Well, the methodology is flawed, or at least it doesn't actually reflect what it says on the tin. It's "consumer-initiated", i.e. what they are really measurin

Well, the methodology is flawed, or at least it doesn't actually reflect what it says on the tin. It's "consumer-initiated", i.e. what they are really measuring is which contracts people are using, not which speeds are available. Example: I could book a 500/100 line at my house (1000 symmetrical available if I switched providers), but I stick with the cheapest plan, which is 50/12 simply because it's good enough for my personal use. Same for most of my neighbours. Most don't have more than 100/20, though higher speeds are readily available. With StarLink there's only type of contract: you get what you get and that's that. In Europe for example, there's usually more than one option per ISP and you pay more for higher bandwidth. So if you're fine with the smallest plan (usually around 50mbps), why pay more? So in essence they actually answer a different question.