I always make this argument, and it usually falls on deaf ears. But, this is a cultural and industry-norm-driven pattern — and Netflix is in prime position to upset industry norms (as it’s already done by releasing whole seasons at once).
In the anime industry, outside of a handful of long-running high-profit and usually every episodic shows aimed at younger audiences, there’s a very different attitude toward extending series. In the US, we tend to make new seasons of a show until it gets so bad that it gets cancelled; in Japan, because cancellation is such a big deal (it’s so rare that I’ve only ever seen one cancelled show), you have a very different pattern: a single season is planned out and made, and if it’s sufficiently popular and the source material allows for it, a decision is made about whether or not to create a sequel, usually several years later. The average show is between 12 and 26 episodes long; most shows have exactly one season, and when new seasons appear, whether or not they are a continuation of the same story depends heavily upon whether or not the show is an adaptation of a continuing story in another medium. As a result, a season is like a miniseries: good shows are finely crafted single-season stories that are never revisited, and a show with more than three seasons is almost always considered mediocre at best.
If Netflix adopted this style, they could push a change in the industry. After all, television has already changed: serious character-driven stories with complex ongoing narratives have become valued only in the past fifteen years or so, displacing the normal highly-episodic series well-adapted to syndication by channels that only license a small portion. If the expected form of television changed in this way, we’d see fewer shows that follow the pattern of Buffy — you know, an excellent first season followed by increasingly mediocre and convoluted follow-up seasons designed to pander to an obsessive core fanbase.