Ukraine Military History Institute
Everyone wants a piece of deep strike, and that is the problem
The most concrete criticism Rob Lee makes of Ukraine's deep-strike effort is not about targets or effects. It is about how many organisations are running it.
He counts them off in one breath. Three separate bodies doing deep strike, the SBU and the Unmanned Forces among them, and then he corrects himself mid-sentence to add special operations as a fourth. Dmytro Putiata interjects to point out that the same organisations are doing middle strike too. And inside the Unmanned Forces, Lee adds, there is more than one unit on the mission.
His expectation is that the list will grow. He is sure new organisations will turn up doing deep strike later, because everyone wants to be involved in it.
The motive is not cynical, and Lee does not present it as one. Everyone wants a role in undermining Russia's capacity to wage the war, which is a reasonable thing for a service to want. His judgement is simply that it would be much more efficient to have one organisation dedicated to the mission.
The mechanism behind the duplication is bureaucratic, not operational
Lee is specific about why this happens, and it is worth separating from the usual complaint about waste.
Some units, he says, got moved to one direction for bureaucratic reasons and for political reasons. There is always internal jockeying between different branches of the military, one wanting to be stronger than another, and the product of that jockeying is redundant capability.
That is a different diagnosis from inefficiency. Redundancy created by competition is stable, because each organisation has an institutional interest in keeping its own version of the capability. It does not get consolidated by anyone noticing the overlap.
Putiata's version is blunter
Putiata agrees fully, and his framing is the one that will sting.
Right now, he says, everyone is doing what he thinks is right, and what looks good. His example is the one everybody has seen: oil refineries are burning, and it looks nice.
Then the line that matters. Nobody is asking about the efficiency, or how it was implemented.
That is a measurement problem rather than a targeting problem. Spectacular imagery is self-validating. If four organisations are each producing spectacular imagery, and none of them is being assessed on cost per effect, the system has no mechanism for discovering which of the four is actually good at this.
Drones augment, they do not replace
The second half of the argument is addressed outward, to the militaries watching this war and deciding what to buy.
Lee's rule is restrained. For a Western military, and particularly for a United States military with a very strong air component and a range of unique capabilities, drones should be understood as something that augments and complements what already exists. Not something that changes everything immediately.
His procurement test follows from that: do not procure drones for the sake of procuring drones. Procure them when they make units more lethal, more effective and more flexible.
Putiata arrives at the same place from the Ukrainian side, and adds the arithmetic. This war showed that you do not need a 200 thousand dollar HIMARS missile to do the job. Much cheaper drones will do the same. His conclusion is not that drones replace the rest, it is that drones belong in the ecosystem rather than beside it: combined strikes in which deep-strike drones work alongside ballistic and cruise missiles against a single target, and alongside artillery and tanks on the battlefield.
Drones, in his phrase, are not absolute.
He names the two wrong positions explicitly. One holds that drones will destroy everything and do everything. The other holds that drones are for less developed militaries and that serious armies will fight differently. He is willing to concede part of the second: other nations will find a different way. What he will not concede is ignoring drones, and ignoring detection and protection against them.
Why Ukraine's model does not port cleanly
The distinction that matters is the one explaining why copying Ukraine would be a mistake even if everything Ukraine does works.
Most of Ukraine's structural adaptations have happened since 2023, Lee points out, and throughout that period Ukraine has been strategically on the defensive. So the drone-era adaptations are the adaptations of a defensive military. The Unmanned Forces mission, as he describes it, is to kill as many Russians as possible and inflict losses Russia cannot sustain until it can no longer continue the war.
That objective is not always tied to manoeuvre. Plenty of Ukrainian units and operations, he says, have no manoeuvre component at all.
For the United States military, the relationship runs the other way. Fires by itself is never the purpose. The purpose is fires plus manoeuvre, and the question asked of any new capability is how it enables manoeuvre.
So Ukraine's answer is correct for Ukraine's strategy and would not necessarily be correct for anyone else's. Putiata's recommendation to NATO and non-NATO countries alike is the two-step version of the same point: learn as much as possible about what Ukraine and Russia are actually doing, and only then think about how to implement it at home.
Lee closes with a reminder of what the gap looks like from outside. Israeli forces in Lebanon have had difficulty with Hezbollah FPV teams operating at what he estimates is a 2023 level of capability. Putiata cuts in: not even close to what Ukraine and Russia are doing now.
What to watch
- Whether the deep-strike organisation count goes up or down. Lee predicts it will go up. A consolidation into one dedicated organisation is the falsifiable alternative, and either outcome is observable.
- Whether anyone publishes an efficiency measure for strike missions. Putiata's charge is that nobody is asking how it was implemented. A cost-per-effect assessment appearing anywhere would settle whether that has changed.
- Whether Western drone procurement is justified by capability gaps or by volume. Lee's test is lethality, effectiveness and flexibility. Programmes that cite quantities instead are the ones failing it.