Ukrainian Disclosures Offer Insight Into Scale of Russian Strike Drone and Decoy Decoy Launches, Projected Production – Part Two

Note: This text is the second part of a SPAS Consulting analysis examining the scale of Russian strike drone and decoy drone launches and projected production. Part one focuses on the current composition of Russia's arsenal of strike drones and decoy drones. Part two focuses on the question of numbers.

Part Two of this SPAS Consulting analysis on Russia’s employment of single-use propeller-driven strike drones focuses on the number of strike drones that Russia has employed to date and is projected to manufacture and perhaps employ in 2025. This analysis relies on Ukrainian disclosures of the number of strike drones – as well as decoy drones – launched by Russia against targets located in Ukraine. Regular Ukrainian press releases provide observers figures for the number of strike munitions of a given type including the Shahed-136/Geran-2 single-use propeller-driven strike drone launched by Russia as well as the number of strike munitions that Ukraine claims to have intercepted. These regular – often daily – Ukrainian press releases collectively amount to a dataset that, while imperfect, purposefully curated, and subject to considerable uncertainty, observers can use to track the effects and effectiveness of Russia’s employment of strike munitions including strike drones in the Russia-Ukraine War. It is important to note that the Russian military does not issue comparable press releases on how many strike drones it launches against targets in Ukraine (or how many strike drones Ukraine launches against targets in Russia). As a result, observers are forced to rely almost exclusively on Ukrainian disclosures for insight into the scale of Russian strike drone employment. The remainder of this analysis draws upon an open-source dataset developed and updated by Petro Ivaniuk that compiles the regular disclosures of Russian strike drone activity made by the Ukrainian military.

A Note on Ukrainian Data on Russian Strike Drone Launches

This SPAS Consulting analysis draws heavily on an open-source dataset that compiles the data contained in regular Ukrainian military press releases on the number of strike munitions, including single-use propeller-driven strike drones, that Ukraine claims Russia to have launched and Ukraine to have intercepted. These Ukrainian disclosures are, of course, selective, curated, and intended to serve Ukrainian purposes and should, therefore, be interpreted with caution not least in a context in which observers are in no position to independently verify Ukraine’s claims as to the number of strike munitions reportedly launched by Russia let alone the number reportedly intercepted by Ukraine. The sheer number of strike munitions employed in a high-intensity conflict that has entered its fourth year only compounds the analytical challenge encountered by observers.

As with all manually compiled datasets, there is scope for human error and a possibility exists that the Ukrainian military’s regular press releases are of less than perfect accuracy. Uncertainties notwithstanding, the open-source dataset compiling the Ukrainian military’s regular press releases amounts to the best data available to observers and the situation is unlikely to change until analysts and historians look into the matter following a durable ceasefire if not a final resolution to the Russia-Ukraine War. There is, however, one major discernible flaw in the Ukrainian military’s regular press releases that is carried over into the open-source dataset, a dataset that, like the underlying official data, has been widely used in media reporting and open-source analyses. As explained in part one of this SPAS Consulting analysis, Russia has since July 2024 launched ever-increasing numbers of armed and unarmed propeller-driven single-use decoy drones alongside its single-use propeller-driven strike drones. Ukraine’s regular disclosures of the launch and interception of Russian strike drones do not disaggregate strike drones from decoy drones, and the open-source dataset and the underlying data contained in regular Ukrainian press releases are as such increasingly polluted. 

While it is possible that the Ukrainian military has perfect knowledge of the exact type and number of Russian drones, whether strike drones, decoy drones, or any other type of propeller-driven drone, operating in the airspace above Ukrainian-controlled territory at any given time, observers must take into account the fact that Ukraine is highly reliant on passive and acoustic sensors to detect and track the activity of propeller-driven drones, whether these are strike drones or decoy drones. Russia’s Gerbera and Parodiya decoy drones, which are examined in part one, likely amount to medium-fidelity acoustic decoys and can therefore be challenging for Ukraine’s human air defence observers and Ukraine’s network of acoustic sensors to discriminate from Russia’s strike drones. Ukraine is therefore unlikely to have perfect knowledge of the exact type and number of Russian drones operating in the airspace above Ukrainian-controlled territory at any given time. 

Given this, there is a case to be made that Ukraine’s regular press releases deliberately inflate the number of strike drones that Russia reportedly launches and Ukraine reportedly intercepts by purposefully not excluding known let alone unknown decoy drones, whether armed or unarmed, in the regularly disclosed press releases. It is, of course, possible, even likely, that the Ukrainian military itself cannot always discriminate between propeller-driven strike drones and propeller-driven decoy drones or at least cannot discriminate between these fast enough to facilitate accurate daily reporting. The increasingly polluted state of the open-source dataset since the autumn of 2024, which is after Russia began to deploy decoy drones and the time when Russia initiated its winter 2024-2025 strike campaign against Ukraine’s energy infrastructure, may therefore primarily reflect the challenges of discriminating between strike drones and decoy drones with similar acoustic signatures even in a context in which Ukraine benefits from the non-disaggregation of decoy drones from its regular disclosures of Russian strike drone launch activity.

How Many Shahed-136/Geran-2/Garpiya Strike Drones Has Russia Launched to Date?

According to Petro Ivaniuk’s open-source dataset (hereafter “open-source dataset”) compiling the Ukrainian military’s regular press releases, Russia launched a total of 21,187 strike drones and decoy drones from 29 September 2022 to 28 February 2025. Given the limitations of the open-source dataset and the underlying data contained in Ukraine’s regular press releases, it is important for observers to consider these figures to be approximations. Figure 1 plots the running total of reported Russian strike drone and decoy drone launches from September 2022 to the end of February 2025.

Over an 884-day timeframe, Russia launched an average of 23.96 strike drones per day and an average of 719 strike drones per 30-day month. These average figures are of course misleading for two reasons. First, average daily launch rates were much lower throughout 2022 and 2023 relative to 2024 and the first two months of 2025. Second, the open-source dataset and Ukrainian disclosures do not distinguish between the increasing numbers of Gerbera and Parodiya decoy drones that Russia launched in increasing alongside its strike drones from July 2024 onward. Figure 2 disaggregates the data from the open-source dataset to reflect reported Russian strike drone and decoy drone launches by year.

As previously noted, the total number of strike drones and decoy drones – the latter being launched in increasingly large numbers from July 2024 onward – that Russia reportedly launched throughout 2024 dwarf the figures for 2023 and 2022. According to the open-source dataset, Russia launched nearly two-thirds of the total number of strike drones and decoy drones (launched up to the end of February 2025) in 2024. Should Russia sustain anything close to the launch rate observed in the first two months of 2025 over the rest of the current year, then the total figure for 2025 – including a very large number of decoy drones – may similarly dwarf the total figure for 2024. It is productive to visualize reported strike drone and decoy drone launches by year in a pie chart, which is provided in Figure 3.

Although the aggregated annualized data is interesting, strike drones – and decoy drones – are not uniformly launched over time or across space. There is insufficient data to disaggregate the open-source dataset across space and Figure 4 therefore plots reported Russian strike drone and decoy drone launches by month. One advantage of disaggregating the open-source dataset by month is that it allows observers to examine the figures for the months before July 2024, which are likely to almost exclusively refer to the employment of Shahed-136/Geran-2/Garpiya strike drones and few if any decoy drones.

The disaggregated monthly launches plotted in Figure 4 highlight the fact that reported Russian strike drone and – from July 2024 onward – decoy drone launches are clustered around two campaigns targeting Ukraine’s energy infrastructure in the winter of 2023-2024 and 2024-2025 (the latter started earlier in autumn 2024). These winter peaks not only highlight the focus of Russian targeting, which when successful is more effective in cold weather, but are also indicative of stockpiling. Russia appears to throttle launch rates for months on end in preparation for these winter strike campaigns targeting Ukraine’s energy infrastructure. Stockpiling and just-in-time delivery and the immediate expenditure of strike drones are, however, not mutually exclusive and Russia is likely to rapidly launch/consume newly produced strike drones during its winter campaigns targeting Ukraine’s energy infrastructure.

The monthly launch data drawn from the open-source dataset can also be viewed from the perspective of Ukraine: Ukraine has months to prepare its air defences for Russia’s winter strike campaigns targeting its energy infrastructure. It is an open question whether Russia’s employment of the Shahed-136/Geran-2/Garpiya would have been more effective to date if Russia could sustain a higher average number of launches outside of its winter campaigns against Ukraine’s energy infrastructure. This, of course, would require a much larger stockpile of strike drones and, as such, a higher production rate. It is worth noting that there is no indication that Russia is currently stockpiling strike drones for employment beyond the Russia-Ukraine War – all strike drones that Russia has produced to date during the war appear to be intended for near-term if not immediate use against Ukraine.

Although variation in reported launch rates is likely to be primarily reflective of Russian production output and preparations for winter campaigns targeting Ukraine’s energy infrastructure, some of the variability is likely to result from other factors including weather. While weather systems tend to be localized and are unlikely to affect the entire expanse of the Russia-Ukraine War on a given day, it is possible that Russian targeting and mission planning lacks flexibility such that it amounts to something close to an all-or-nothing approach in which poor weather around one launch area can affect launch activity in another area that is unaffected by poor weather. It is therefore productive to shift the analytical aperture from calendar months to other time intervals. Figure 5 plots the reported rolling three-month average of Russian strike drone and decoy drone launches.

The three-month rolling average plotted in Figure 5 underscores the clustering of Russian launch activity in two winter strike campaigns targeting Ukraine’s energy infrastructure. The clustering of reported launches in the winter 2023-2024 campaign is particularly notable in that this strike campaign took place before Russia is understood to have begun launching ever-increasing numbers of decoy drones alongside its strike drones. That is, Russia appears to have spent months developing a large stockpile of strike drones in order to have launched so many during the winter 2023-2024 strike campaign. Disaggregating reported launch activity by day allows observers to further visualize variability in launch activity across time. Figure 6 plots reported daily Russian strike drone and decoy drone launches from September through the end of February 2025. 

Disaggregating the open-source dataset by day highlights the relative and absolute paucity of strike drone launches on a given day in 2022 and 2023, which was long before Russia greatly increased production of the Shahed-136/Geran-2 let alone launched increasingly large numbers of decoy drones alongside its strike drones. Figure 6 also highlights the intermittence and variability of launch activity prior to July 2024. The open-source dataset suggests that Russia lacked a sufficient number of strike drones to consistently launch one let alone twenty strike drones per day until increasingly large numbers of decoy drones, which can not be disaggregated in this open-source dataset or the regular – often daily – Ukrainian disclosures that the open-source dataset compiles, became available. Above all, Figure 6 highlights a dramatic spike in launch activity in the second half of 2024 and the first two months of 2025. This is, of course, a timeframe in which Russia is understood to have launched decoy drones in ever-increasing numbers and it therefore stands to reason that this dramatic sustained increase in launch activity is likely to be primarily driven by the production and employment of decoy drones in ever-increasing numbers with the stockpiling of strike drones likely to be of secondary if not tertiary importance.

With some two-thirds of Russia’s reported strike drone and decoy drones having been launched in 2024, it is productive to separately visualize daily launches in that year. Figure 7 plots reported daily Russian strike drone and decoy launches in 2024.

Figure 7 highlights the intermittence and variability of strike drone – and from July onward decoy drone – launch activity prior throughout 2024. According to the open-source dataset compiling regular Ukrainian press releases, Russia launched fewer than fifty strike drones and/or decoy drones on most days in 2024.

Collectively, the charts presented in part one and part two of this SPAS Consulting analysis highlight that Russia has so far not had at its disposal a sufficiently large number of strike drones to realize its ambitious goals. The two winter campaigns targeting Ukraine’s energy infrastructure have doubtless had a non-zero effect on the country but Ukraine claims to have neutralized most of the strike drones and other strike munitions employed in these campaigns. Not only did these two winter strike campaigns, and the more haphazard and smaller-scale strike drone launch activity undertaken from autumn of 2022 through early 2023, fail to bring Ukraine to capitulation, but the employment of strike drones against targets near population centers in general and the targeting of Ukraine’s energy infrastructure in particular only galvanized diplomatic and military support for Ukraine. This includes what amounts to the ongoing open-ended transfer of air defence systems to Ukraine, transfers that have had the effect of further diminishing the per-unit effectiveness of Shahed-136/Geran-2/Garpiya strike drones as well as other Russian strike munitions. 

It also bears emphasis that although the winter strike campaigns targeting Ukraine’s energy infrastructure have been impressive in terms of the total number of strike drones and/or decoy drones that Russia reportedly launched, this reflects in part an intertemporal tradeoff resulting from stockpiling. Simply put, Russia forwent launches in the rest of 2023 and 2024 – and thereby lowered the maximum potential effectiveness of its strike drones at other times of these years – in order to undertake its concerted and inconclusive winter campaigns targeting Ukraine’s energy infrastructure. In a world of infinite resources and infinite production capacity, Russia would not have needed to accept such a tradeoff and dilute the effects of its strike drones over time. The remainder of this analysis will focus on recent disclosures about the number of strike drones and decoy drones that Russia is projected to manufacture in 2025. Realized production figures are very likely to play a large part in determining the effects and effectiveness of Russia’s employment of single-use propeller-driven strike drones over the course of the fourth year of the Russia-Ukraine War.

