Buy the data before you buy the satellite
Most organizations that ask about owning a satellite need the result a satellite produces, and the result is usually for sale by itself.
The conversation often opens with hardware: we want our own satellite, for imagery, for connectivity, for presence. The first honest question runs the other way. What would you do with the result, and can that result be bought as a product? Imagery over your sites on a revisit schedule, connectivity for your assets, monitoring feeds: these exist as services, priced like services, with someone else carrying the launch risk.
Owning is a different life. The spacecraft is often the cheapest line on a list that includes the ride to orbit, the ground segment or a contract for someone else's, the licensing, the operations staff or the operations vendor, insurance decisions, and a lawful plan for the end of the asset's life. Each is its own procurement with its own specialists. None of it is impossible, and organizations do it well, but nobody should walk into it because a slide said satellites are cheap now.
There are honest reasons to own. Tasking control, when you need the sensor pointed where you decide, on your schedule, and the commercial constellations cannot promise it. A proprietary sensor that does not exist as a service. A sovereignty or program requirement written by someone above your pay grade. When one of those is real, ownership stops being romantic and becomes an engineering program with a defensible purpose.
The trap between the two is the pilot satellite bought to learn, without a plan for what success would change. Learning is real, but a data subscription teaches most of the same lessons for a fraction of the exposure, and a subscription can be cancelled. A spacecraft cannot be un-launched.
This is a market where we make introductions rather than recommendations, so the useful thing we do is exactly this sorting: buy the result, or fly the hardware. Arrive with that question answered, and every conversation with the real specialists gets shorter and more productive.
Ask what you would do with the result, then try to buy the result. Own the hardware only when tasking control, a unique sensor, or a mandate genuinely requires it.
Put it to the test on your own job
Tell Joe what you are running and he will apply this to your numbers, including telling you when a robot is the wrong answer.
More notes