Why it matters
Reflect Orbital, a Hawthorne, California company led by co-founder and CEO Ben Nowack, wants to sell sunlight after dark. Satellites would unfurl mirrors in low Earth orbit and steer reflected sunlight onto a chosen spot, so a customer could order light without new ground equipment. The homepage markets that spot as highly localized, dimmable from moonlight toward noon once more satellites are flying, and usable for solar farms, disaster response, industry, agriculture, defense, and events. The service is not operating.
On November 10, 2026 the company said it expects its FCC-approved inaugural demonstration, a mission to collect data on how an orbital reflector deploys and performs, in the second quarter of 2027. That flight is a test. The same post said the new funding supports initial launches, further reflector development, and commercial partnerships in energy and lighting.
The launch plan and the homepage disagree
The homepage still publishes two near-term fleets. One block reads 2 satellites in 2026, 36 in 2027, more than 1,000 by 2028, more than 5,000 by 2030, and more than 50,000 by 2035. A later block on the same page lists 38 satellites for 2027. Those cells are company targets, and the 2026 pair is already behind the company's own later posts.
On October 31, 2026 Reflect said it had filed on October 30 for Earendil-0, a new developmental satellite with no orbital reflector, to fly as a commercial rideshare passenger no earlier than April 2027. The post calls Earendil-0 the first Reflect spacecraft in orbit and says it will test control and communications ahead of Earendil-1, which the FCC licensed earlier in 2026. The November 10 post still describes that developmental satellite as subject to regulatory approval, then says the already approved reflector demonstration is expected in the second quarter of 2027. It does not print the name Earendil-1 or a rocket. It does say the company has other 2027 bookings and has purchased two dedicated Skyroot launches. On October 9, FCC order DA 26-1101 still described Earendil-1 as scheduled within several months of the July 9 grant. That sentence and the November demonstration date are different schedules.
What the first mirror is designed to do
Science News reported on September 17, 2026 that Earendil-1 is an 18-by-18-meter reflective sheet, folded for launch and unfurled in orbit at about 600 to 650 kilometers. At that size the intended spot is about five kilometers wide, about as bright as a full moon overhead, and about four minutes long, because the satellite moves at roughly 7.5 kilometers per second. The homepage still lists the 2026 lighting specification as 0.1 lux for five minutes, which it compares to a full moon. None of this has been shown from orbit. Later homepage steps, including 2 lux for 2.5 hours in 2027 and daylight-level bursts in the 2030s, assume many more satellites. Science News reported that the company wants those later mirrors to be about 54 meters across.
The license on the books is for one satellite. DA 26-1101 reissued the Earendil-1 grant, call sign S00711, file SAT-LOA-20250701-00129. It authorizes an insertion orbit near 510 kilometers and operations near 625 kilometers at about 88 degrees inclination, and it authorizes the radio links used to deploy and test a steerable sunlight reflector. It is not a license for the 36-satellite step or the 50,000-satellite roadmap.
How the company says it would get paid
Reflect describes two uses, and only one of them is supposed to justify a huge fleet. Lighting is the early product: a spot of sunlight for a place and a time. Energy is the stated reason to fly thousands of mirrors. The homepage keeps the energy service in testing through 2028. Around 2030 it projects about a 1 percent capacity-factor gain from 50 watts per square meter for 20 minutes, with a footnote that a typically overbuilt farm might see about 0.75 percent in Southern California and about 1.4 percent in Germany. By 2035 it projects a 20 percent capacity-factor gain from 300 watts per square meter for three hours. Those are company estimates. The homepage and the November funding post do not name a signed solar customer or a measured trial.
The November 10 round is aimed at the step before revenue: first launches, reflector development, and the start of commercial partnerships. Until a satellite flies and a customer pays, both lines are targets.
Funding and the factory claim
On November 10, 2026 Reflect Orbital said it raised $73.5 million, taking total funding above $100 million. The post gives this round no series label. It says the money follows the Series A in May 2025. Named participants are Sequoia Capital, Lux Capital, and Starship Ventures, plus new investors including Buckley Ventures, Convective Capital, The Uncommon Group, and Salesforce Ventures. The post adds "among others" and does not publish a valuation.
The same announcement describes a production rhythm: one large reflector quadrant every 3.5 days, a membrane the company says is 28 times thinner than a human hair and folds to a shoebox; one four-boom structure each week; and a 12,000-part satellite every two months. Those are company statements about the factory, not audited deliveries, and they sit next to a flight plan that still has no spacecraft in orbit.
Regulation and the night sky
The constraint on a larger fleet is optical as well as financial. Science News reported that European Southern Observatory astronomer Olivier Hainaut found a full Reflect fleet could make the night sky three to four times brighter within a few hundred kilometers of a beam, and that the operational mirrors would be about 54 meters across rather than the 18-meter demonstrator. Reflect disputes those findings. Its homepage says the light stays in a defined spot, can be turned off, is not bright enough to start fires or harm eyes, and that exclusion zones will keep it away from observatories and sensitive environments. There is no on-orbit measurement yet. The September reporting also said the company expected the first demonstration to be the first of several test missions.
The license challenge is unfinished. DA 26-1101, released October 9, 2026, left pending an August 6 Application for Review filed by Earthjustice for DarkSky International, the American Bird Conservancy, Environment America, and Public Employees for Environmental Responsibility, and an August 7 petition for reconsideration filed by an individual. The order moves the Earendil-1 surety bond to January 5, 2027 and does not decide those filings. This pass did not find a later Commission order closing them.
What to watch
Three dates now matter more than the homepage's 2026 pair. Earendil-0 is a filed application for a reflector-free rideshare no earlier than April 2027, and the November 10 post still calls that developmental satellite subject to regulatory approval. The FCC-approved reflector demonstration is company-timed to the second quarter of 2027, with no rocket named in that post. The Earendil-1 bond is due January 5, 2027, and the August challenges were still pending when the Bureau reissued the license on October 9.
The result that would make the energy business more than a roadmap cell is a named solar customer and a measured pass. The homepage still marks energy service as testing until the 2030 claim of about a 1 percent capacity-factor effect, and it still shows two satellites in 2026. Two dedicated Skyroot launches are purchased, with no dates given on November 10. The next public proof points are an Earendil-0 grant, deployment of the first reflector, and any full-Commission action on the August 6 review.