The Industry Counts Every Robot It Ships. It Cannot Count the Workers Who Deploy Them.

The robotics industry can tell you exactly how many robots shipped last year — it cannot tell you how many qualified workers exist to deploy them.

That asymmetry is the story. A robot that arrives at a factory from FANUC or ABB or Yaskawa is not yet a working automation system. It requires mechanical mounting, a gripper or welding tool at the end of the arm, the programming that tells it what to do, safety systems, connection to the factory's existing controller, and operator training. That work falls to system integrators — and nobody has a reliable count of them. The Robot Report, a trade publication that has spent fourteen years trying to map the integrator layer, has documented fewer than 450 integrator profiles globally and describes the real landscape as far larger, scattered across thousands of small firms, most employing fewer than 50 people, operating regionally, specializing in specific industries or applications.

The International Federation of Robotics, an industry data group, reported in September that 542,000 industrial robots were installed globally in 2024 — the fourth consecutive year above the half-million mark, according to IFR data. The figure is precise and widely cited. It describes one end of a supply chain. The other end — the people who make the robots actually work — remains unmeasured.

A new version of the VDA 5050 standard, an open interface protocol meant to let robots from different manufacturers operate under a single fleet controller, was published in March 2026 by the German Association of the Automotive Industry and the VDMA Materials Handling and Intralogistics Association, with technical supervision from the Karlsruhe Institute of Technology. Version 3.0 expands the standard to cover mobile robots with higher autonomy levels, according to automationspraxis.industrie.de. It is open source and available on GitHub. BMW has made earlier versions of VDA 5050 a mandatory requirement for its automated guided vehicle integrations, according to a press release from the German Association of the Automotive Industry.

The standard is a signal of where the industry wants to go: fewer custom software bridges between robots, more plug-and-play fleet management. But the deployment reality lags the ambition. Most factory floors run robots from multiple manufacturers, each with its own control system, bridged by integrators who have built custom connections between them. "Integrators need to unify the fragmented automation landscape," ABI Research noted in a December 2025 analysis of warehouse automation. The robotics system integration services market is, according to Fortune Business Insights, "moderately fragmented, such that there is a large number of regional and specialized system integrators."

The integrator layer represents both a dependency and a strategic puzzle for robot makers. Every major OEM maintains a certified partner program for integrators, but these lists capture only firms that have sought certification and met vendor-specific requirements. The Robot Report's count of 4,296 integrators across 64 countries is a partial picture — it captures the firms it has found, not all the firms that exist. A robot maker expanding into Southeast Asia or Eastern Europe cannot call a single number and get a verified list of qualified local integrators. They ask their existing partners, who ask their partners.

For robot makers, the question of who deploys their machines is becoming a genuine constraint on growth. Robot installations are projected to grow 6 percent to 575,000 units in 2025, with the 700,000-unit mark expected by 2028, according to IFR data. If the integrator layer cannot scale proportionally, those numbers represent backlogs, not deployments.

The geographic breakdown from the IFR data shows where the hardware is flowing and why deployment capacity matters: China installed 295,000 robots in 2024, more than the rest of the world combined, and for the first time, Chinese manufacturers held a majority of their domestic market at 57 percent. Asia accounted for 74 percent of all new global deployments. Europe installed 85,000 robots, down 8 percent from the prior year. The Americas installed 50,100, also down. These declines occurred even as global robot sales hit new records — which suggests the supply of robot hardware is not the binding constraint on automation adoption.

The 542,000 figure will continue to be cited. It belongs in any accounting of industrial automation. But it describes one end of a supply chain whose other end — the people who actually make the robots useful — remains unmeasured and, by most accounts, under pressure. The machines are getting cheaper and more capable. The humans who deploy them are not being counted.