I know that many people have had a lot of success in Select, or attribute part of their success to it. However, if you look at it from a reader's
point of view, there's a certain logic to the program that begins to seem alarming.
First of all, free days. One of the "perks" for writers is that each enrolled book can be priced at $0 for five days during the enrollment period. From the writer's point of view, it means a chance to artificially boost the book's rankings, thus putting it in front of more eyeballs and potentially getting more readers.
However, from a reader's
point of view, this means that any book labeled "KDP Select" is going to be free at some point in the next month or two. If they're willing to wait, all they have to do is track it--not a difficult task, considering how eager most of us are to get our books listed on the free ebook sites.
If the reader is signed up for Select, they can borrow the book, and the author gets paid. However, if the reader is not signed up for Select, they have little or no incentive to actually buy the book, since they know they can get it for free.
Thus, readers come to see KDP Select as a revolving bargain bin for authors who don't value their own work.
Since newer writers who have yet to build a fanbase are the ones who benefit most from the Select program (or in other words, the ones who stand to lose the least), they make up a majority of Select authors. However, they're also the ones who have yet to master their craft. Thus, over time, readers come to associate Select with cheap writing. This only solidifies the bargain bin stigma mentioned above.
If you follow the logic, it's hard not to conclude that authors who sign up for select may actually lose a fair amount of sales that they would've otherwise had. This is because, according to the logic, readers who aren't signed up for Select 1) know they can get it for free if they hold out long enough, and 2) associate Select books with lower quality.
An interesting comment
I recently read on a blog post about Amazon's review policy seems to confirm as much: