The past several months have produced a steady stream of announcements and market signals that point in one direction: the hardware-centric model of industrial automation is giving way to something more fluid. The shift is not a single event but a convergence of product launches, pricing experiments, and strategic positioning by companies that see recurring revenue as the path to broader adoption.
One of the clearest signals came from Dwbrobot, a France-based robotics provider, which in early April 2026 formally introduced what it calls a “zero-investment” robot model. The company’s platform is designed to support both repetitive tasks and more complex industrial processes, and the headline feature is its Robotics-as-a-Service (RaaS) offering. Under this structure, customers do not pay a large upfront capital sum for hardware. Instead, they gain access to robotic systems through a subscription or pay-per-use arrangement that bundles the equipment with software, maintenance, and support. Dwbrobot’s stated goal is to let companies deploy automation with minimal financial risk, optimising cost structures while also improving productivity and workplace safety. The company has also extended the concept with a variant it calls RaaStp, or Robotics as a Service to people, which incorporates elements of the sharing economy. The specifics of that extended model — how sharing is structured, who participates, and what the economics look like — were not detailed in the announcement.
Across the Atlantic, Figure AI has been making headlines for a different reason. The company’s humanoid robot was seen accompanying First Lady Melania Trump into the East Room of the White House last month, a moment that was described as likely the first time a humanoid has walked those halls. But the more consequential development for the industry may be the business model underneath. Figure charges roughly $1,000 per month per robot under a “Robot-as-a-Service” subscription that covers hardware, software updates, and maintenance. That price point is notable not because it is cheap — for a fleet of dozens or hundreds of units, the monthly costs add up — but because it represents a clear, publicised attempt to make humanoid robots accessible without a capital purchase.
The broader context is also worth noting. By the end of 2025, more than 140 humanoid robot manufacturers had collectively launched over 330 different models. That is a crowded field, and it suggests that the technology is maturing to the point where differentiation will increasingly come from service models and deployment flexibility rather than hardware specs alone. In China, the strategic importance of this category has been elevated further. Premier Li Qiang’s 2026 Government Work Report included “embodied intelligence” as a strategic national priority for the first time, placing humanoid robots alongside quantum computing and 6G in the country’s 15th Five-Year Plan. That is a policy signal with real consequences for supply chains, standards, and export dynamics.
There is also a less obvious but telling data point from outside the robotics sector. Starlink, the satellite internet service, has shifted its hardware pricing model. It now shows an upfront hardware cost of $0 and a monthly kit fee of $10, a departure from its previous practice of selling hardware for a one-time charge. The monthly fee is in addition to service prices, which were recently raised by $5 to $10 per month. Starlink also offers professional installation for a one-time fee of $199. The relevance to robotics may not be immediate, but the pattern is familiar: hardware becomes a service, capital expenditure becomes operational expenditure, and the vendor retains ownership of the physical asset while charging for access and upkeep.
Why it matters for European robot service
For European service operators, the RaaS shift is not a theoretical discussion. It changes the fundamental economics of automation adoption in a region where capital budgets are often constrained and where the justification process for new equipment can be lengthy and politically fraught within organisations.
The traditional model — buy a robot, integrate it, maintain it, and hope it pays for itself over a five- or seven-year horizon — requires a significant upfront commitment. That commitment includes not just the purchase price but also the cost of installation, training, spare parts inventory, and the internal expertise needed to keep the system running. For many small and mid-sized operators, that barrier has been prohibitive. RaaS removes the largest hurdle by converting a large capital expense into a predictable operating expense. Instead of asking for a budget approval for a six-figure robot system, a manager can approve a monthly fee that is easier to model against labour savings or throughput gains.
The subscription model also changes the risk profile. If a robot underperforms, or if the operational requirements change, the customer is not stuck with a depreciating asset. The vendor retains the responsibility for uptime, software updates, and maintenance — at least in theory. That is a meaningful shift in accountability. In a traditional sale, once the equipment is handed over, the buyer owns the risk of failure. In a service model, the vendor’s revenue depends on the robot actually working, which creates a structural incentive for the provider to keep the system operational.
There is also a fleet-scaling argument. With RaaS, a company can start with one or two units, prove the business case, and then expand. The marginal cost of adding a third or fourth unit is simply the monthly fee. This is particularly relevant for European operators with seasonal demand or fluctuating order volumes. Pay-as-you-go options reduce the barrier to automation because the cost structure can flex with actual usage. The source material notes that shared user facilities run by third-party logistics providers (3PLs) make it easier for organisations to justify new investments and recover costs of pay-as-you-go options. That suggests the model is not just for individual companies but also for shared infrastructure where multiple users can access automation without any single user bearing the full cost.
The 3PL angle is important for Europe, where the logistics sector is fragmented and many operators rely on external partners for warehousing and fulfilment. The source material highlights that operators are beginning to recognise the need to award 3PLs longer-term contracts, allowing them to make the substantial infrastructure investments required to deliver futureproof solutions. In other words, the RaaS model and the 3PL model are complementary. A 3PL that commits to a five-year contract with a customer can justify investing in automation infrastructure, and that infrastructure can then be offered to multiple customers on a pay-as-you-go basis. This creates a virtuous cycle: longer contracts enable infrastructure investment, and shared infrastructure lowers the barrier for smaller users.
For larger organisations with established infrastructure and strong capital investment capacity, the source material suggests a blended approach is more effective. Rather than relying on a single automation model, these organisations can spread risk across a mix of automation models within their distribution network. Some sites might use RaaS for flexibility, while others might use traditional purchases for core, high-utilisation operations. This blended strategy acknowledges that RaaS is not universally superior; it is a tool that works best in certain contexts.
