3 ms·
You can think of Netflix as following a third generation of ratings. The first generation of ratings was just raw data: how many TV’s were tuned into a given s
by philwelch 8y ago
You can think of Netflix as following a third generation of ratings.
The first generation of ratings was just raw data: how many TV’s were tuned into a given show at a given time? More viewers, more ad revenue. Simple as that.
The problem was that some viewers are more lucrative than others. The second generation of ratings added demographic data. This happened sometime in the late 60’s/early 70’s, just in time to discover that the recently-cancelled Star Trek, despite its low ratings, appealed to a very lucrative demographic.
Netflix has user-by-user, viewer-by-viewer ratings. Not just statistics per the old Nielsen sample, but the entire population of data. Maybe even minute-by-minute (unless you doze off). This is way more powerful than demographics, because instead of targeting shows to “18-35 middle class white urban women”, for example, you can target shows to “people who binge watch space operas” or “people who enjoyed only the first two seasons of <<some show>>” or any other large enough group you can tease out of there. And Netflix might only be scratching the surface of what is possible with this data.
- mmt 8y ago> Netflix has user-by-user, viewer-by-viewer ratings. (Minor nit: They don't necessarily know how many viewers are watching a screen) But maybe that's because they don't care, since, does it really matter? If they're not running ads, isn't every unit of screen-time worth essentially the same amount, ignoring the variance in subscription prices for different regions and resolution options. Actually, I'd go so far as to argue that viewer time is worth, incrementally, a negative amount. I'd expect their goal is to have just enough content to keep subscribers interested, rather than to maximize viewing time. As such, it seems to me that their viewer data goes beyond being a different generation of ratings, into a completely different thing. Maybe it's closer to box-office (seats, not revenue) statistics for movies. Now, I know they're not actually ad-free, what with product placement, in which case traditional "ratings" can translate to real money, but I'm not sure if that's a large enough proportion of revenue (or reduction in content production cost) to change their behavior.
- philwelch 8y ago> Actually, I'd go so far as to argue that viewer time is worth, incrementally, a negative amount. I'd expect their goal is to have just enough content to keep subscribers interested, rather than to maximize viewing time. That might be going a LITTLE far. Netflix subscribers who don't actually watch Netflix not only don't give Netflix enough data to make programming decisions from, but I'm willing to bet that an engaged Netflix subscriber is much, much, much less likely to cancel than someone who barely ever watches Netflix. If you think of viewer time as inversely correlated with probability of cancelling, the obvious metric to optimize for is viewer time. You're never going to watch so much Netflix that the marginal infrastructure cost is meaningful to them.
- mmt 8y agoKeep in mind, though, that I mentioned incremental value, though maybe what I meant was marginal? I do agree that NF's business model can't really be compared to that of a gym, with a goal of maximizing subscribers that (eventually) don't use the service at all. One of the big mysteries is how good a deal did they got for that outsourced delivery infrastructure, as well as how much of it is metered (and how) versus flat-rate.