11–20 of 145 posts
Someone should try selling a book that there are limited copies of (ie one that can't be procured easily) to see what happens if someone buys from a seller who doesn't have it.
Now, if you'll excuse me I'm off to weep over my broken dreams of belated celebrity.
These things are /wonderful/ when it comes to making the market more efficient. Really, though, the shipping costs eat up most of the efficiency. Amazon needs an easy way to say "I want these 10 books, used. Find me the lowest price (including shipping)" - the idea is that the more books you could buy from one seller, the less shipping friction would be involved, but amazon isn't really set up that way, which makes it much less efficient for the low end used books.
Earlier quoted context omitted.
Any 3rd party tools available to 3rd party sellers? Offering something that beats the current algorithms could be a good deal (if its not illegal).
I have code that does ecommerce price point optimization using machine learning. In an ideal setting, it combines a variety of signals, including prediction using previous sales data. It's also able to "smooth" over different products, so that you can make good price point choices even for products that have very little sales history.
Edit: I'm puzzled by the downvotes to my honest question. "machine learning" is not an algorithm, it's a field!
oh man. reminds me of the days when I sold books on amazon and half.com. I wrote a script that took the 'nickel less than the other guy' approach. These things are /wonderful/ when it comes to making the market more efficient. Really, though, the shipping costs eat up most of the efficiency. Amazon needs an easy way to say "I want these 10 books, used. Find me the lowest price (including shipping)" - the idea is that…
Someone should try selling a book that there are limited copies of (ie one that can't be procured easily) to see what happens if someone buys from a seller who doesn't have it.
(For an example, read up on Porsche & Volkswagen: eg. http://www.nytimes.com/2008/10/30/business/worldbusiness/30i... )
oh man. reminds me of the days when I sold books on amazon and half.com. I wrote a script that took the 'nickel less than the other guy' approach. These things are /wonderful/ when it comes to making the market more efficient. Really, though, the shipping costs eat up most of the efficiency. Amazon needs an easy way to say "I want these 10 books, used. Find me the lowest price (including shipping)" - the idea is that…
Everyone always mentions that buying more items from a single Amazon store would make shipping cheaper? Is that ever true? Of all the online stores I've made purchases from, I believe monoprice is the only one that appears to base shipping on tiers based on item size/mass. All the rest simply charge a set rate for each item, and it doesn't matter how you divide your shipments up into different orders.
Earlier quoted context omitted.
I have code that does ecommerce price point optimization using machine learning. In an ideal setting, it combines a variety of signals, including prediction using previous sales data. It's also able to "smooth" over different products, so that you can make good price point choices even for products that have very little sales history.
What's the algorithm? Edit: I'm puzzled by the downvotes to my honest question. "machine learning" is not an algorithm, it's a field!