While there is a lot of discussion about what SpaceX Starship is doing on the next test flight, there still is not enough about how radically Starship will change the face of space once it is fully operational.
It will impact everything in space. Yes, everything. With operational flights just beginning, we are truly on the cusp of the Starship Revolution.
This is an update of a piece first posted in April 2025. The thesis has not changed. The calendar has.
Starship is no longer a paper rocket. Block 3 has flown. Flight 12 lifted off on 22 May 2026. Flight 13 followed on 24 July 2026, deployed 20 Starlink V3 satellites on a suborbital arc, relit a Raptor in space, and put the ship down in the Indian Ocean in one piece.
Flight 14 is the one that changed the sentence. On 28 September 2026, Booster 21 and Ship 41 left Pad 2 at Starbase at 7:48 a.m. CT. One ship engine shut down early on the climb. The team still lit a single Raptor for the insertion burn, and Starship reached orbit for the first time, at about 275 km. It then deployed all 26 Starlink V3 satellites. SpaceX made contact with every one. That is the first operational payload Starship has left in orbit.
The rest of the plan was cut. Six orbits and nearly 10 hours, ending west of Chile, became about 3 hours and 9 minutes, ending north of Hawaii. The ship did its first deorbit burn, came through entry, flipped, and splashed down on target. The booster also lost an engine on the way up, finished its landing burn, and came down in the Gulf. Better than Flight 13. Still not a catch.
What it has not done yet is the unglamorous stuff the whole architecture sits on. It has not yet transferred propellant ship to ship. It has not yet caught a ship on the tower. It has not yet shown the quoted 100 t reusable payload on an operational flight. Twenty-six V3s is a real delivery. It is not a full manifest.
The destination is still the same. A reusable ship in the 100 t class, flown over and over, changes how much can go to space, what can be built there, and how often people can leave Earth.
Starship still relies on in-orbit refuelling for missions beyond LEO. Once that is in place, the Moon and Mars stop being once-a-decade events. They become a traffic problem.
A new era in space is still coming. Orbit was the gate. Refuelling is the next one.
Moon Mission
The most obvious impact of Starship is still getting boots back on the Moon.
That is no longer a speech. Artemis II flew. Orion launched on 1 April 2026 with Reid Wiseman, Victor Glover, Christina Koch, and Jeremy Hansen, looped the Moon, and splashed down off San Diego on 10 April. Humans have been around the Moon again. They have not landed.
NASA's current plan is more cautious than the 2025 briefing charts. Artemis III is now aimed at mid to late 2027 as a low Earth orbit docking demo with lander test articles. The first crewed landing is aimed at Artemis IV around 2028. Both Starship HLS and Blue Origin's Blue Moon are still in the mix. Neither has done an uncrewed lunar landing yet.
The elephant in the room is still there.
Once a tanker-refuelled Starship can land on the Moon and come back, a lot of the supporting stack looks optional. SLS, Orion, Gateway, and a second lander are not required by physics. They are required by the programme NASA is actually flying.
That distinction matters.
Starship HLS, as NASA has contracted it, is a lander. It is meant to meet Orion, take crew down, act as a surface habitat, and return to lunar orbit. It is not, in that design, the ship that also flies the crew from Earth and home again.
A later Block 3 or Block 4 ship, topped off in LEO, is a different animal. That version can leave LEO, land, and come home. That is the simpler architecture. NASA has not chosen it. They kept Orion, added a 2027 dress rehearsal in LEO, and split the landing work between two companies.
So HLS does not make every other part of Artemis "totally redundant" tomorrow. It does make a cheaper, faster lunar architecture sitting in plain sight. If SpaceX can refuel and land, that simpler path gets harder to ignore.
The landing is no longer "sometime in the 2030s" on paper. It is targeted around 2028. Whether 2028 holds is another question. Hopefully it will.
Mass to Orbit
This is still the section that does the most work.
Block 1 was a test article in the teens of tonnes. Block 2 was quoted around 35 t. Block 3 is quoted around 100 t reusable. Block 4 is talked about around 200 t. Treat those last two as targets. Flight 14 proved the ship can reach orbit and drop a payload that stays there. It did not prove the 100 t number.
Even so, it has been improving since 2024.
In 2024, the non-Starlink mass to orbit was on the order of ~400 t. Two to four full Block 3 flights would have matched that.
In 2025 the picture split in two. SpaceX delivered about 2,413 t of known payload. Starlink was roughly 1,975 t of that. The paid launch market, the tonnes customers actually bought, was about 347 t.
So the old rocket industry is not competing with Starship for "all mass to orbit." It is competing for a few hundred tonnes a year while Starlink already moves thousands on Falcon. Starship's job is to make those thousands cheaper, and then make tens of thousands possible.
