That article is a riot. Also disturbing and depressing. And confusing, but not because of the author, but because of Amazon.
I get that balancing out the different factors that play into the likelihood of buying it again is difficult, but with all the data they gather, it really shouldn’t be impossible to figure out that the mode of “purchases per customer” for a given product cluster is 1 or that a frequent recommendation has not actually gotten any interest and use those for weighing the random recommendation selection.
Even within some product group you buy more frequently, there might be certain brands / vendors you prefer, so buying something once to try (particularly if it’s new) and never again should also reduce the likelihood.
But I’m just a Junior Data Analyst, what do I know.
This article about a person who bought an urn online is hilarious. Amazon thinks I’m a serial killer now (cracked dot com)
That article is a riot. Also disturbing and depressing. And confusing, but not because of the author, but because of Amazon.
I get that balancing out the different factors that play into the likelihood of buying it again is difficult, but with all the data they gather, it really shouldn’t be impossible to figure out that the mode of “purchases per customer” for a given product cluster is 1 or that a frequent recommendation has not actually gotten any interest and use those for weighing the random recommendation selection.
Even within some product group you buy more frequently, there might be certain brands / vendors you prefer, so buying something once to try (particularly if it’s new) and never again should also reduce the likelihood.
But I’m just a Junior Data Analyst, what do I know.