Amazon Does The Math, Anticipates Your Needs

Originally published on January 22, 2014 8:12 am
Copyright 2018 NPR. To see more, visit


Amazon got a lot of publicity and not a few jokes out of its idea to drop packages on your doorstep with drones. Now the company's quest to cut deliveries to the shortest time humanly possible has taken things to a higher level. Amazon is planning to ship goods your way before you even hit the buy button. You could say it will be reading your mind. In a patent Amazon calls it anticipatory shipping.

Tim Stevens, editor at large for CNET, has been following this story. Thank you for joining us.

TIM STEVENS: Thank you for having me.

MONTAGNE: What are we talking about here? I mean this is sort of the headline. Amazon is going to ship stuff you want before you even know you want it or certainly before you ordered it.

STEVENS: That's right. It may be things that you want. It may be things that you've added to your wish list, but it's not things that you've committed to buying yet, at least not knowingly. Amazon's going to be basically looking at what you're shopping for, things that maybe you've added to your cart but decided not to check out for yet, or even things that you've maybe hovered your mouse cursor over longingly, but decided not to actually buy.

Amazon wants to develop a system that will allow it to actually use those cues as indicators that you're going to buy this thing at some point in the future and then go ahead and ship that thing to you so that when you do decide to buy the package, it will just be either on a truck ready to go to your house or in a warehouse not far from you, so instead of even having to wait two days, you might even be able to get it same-day.

MONTAGNE: Clearly this is not a totally new concept, the idea of purchasing patterns, what people in some parts of the country want or need. How is Amazon fine-tuning this so much that it can ask for a patent on it?

STEVENS: Specifically what they're patenting is the monitoring of you buying purchases and the application of your history as applied to other people in your area with similar interests to you, and then the other part of it is actually the logistics of getting those packages to these shipping centers and getting them closer to you.

So it's basically the whole package describing the algorithms for finding out what you're going to buy and then finding out how to figure out the best way to get that close to you in a way that's not going to cost Amazon a whole lot of money, of course. In fact, they describe a scenario in which they ship a package to you that you decide you don't want after all and they may actually just give that package to you for free because it might be cheaper than shipping back to Amazon.

And they consider that an act of good will.

MONTAGNE: So what is the patent for exactly?

STEVENS: The primary focus is definitely the algorithms where they can rank and score the different indicators of your behavior and then once that score crosses a given threshold, ship that package to you before you even know that you want it. So that's the primary focus of this patent.

MONTAGNE: You follow this area of the news. Does anything strike you in this as not doable?

STEVENS: Parts of this definitely seem like the kind of thing that Amazon could implement very quickly. In fact, I wouldn't be surprised if they're using some aspects of this right now to move packages between fulfillment centers just to make that part of the process a bit smarter. But they also describe things that would require a lot of partnerships with UPS, having these packages sit at UPS shipping depots without a proper address on them and then having someone at UPS basically go in and apply the correct address and ship that package out.

I haven't heard any deals like that going on at this point, but that sort of thing, I think, UPS and the USPS would be very open to because they need all the money they can get.

MONTAGNE: Tim Stevens is editor at large for CNET. Thanks very much for joining us.

STEVENS: Thank you for having me. Transcript provided by NPR, Copyright NPR.