That is the radical bit. It is not one big launch. It is many big launches.
SpaceX still talks about ramping production from months to weeks to days, and about flying each ship often once reuse is real. That remains the plan. It is not the present. Fourteen integrated flights since April 2023 is serious progress.
Get reuse and refuelling working and those numbers stop looking silly. Goal now, reality soon.
Space Communications
Starlink has already changed space comms around Earth. Crewed flights have used it. Launch vehicles use it. Suits and rovers are being designed around ordinary 4G/LTE style networks rather than a unique deep-space radio for every object.
Starship is now part of that story, not a future dream. Flight 13 deployed 20 Starlink V3 satellites on a path that brought them back down. Flight 14 put 26 operational V3s into orbit, and SpaceX contacted all of them. V3 is the satellite that only makes sense if Starship is flying often. The first batch is up.
Moonlink and Marslink are still in the future mix. They are not here yet.
A lunar Starlink-class network is being talked about as off-Earth expansion. No dedicated Moon constellation has been deployed. One Block 3 class ship could, in principle, drop enough satellites for a useful first network and come home. A game changer.
Mars is further out. A Marslink ship probably does not come home. It could stay in Mars orbit as a relay, or it could be a test lander. That is a good use for an early cargo flight. Proven space mesh with an Earth link.
Get the network in early and everything after it gets easier. Suits, androids, habitats, and ground teams all speak the same language. That part of the original case still stands. The date does not.
Space Stations
Starship as a station would be a true revolution.
The ISS is still headed for a controlled deorbit around the end of 2030 or early 2031. SpaceX is building the US deorbit vehicle for that job. After that, LEO is supposed to belong to commercial stations.
Vast and Axiom have real space station programmes. Vast's Haven-1 is being built in Long Beach, aimed at Q1 2027 on a Falcon 9, with about 45 m³ of habitable volume. Useful. Small. Haven Demo already flew and deorbited. Axiom is still assembling a future out of modules. China keeps flying its own station.
None of that matches a single Starship hull.
One launch can put up a station with on the order of 1,000 m³ of pressurized volume, in the same class as the ISS. Two launches and you can start talking about a spinning pair and artificial gravity. No months of docking little cans together. The station is usable after arrival.
Early versions can be cousins of the HLS lander. Later versions can be built for the job. Propellant tanks, once in space, can be cleaned and used as water stores. On a gravity station that is how you get a shower that is not a science experiment.
Other stations may still fly, and some will. They will just look modest next to a hull that is already the size of the destination.
It is still a logical step. Nobody has flown it. That is the update.
Mars Missions
This is the chapter that aged the fastest, and the architecture that aged the least.
The original plan was five uncrewed Starships in the 2026 transfer window, more in 2028, and boots around 2029. That opening mission is not happening.
The cheap Earth-Mars window in late 2026 is not being used. Starship has reached orbit once. Tanker-to-ship transfer has not been demonstrated. In early 2026 SpaceX put the near-term focus on the Moon. In July, Elon Musk put people on Mars in roughly 5 to 7 years, with a lander a few years sooner. Take that as Musk-time. It points at the early 2030s, not 2029.
The next honest sequence looks like this. Orbit is done, once, and cut short. Next is a full-duration orbital flight, then a propellant transfer demo, then an uncrewed lunar landing. Then, if the ships and the tankers are real, the next Mars windows around late 2028 / early 2029 and 2030 / 2031.
Still a revolution that only Starship can bring, just with a window shift.
What has not changed is the force multiplier.
Only a reusable heavy ship, built in a factory, makes a thousand-ship window thinkable. Five cargo ships to learn landing. Then 20. Then 100. Then, incrementally, the 1,000-ship cadence. If half of those later ships carried people and half cargo, you get tens of thousands of tonnes and a lot of seats. That is still radical planning. No other rocket is even in close.
Early uncrewed ships can carry Optimus androids to walk the surface, survey ice, and fail cheaply before crews show up. That is a better first image than another empty crater.
It still sounds like a dream. The difference from 2025 is simple. The Moon Mars, but hopefully for the better.
Conclusion
Starship is still the revolution that changes the shape of space. The 100 t class reusable payload still radically effects the old mass-to-orbit math. A tanker-fed ship can still do lunar work without the rest of the Artemis stack. Starlink V3 is already being delivered from the vehicle, not just talked about. An ISS-scale station in one launch is still sitting there unused. A thousand-ship Mars window is still the only colony architecture that is not a pamphlet.
What 2026 added is honesty about the gates.
Block 3 is flying. Orbit has been reached. Refuelling is next. The Moon is the near job. Mars has slipped to the next windows, yet still soon.
Embrace it as it is, not as last year's calendar. The adventure is still out there. 🚀
Note: Cover photo from SpaceX Flight 14 video.