I've got a substantial commute to my day job and one of the podcasts I listen to (usually on the way home because in the morning I'm concerned I'm too tired to listen carefully) is the Sell More Books Show. The podcast's two hosts Bryan Cohen and Jim Kukral have provided lots of useful advice on how independent authors can sell more books. Much of their advice has not been applicable to me--I've only got a few short stories and one collection thereof--but some is, and the time may come where I self-publish something more substantial.
In any event, I was listening to Episode 96 on the way home today and they discussed the concept of "the long tail." It's basically the idea that one's earnings are more lucrative in the long run than in the opening burst of sales, especially if one has an extensive backlist. Here's one article explaining how Amazon, e-publishing, etc made this possible. Here's another, the first in a series explaining the process. This article here points out that the theory may not be all it's cracked up to be, but it's something I've noticed in my own sales.
I published my rather gruesome horror tale "I am the Wendigo" in 2012 alongside the less gruesome (but funnier) "Melon Heads." Both stories serve as examples of "the long tail" in action. "Melon Heads" started with a bang--16 copies in the first month--and sold seven copies in 2013, six in 2014, and four copies in 2015. So far I've sold one copy in January and one in February. That's 19 sales so far, with three years, one month, and two weeks' earnings nearly equal to the opening explosion. Meanwhile, "Wendigo" has proven a bit more jagged--14 in 2012, 23 in 2013, 14 in 2014, 19 in 2015, and three so far in 2016. Although it's not the classic "long tail," it's still a case of later sales exceeding the premiere. And this can go on and on and on for years...
So long as Amazon stays running, of course. But I don't anticipate it going anywhere.