Follow the business model and the patent makes sense. Building a satellite to fly one instrument is expensive and slow. Loft Orbital's pitch is that many customers do not want a satellite, they want their payload in orbit with power, pointing, and a data pipe. Sell that as a service and you change who can afford to operate in space.
The grant US10981678B2 (inventors Pierre-Damien Vaujour and Lucas Bremond), classified in B64G 1/1085 (constellation and platform arrangements) with B64G 1/242 (attitude control) and H04B 7/18521, claims the system that makes hosting standardized: a spacecraft that provides services, power, attitude control, communications, to payloads it did not itself design.
The mechanism is abstraction. Just as cloud computing hid the server behind an API, this filing hides the satellite bus behind a service interface, so a payload provider can treat orbit as infrastructure rather than as a bespoke engineering project. The standardization is the product.
For a private company that files little with the SEC, the patent is one of the clearer public traces of the strategy. It tells you Loft Orbital considered the service-provision architecture, not any single satellite, to be the defensible asset, which is exactly what you would expect from an infrastructure-as-a-service play.
The caution is that a business-model-shaped patent describes intent and architecture, not market traction. It does not tell you how many payloads flew or what the unit economics were. It tells you what Loft believed was worth protecting: the interface between a customer's instrument and a shared ride to orbit.