In Ukraine, Russia’s low-cost drones are forcing defenders into a costly equation: each interception can cost ten times as much as the weapon it destroys. For European armies, military advantage increasingly depends not only on performance, but also on the cost of attack and defence, industrial capacity and the speed of replacement. Boris Laurent, Defence Manager at Sopra Steria Next, and Christophe Eugène, Defence Adviser at Sopra Steria, examine the changing economics of modern warfare.
Is it better to have 15,000 bayonet-armed infantrymen or three cutting-edge cannons? The question is as old as war itself, and the war in Ukraine might seem merely to revive it, with drones taking the place of infantrymen. Rear Admiral (ret.) Christophe Eugène, Defence Adviser at Sopra Steria, does not dispute the comparison: “Saturation has always been sought after, because to saturate the enemy is to force them to make choices, reduce their ability to defend themselves, and in a sense, to surprise them.”
What has changed in Ukraine, according to Eugène and Boris Laurent, is the cost of deploying on mass, who bears that cost and how long each side can sustain it.
The data tsunami
The first shift is industrial. Few anticipated how far drone costs would fall, or how many could be produced. A mass salvo also seeks to overwhelm command structures: “The aim is to saturate enemy command to prevent it from seeing what’s happening behind or nearby. While it’s focused on the saturation, it loses sight of the rest of the battlespace,” explains Eugène.
The second shift is more alarming and concerns the digital realm, which the admiral describes as submerged under a “data tsunami effect”. Data arrives from all directions: the adversary floods the defender with information, while the defender’s own sensors and effectors generate still more. “We’re going to have to sort through all of this.” Artificial intelligence becomes indispensable for processing these torrents, but as Eugène notes, every AI system connected to a network also creates potential vulnerabilities that must be secured.
Laurent adds an important qualification drawn from military history: numbers matter only if used effectively. “You can send 10,000 men into battle in complete disorder, achieving nothing – or send fewer, better organised, with a clearer, more precise effect in mind.”
Over 75% losses – and rising
The core question is therefore economic. The CSIS illustrated this imbalance in its analysis of Russia’s Shahed campaign: drones costing between $20,000 and $50,000 face interceptor missiles costing several hundred thousand dollars apiece. Between early 2024 and March 2025, launches rose from about 200 to more than 1,000 per week. Moscow accepts losses exceeding 75% of the drones it launches.
Does this mean saturation always benefits the attacker? Laurent disagrees. “It benefits whoever achieves the most favorable ratio. If the attacker spends far more than the defender, they’ll eventually face industrial, economic, or supply-chain constraints.” The battlefield reveals only part of the equation. “Operations are just the tip of the spear.” Behind that tip lie financial and human costs, including soldiers’ cognitive fatigue.
Russia continues to raise the stakes: its newer, faster drones are more expensive, yet Moscow continues to accept high loss rates – a sign, according to Laurent, that its industrial base can still keep pace – for now.
Eugène places this calculation within the traditional framework of military planning, which combines diplomacy, information, armed forces and economics. There is no magic threshold. “There’s no ideal formula that says: above a hundred, you’re in a position of strength. The equation is never that simple.” Human factors complicate any purely mathematical calculation.
Moscow is waging a war of attrition, described in Russian doctrine as “nonlinear”, aimed at exhausting Ukraine and, indirectly, its European supporters. Saturation thus becomes cognitive, informational and cyber as much as physical or aerial. “They send soldiers to the front like logs into a fire,” summarises Laurent. “They don’t view human losses the way we do.”
The right effect at the right price
Are Ukraine’s low-cost interceptors – costing only a few thousand dollars – a model to follow? Only partly. Laurent’s lesson is to counter each threat with the least costly means capable of neutralising it: electronic warfare, jamming, guns, interceptor drones and, only when necessary, missiles. “We must keep highly capable systems for the threats that deserve them – but avoid using them for every threat.”
This approach has limits, as the enemy adapts quickly. In late September, Volodymyr Zelensky reported that Ukrainian defences were intercepting only 57% of jet-powered Shahed drones, compared with 70% across all drones. A low-cost countermeasure can lose its effectiveness within months.
The defender therefore needs a range of responses – and, above all, a system capable of selecting the right one. That is the logic behind BOREADES, the anti-drone warfare system developed since 2015 by CS GROUP, a Sopra Steria subsidiary. Its open, modular architecture combines radar, radiofrequency and optronic sensors to build a coherent tactical picture, giving operators the information they need to assign the appropriate effector – from jamming to physical neutralisation – to each detected threat.
Its interoperability, through standards such as Link 16, also allows it to share situational awareness and commands with other air-defence units and command centres, improving resource allocation under saturation and reducing the risk of friendly fire. In late August, during NATO’s Baltic Trust 2026 anti-drone exercise in Latvia, the system received the top performance award for its role as a level-2 command-and-control (C2) system within a multinational architecture. “In saturation scenarios, value shifts from the missile itself to the system that decides which missile or other effect to use,” summarises Laurent.
A wartime procurement code?
In France, Eugène identifies a widely shared lesson: “The decision loop must move faster.” Between detection and response, digitalisation and AI can save critical time. Ukraine’s Delta command system illustrates this: by harnessing data, Ukrainian forces can respond to threats almost instantly.
Yet Eugène warns against overcorrecting. The French military model remains built around nuclear deterrence, and “the war of tomorrow will not be the war of today.”
There are signs of adaptation, but the admiral believes there is still a long way to go. “People say we need to move to a war economy. But a public procurement code and a wartime footing – those two don’t go together.” Specifications from France’s defence procurement agency, the DGA, will have to evolve: “A 50-page requirement document for a need that must be met tomorrow – that just doesn’t work.”
Laurent argues for two complementary tempos. Major programmes (SSBNs, attack submarines and aircraft carriers) remain essential. At the same time, industry must learn to operate in shorter cycles – something many manufacturers recognise but have yet to achieve at scale.
Eugène sees a risk here that the state must accept, and a role for major groups: supporting start-ups “without being swallowed, giving them a reason to live,” long enough for them to move from the garage to industrialisation. Orders, however, must follow announcements. “Declarations promising a hundred billion when there’s no money can end up killing small firms.” Without contracts, warns Laurent, that execution talent goes to the highest bidder – “and that’s often the Americans.”
Sustaining this effort over time also requires a change of mindset. “We’re coming out of 30 years of peace, from a world where Montesquieu’s ‘doux commerce’ was expected to solve everything,” remarks Laurent. “The awakening is brutal.” Eugène cites Finland and Estonia, which recognised much earlier that resilience mobilises all of society, not just the armed forces.
Being ready for the next war
Before buying equipment, armies must first know how to deploy equipment effectively at scale. “The first step is doctrine,” insists Laurent. “What type of drone? For strike or observation? Aerial or tracked robot?” Military capability is defined by the DORESE model (doctrine, organisation, human resources, equipment, support and training), and even general staffs will need to reinvent themselves – operating in smaller formations with the smallest possible electromagnetic footprint.
In this approach, sophistication still has its place – but at a different level. “High technology doesn’t disappear; it shifts position,” summarises Laurent. Satellites, radars and command software coordinate large numbers of cheaper systems. In September, CS GROUP acquired the technologies and teams of UAVIA, a specialist in supervising fleets of drones and robots from multiple manufacturers and linking them to command centres.
Producing 10,000 drones is of little use if you cannot coordinate them or exploit the data they generate. “It’s far from simple,” concludes Eugène. “But just because it’s far from simple doesn’t mean it shouldn’t be done. The path is there – and we must follow it to be ready for the next war.”