This should be a golden era of launch.

During the last three years, an average of 270 orbital rockets have launched from Earth, a more than three-fold increase from only a decade ago. By every metric available, the launch industry is crushing it: Prices have never been more competitive, launches never more frequent, and access to space never more rapid.

Paradoxically, though, there is a growing crunch in launch availability.

As part of his job as the director of research at Quilty Space, Caleb Henry speaks with satellite operators all the time. And he is hearing a constant and increasingly loud refrain from companies that want to put stuff into space.

“What we have is an industry in panic,” Henry said.

Anecdotes abound. A Canadian satellite company, Telesat, recently received inquiries from other companies asking if they would consider sharing some of the 11 Falcon 9 launches Telesat has booked for its new constellation. (Sorry, no, is the reply). The Amazon LEO constellation has had to throttle back production of satellites because few of the dozens of launches it booked half a decade ago are ready. And during an earnings call last Monday, officials from AST SpaceMobile said launch availability was now the pacing item for deploying its constellation.

With a growing number of constellations big and small, there is a huge and growing demand signal. The Commercial Space Federation predicts demand for thousands of satellite launches annually in less than a decade. Analysys Mason forecasted that more than 37,000 satellites will need to be launched between 2023 and 2033. Henry’s own firm, Quilty Space, has charted a change in the discussion from how many launch companies the US market can support to who can execute the fastest to meet surging satellite needs.

And if things are bad now, there are reasons to believe they may get even worse over the next two to four years.