here.
Even if a drone could carry the energy equivalent of a 20 gallon fuel tank with it, it would waste the majority of that energy both keeping itself and its energy aloft.
Tires and other ground plane vehicles, mitigate the majority of that waste in exchange for increased navigational difficulty (regarding obstacles like buildings and trees but also moving objects like vehicles in its path).
The self-driving robots made some sense but only for very local (couple of block radius) delivery.
A paraglider drone with a tiny parachute would be more efficient than any 3/4/6/8 motor drone copter
the rest of the money went into the gas tank of the truck
you and everyone else who ordered a cable that day on that route chipped in to fill the truck up
The economics are totally different, though. The "product" is very high value and has a short shelf life, so being able to deliver it quickly is a big benefit.
On top of that, using a fixed-wing delivery vehicle and launching it with a human-powered slingshot makes it actually reasonably energy-efficient.
See also https://youtu.be/DOWDNBu9DkU From 3 months ago
Maybe it's my proximity to large bodies of water, but there are many island communities where I live that have limited (or no) road access requiring either a very out-of-the-way route to reach a destination or -- worse -- a ferry. A trip out to some communities could be the difference between a mile travel by air vs 10 miles travel by vehicle.
That's an extreme outlier, obviously, but if you dial that back a bit and use, say, "a trip from my home to my girlfriend's" which involves working around two rivers that roads only cross at certain points, the difference is 5 miles "as the crow flies" vs almost 14 miles of windy roads.
When all of the other problems are worked out, a drone delivery service could augment regular delivery to reduce costs/improve delivery time for products.
I think products that greatly benefit from improved delivery time, though, is the place where drone delivery can be successful. It's currently being used in some countries for blood/medical deliveries and that was the first thing that came to mind when I thought of places where drone delivery would be able to be compete/displace existing providers.
I had a friend who flew a helicopter for a company that moved people/organs/other emergency-releated activities for hospitals. Organ transplants, as I understand them, are extremely time-sensitive. One day he was called on to deliver a liver from a hospital nearby to a hospital mid-state for a waiting patient. Weather, wind, clouds and other variables were excellent; he was on-call and a ways out but made it. Shortly after take-off the helicopter suffered a failure (memory escapes me), it crashed resulting in the death of my friend, (I think) a passenger and ultimately the patient waiting for the organ.
This was 20-ish years ago and in the time since then it was discovered that part of the pre-flight check was skipped and IIRC[0] the crash was determined to be related to condensation in the fuel tank or something along those lines -- basically, he was on-call with a tight delivery schedule and cut a corner to account for the time it took him to get to the hospital for the delivery.
Drone delivery would improve several things in this model. The most obvious one being "no dead pilot or passenger" when something goes awry. If we're "flying things around" especially using small aircraft and GA pilots, there's going to be crashes -- drones or people involved -- it's a guarantee. Using drones would enable deliveries to take place without delays involved in "waking up a pilot on-call/waiting for their arrival" -- allowing hospital staff to prepare the delivery vehicle, make a call, and hand off to a remote drone pilot to get it to its destination. Assuming software/user interfaces can make piloting a drone remotely as safe as flying a helicopter "in person", the benefits would be enough to offset the higher delivery costs.
[0] I tried to find a reference online to refresh my memory but was unable so forgive my memory if I've got some details wrong here.