The strategic elevation of humanoid robotics in China’s Five-Year Plan also has implications for Europe. If China is prioritising embodied intelligence as a national strategic goal, it is reasonable to expect accelerated development, lower production costs, and potentially more aggressive pricing in export markets. European operators may benefit from a wider range of options at lower price points, but they may also face questions about supply chain resilience, data sovereignty, and the long-term viability of vendors that are dependent on state support. The source material does not provide details on these risks, so they should be flagged as open questions rather than established facts.
What buyers and operators should know
The RaaS model is attractive, but it is not free money. The trade-off is a higher total cost over long horizons. A subscription that bundles hardware, software, and maintenance will, over several years, likely cost more than an outright purchase followed by a third-party maintenance contract. The vendor is taking on risk and providing services, and that has a price. Buyers should model the total cost of ownership over the expected life of the deployment, not just the monthly fee.
The second trade-off is dependence on the vendor for uptime. In a traditional purchase, the buyer can shop around for maintenance providers, keep spare parts in stock, and control the service schedule. In a RaaS model, the vendor controls the maintenance. If the vendor has a slow response time or a poor spare-parts supply chain, the customer’s operations suffer. The source material does not disclose any specific service-level agreement (SLA) numbers, response times, or spare-part lead times for any of the providers mentioned. That is a significant gap. Buyers should demand these details in writing before signing. If a vendor cannot commit to specific response times and uptime guarantees, that is a red flag.
Another consideration is the contract term. RaaS is not a month-to-month rental in most cases. Vendors need to recover their hardware costs over a defined period, so contracts are likely to run for multiple years. Buyers should understand the exit terms. What happens if the robot does not perform as expected? Is there a trial period? Can the contract be terminated early, and at what cost? The source material does not address these questions, so they remain open items for negotiation.
The Dwbrobot announcement mentions that its RaaS model enables companies to deploy robotic systems with minimal financial risk. That is a marketing claim, not a guarantee. The actual risk depends on the contract terms, the vendor’s financial stability, and the performance of the robot in the specific application. A robot that works well in a showcase video may not perform to the same standard on a dusty factory floor with variable lighting and unpredictable human behaviour.
The Figure AI pricing of roughly $1,000 per month per robot is a useful benchmark, but it is not a universal price. Humanoid robots are a different category from fixed industrial arms or mobile robots. The price will vary based on the robot’s capabilities, the software included, and the level of support. Buyers should not assume that $1,000 per month is the market rate for all RaaS offerings. It is a data point for one company’s product at one point in time.
The Starlink hardware rental model is a useful analogy for what is happening across the hardware-as-a-service space. Starlink has moved from selling hardware to renting it, with a $10 monthly kit fee and a $199 professional installation option. This shift normalises the idea that hardware can be rented rather than owned. For robotics buyers, this is a cultural change as much as a financial one. Many procurement departments are used to capital purchases. Shifting to an operational expense model requires changes in budgeting processes, accounting treatment, and internal approval workflows.
The source material also notes that the country has over 140 humanoid robot manufacturers that collectively launched more than 330 different models by the end of 2025. That is a crowded market, and it suggests that consolidation is likely. Some of these manufacturers will not survive. Buyers who sign long-term RaaS contracts with a vendor that later goes out of business face the risk of stranded assets — robots that stop working because the vendor is no longer there to maintain them. This is a critical due diligence point. Buyers should investigate the vendor’s financial health, funding runway, and customer base before committing to a multi-year contract.
The Ukraine deployment of two Phantom MK-1 humanoid robots by Foundation, a San Francisco startup, for frontline reconnaissance in February is described as believed to be the first humanoid deployment to any combat theater. This is a striking data point, but its relevance to European service operators is indirect. It does, however, underscore that humanoid robots are moving from laboratory demonstrations to real-world deployments in challenging environments. If robots can operate in a combat zone, they can likely operate in a warehouse. But the operational requirements are different, and the source material does not provide any performance data from that deployment.
For European buyers, the practical advice is to model the economics before signing. Compare the total cost of a RaaS contract over the expected life of the deployment against the cost of an outright purchase plus a separate maintenance contract. Factor in the cost of capital, the expected utilisation rate, and the potential for downtime. Consider the vendor’s track record and financial stability. Ask for specific SLA commitments in writing. Understand the exit terms and the consequences of early termination. And do not assume that a low monthly fee is the whole story — read the fine print for additional charges, such as installation fees, training fees, or overage charges for excessive usage.
The blended approach mentioned in the source material is worth taking seriously. For larger organisations, a mix of ownership and subscription models may be the most effective way to manage risk. Core, high-utilisation assets might be purchased outright. Flexible, lower-utilisation assets might be subscribed on a pay-as-you-go basis. This approach allows an organisation to optimise its cost structure while maintaining the flexibility to scale up or down as demand changes.
Finally, buyers should recognise that the RaaS market is still young. The source material does not provide data on the total market size, growth rates, or customer satisfaction levels. The information available is largely anecdotal and promotional. That does not mean the model is flawed; it means that buyers should do their own due diligence and not rely on vendor claims alone. The shift from capital expenditure to operational expenditure is real, and it is likely to accelerate. But the details — contract terms, SLA commitments, vendor viability, and total cost — are where the value will be won or lost.
Published by Vigla Media OÜ (Estonia).