How Many Shahed-136/Geran-2/Garpiya Strike Drones Is Russia Projected to Manufacture In 2025?

The available data on Russian strike drone employment over time, which in this analysis is entirely drawn from an open-source dataset compiling the regular disclosures made by the Ukrainian military, amounts to a lagging indicator. Observers require a leading indicator to assess the potential effects and effectiveness of Russian strike drone employment going forward. Estimating Russian strike drone production requires observers to turn to irregular official and unofficial disclosures made by Ukrainian officials often through Ukrainian media.

According to a Ukrainian media report from March 2025 that claims to draw upon sources in the Ukrainian government, Russia has the capacity to manufacture a combined total of more than 3000 strike drones and decoy drones per month. The production output figures offered in this Ukrainian media report are not only subject to uncertainty but may of course also increase or decrease over the course of 2025 for any number of reasons. The Ukrainian media report specifies monthly production capacity, which is laid out in the second column of Table 1. The figures for assumed annual production capacity laid out in column four reflect the multiplication of the reported monthly production capacity figure by twelve and are therefore subject to even greater uncertainty than the monthly production capacity figures provided in the Ukrainian media report.

Uncertainties and the presence of suspiciously round numbers notwithstanding, the figures laid out in Table 1 are of considerable analytical value. Supposing that the monthly production capacity figures laid out in the Ukrainian media report are accurate and that Russia’s production output either remains constant or increases and does not decline, then Russia may be able to build a combined total of 1300 Geran-2/Garpiya strike drones per month, which is to say 15,600 per year. Assuming a stockpile of zero Geran-2/Garpiya strike drones on 1 January 2025 and 1 January 2026, which are not realistic assumptions, Russia will be able to launch an average of up to 42.73 Geran-2/Garpiya strike drones per day throughout 2025. For context, the open-source dataset compiling regular Ukrainian press releases indicates that Russia launched a combined total of 17,734 strike drones and decoy drones in 2024, which is to say an average of 48.45 strike drones and decoy drones per day in 2024. Many of these are understood to have been Gerbera and Parodiya decoy drones. The figures laid out in Table 1 suggest that Russia will be able to launch a much larger number of Geran-2/Garpiya strike drones alone over the course of 2025.

It bears emphasis that there is considerable uncertainty associated with the projections for Geran-2/Garpiya production in 2025. It is not clear if the factory in Alabuga will continue to produce the Geran-2 under licence now that the original commercial contract with Iran, which may have been extended or replaced, has apparently been completed. Production of the Garpiya, which is reportedly a reverse-engineered Shahed-136, at a factory in Izhevsk appears to be neither bound up in the commercial contract with Iran nor reliant on the import of components from Iran. Although production of the Garpiya is reportedly reliant on imported foreign components including commercial-off-the-shelf components from China, it is possible that the Izhevsk production line will scale up production much faster than the Alabuga production line and perhaps dwarf the latter’s production output. While the Alabuga factory is likely to continue manufacturing Geran-2 strike drones and may transition to the Garpiya, it is important to note that the Alabuga factory is also responsible for the production of the Gerbera decoy drone and may shift production capacity to one or more other new strike drone and/or decoy drone designs.

The estimated production rates for the Gerbera and Parodiya decoy drones laid out in Table 1 are even more impressive than those for the Geran-2/Garpiya. Supposing that the monthly production capacity figures disclosed in the Ukrainian media report are accurate and that Russia’s production output either remains constant or increases and does not decline, then Russia may be able to build 1600 Gerbera and 100 Parodiya decoy drones per month, which is to say 19,200 Gerbera and 1200 Parodiya decoy drones per year. The combined estimated production figures for decoy drones notably amount to 56.66% of the combined estimated total for strike drones and decoy drones in 2025. 

The lower cost of production and uncertainties notwithstanding, these figures make clear that it is increasingly problematic to discuss Russia’s production and employment of single-use propeller-driven strike drones without mentioning the closely intertwined production and employment of single-use propeller-driven decoy drones. The shifting priorities of Russian drone production and employment are highlighted by JSC Alabuga’s reported production capacity of 900 Geran-2 strike drones alongside 1600 Gerbera decoy drones per month. All else equal, and all else is not equal when it comes to Russia’s military-industrial capacity, the Alabuga factory can in principle channel its finite resources toward increasing Geran-2 strike drone production to the maximum possible level but the priorities of the Russian military, the only customer for JSC Alabuga’s production output, have clearly shifted toward a combination of strike drones and decoy drones. This observation is instructive in that it offers a window into the Russian military’s perception of the state of its strike capabilities vis-a-vis Ukraine and the state of measure-countermeasure dynamics more generally. Decoys are, all else equal, a waste of resources that could be better allocated toward actual strike drones or other military capabilities. Decoys are only built and employed in such large numbers when the strike munition(s) that decoys are intended to mimic wane in effectiveness.

