And I'm okay with that. Having transparent pricing means that I am voting with my dollars for the content I care about. With a cable subscription, most of your dollars just go to Disney and not the content producers whose content you are personally consuming.
How are Netflix, Prime, and even HBO different? You pay them for access to all of the content they decide to carry. I don't see how Netflix is "a la carte."
You pay them for access to everything, but they know what you are watching -- this is quite unlike typical broadcast/cable where only some very small subset (historically "Nielsen Families") are tracked.
This is not accurate. There are a couple of competitors to Nielsen, but more importantly, most cable providers these days track what everyone is watching. They may not do a ton with that data (depends on the company), but it's not like they don't have the data.
It still isn't quite the same for a number of reasons, including:
1) the broadcaster is a step removed from the cable operator (though with all the mergers in the content/delivery space, they might be part of the same parent company). If Netflix develops a new show like House of Cards, they know exactly how many people watched it for how long, etc.
A cable company might track what station their customers are watching but unless they are reporting that back to the broadcaster it won't change programming in the near-term. It may make the cable company make different decisions about their lineup but such decisions are often part of many-year contracts with forced bundling (eg. Disney saying you must carry all these stations if you want ESPN) so even if they do have data I doubt it has much impact.
2) The cable company can know what channel your settop box is tuned to, but a lot of people have setups where the cable box is essentially always on (and only the TV is switched off) and there is no reasonable way for the cable operator to know if anyone is actively watching or not without resorting to very imprecise heuristics like only counting it if the channel has recently been changed.
This is unlike Netflix, et al, which have less passive interfaces where you can be more reasonably sure that if the user started some content they are actually watching it and not streaming it to a TV set that is off.
1) There's obviously a lot more separation here, but it's worth noting that there is a lot of data out there available to broadcasters that doesn't come from Nielsen, such as broadcasters licensing data that comes initially from cable providers. The number of set top boxes here dwarfs the size of the Nielsen data set.
2) You can get more precise if you want. At a basic level, you can look at a lot more options than just changing the channel. You can also train models based on the overlap with something like Nielsen where you have more information about when they are watching. If you can match up comcast data to Nielsen data, you can build a model for active vs. inactive boxes on the comcast data.
3) To backtrack a bit, the fact that Nielsen is a "very small subset" as you said a couple of comments ago is relevant for some things but not others. For the broadcast network prime time shows, it's certainly a big enough sample to see which shows are popular among what demographics without having the level of detailed information that Netflix or cable operators have.
All in all, I agree with you that there is more separation and it is harder to use, but there's still a key distinction between Nielsen, which is in something like <~0.1% of homes and the other data sources which are many times larger. They don't all seek it out, but it's possible for broadcasters to acquire the sort of data necessary to answer questions about content popularity like Netflix can consider.