The variance in the production capacity for the Gerbera and Parodiya decoy drones in the Ukrainian media report is particularly interesting and, if accurate, will likely have major implications for the effects and effectiveness of Russian strike drone employment for as long as the production of the Gerbera dwarfs that of Parodiya in this manner. It bears emphasis that there is very limited public information on the Parodiya at this time including the factory or factories responsible for its production. Public sources indicate that the Gerbera was deployed as early as July 2024 and that the Parodiya was deployed as early as October 2024. It is possible that the variance in production output is explained in part by the later introduction of the Parodiya. Whatever the facts of the matter, the reported production output of 100 Parodiya decoy drones per month is inadequate for Parodiya decoy drones to accompany the increasingly large numbers of Geran-2/Garpiya strike drones available for launch in salvos toward the intended target – the shorter maximum range of the Gerbera precludes its use in this manner against more distant targets including Kyiv. It is possible, even likely, that Russia will expand monthly production of the Parodiya well beyond the reported figure of 100 units.

Uncertainties notwithstanding, the projected production figures laid out in Table 1 on the basis of a recent Ukrainian media report are both interesting and of considerable analytical value but ultimately constitute one of several ways in which observers can approach the question of numbers. Access to perfectly accurate production figures is of limited benefit to an analyst because it amounts to a numerator without a denominator. Analysts with access to perfectly accurate production figures must still determine the realized probability and estimate the expected probability at which a given strike drone can bypass adversary air defences and detonate upon impact with its intended target, among other things. As the Russia-Ukraine War and other conflicts have put on stark display, strike munitions employed in large numbers and in a protracted conflict – as opposed to small numbers in a one-off surprise attack – are unlikely to have a 100% penetration rate, a 100% accuracy rate, and a 100% detonation rate. An analyst can address the issue of numbers in a different manner by postulating a declared objective to launch x or y strike drones per unit of time independent of the probability of penetrating air defences, accurately reaching the intended target, and detonating upon impact. This approach is, of course, limited but does not require an actual or assumed denominator and helps an analyst make sense of realized and projected production output figures.

Suppose that Russia’s senior leadership decrees that the Russian military must launch an average of 50 strike drones per day over a long timeframe such as a year. Assuming that the Russian military has at its immediate disposal a stockpile of zero strike drones and intends to end the war with a stockpile of zero strike drones, this will require an annual production output of 18,250 strike drones per 365-day period, which is to say 1500 per 30-day month and, of course, 50 per day. Supposing that Russia’s senior leadership decrees that the Russian military must launch an average of 100 strike drones per day, this will require an annual production output of 36,500 strike drones per 365-day period, which is to say 3000 per 30-day month and, of course, 100 per day. Assuming that the projected Russian production figures for 2025 laid out in Table 1 are accurate, Russia may produce a total of 15,600 Geran-2/Garpiya strike drones in 2025. If realized, this will facilitate the launch of an average of 42.73 strike drones over the course of a year. 

Viewed in these terms, Russia has a long way to go before it can sustain the employment of an average of 50 strike drones per day over the course of a year let alone an average of 100 or more per day. The mass production and regular large-scale employment of decoy drones can improve the probability that a Russian strike drone will penetrate Ukrainian airspace, accurately reach the intended target, and detonate upon impact. It will not, however, increase the number of discrete aimpoints that Russia can attack with Shahed-136/Geran-2/Garpiya strike drones or the maximum amount of damage that Russia can inflict on Ukraine’s energy infrastructure and/or any other set of targets. While quantity can have a quality all of its own, the interplay of Russia’s approaches to the employment of single-use propeller-driven strike drones to date and intensifying measure-countermeasure dynamics have dramatically raised the floor – as opposed to the ceiling – for what amounts to a high probability of success strike against targets in Ukraine. Publicly available data indicates that Russia has to date lacked the quantity of relatively low-cost strike drones required to meet its military objectives. Publicly available data also indicates that Russia is unlikely to have such a quantity of strike drones at its disposal throughout 2025 that Russian Shahed-136/Geran-2/Garpiya single-use propeller-driven strike drones can have a quality all of their own in the Russia-Ukraine War.