Sorry!! The article you are trying to read is not available now.
Thank you very much;
you're only a step away from
downloading your reports.

With Predictive Shipping, Amazon May Send Free or Deeply Discounted Products to Show Its Love


Is speculative delivery the perfect optimization of data mining or patently ridiculous?

All hail the clairvoyance of YouTube satirists: "Yesterday Shipping" from Amazon (NASDAQ:AMZN) has arrived.

The world's largest online retailer has kicked 2013 (the unofficial Year of the Delivery Drone) to the curb and entered 2014 with a recently approved patent for an ambitious concept it calls "anticipatory package shipping."

The patent -- recently excavated by the Wall Street Journal and available in its entirety here, if you're looking for fireside reading material -- explains how Amazon will forecast customer demand for products before orders are ever placed online. Amazon will then package and bulk-ship these ostensibly in-demand items from its fulfillment centers to regional hubs, where the items will await shipment to customers who eventually click the "Place Your Order" button.

Products may also receive partially complete address labels and joyride locally in trucks provided by such carriers as FedEx (NYSE:FDX) and United Parcel Service (NYSE:UPS). Once orders are actually placed, the full address will be completed in transit and the product quickly delivered to the specified customer's location.

Amazon claims this system will mitigate e-commerce's main disadvantage when competing against brick-and-mortars by cutting down the delay between order placement and delivery. Product predictions will be based on previous customer purchases, search history, items on customer wish lists and in their shopping carts, and even how long customers' cursors hover over products while they're window-shopping online.

This anticipatory algorithm is a prescient upgrade from Amazon's Local Express Delivery (a limited same-day service based on actual orders) and the company's "release-date delivery" feature, which customers can opt into if they want to preorder items such as high-demand video game or movie releases. The newly patented system is also Amazon's way of optimizing the enormous amount of data it has about its customers in an effort to one-up other existing same-day services offered by the likes of eBay (NASDAQ:EBAY), Wal-Mart (NYSE:WMT), and Google (NASDAQ:GOOG).

The patent does allow for occasions when customers receive a product they didn't actually order. Initially, this sounded suspiciously like the negative option billing used by the mail-order music clubs that helped me build up my R.E.M. collection in the '90s: If I didn't opt out by sending back a special "No, thanks" card, I automatically received (and had to pay for) a club-selected CD every month.

The Amazon system doesn't appear to work that way. The company anticipates there may be instances of accidental delivery or cases in which a speculatively shipped product the company can't dump (and that would be too expensive to return or reroute to a different area) will be delivered to a potentially interested customer for a discounted price or even as a "promotional gift to … build goodwill." According to the patent, however, those recipients would likely be vetted as "valued" customers whose "potential interest" would be gauged by an impressive purchasing history or desirable demographic profile. In other words, you probably won't get a free copy of Jackass Presents: Bad Grandpa if you're a great-grandmother from Biloxi who barely goes online.

Second, consumers shouldn't freak out if they do happen to get an Amazon product they never ordered. According to both the Federal Trade Commission and the US Postal Inspection Service, consumers are under no obligation to pay for unsolicited items received in the mail. Only two types of merchandise may be sent without your consent: merch mailed by charitable organizations to solicit contributions, and clearly marked free samples. Amazon can't just ship me the new Beck album because it thinks I'll like it and expect me to pony up for it (though I probably would anyway, because it's Beck).

Getting Inside Our Heads

According to a recent survey by SAP AG (NYSE:SAP), 85% of US organizations using predictive analytics report it having a positive impact on their business; 88% claimed it gave them the upper hand over competitors, with prediction of customer needs and market trends ranking as a high priority. Amazon wants to mine its meticulously gathered customer info to not only meet demand, but also to steer it.

Some say Amazon's proposal is simply a modification of the "just in time" inventory model often used by brick-and-mortars to control supply and demand: Product is ordered in quantities to meet immediate needs and then delivered exactly when they're needed. Amazon is simply taking the model into a new realm, gazing into a virtual crystal ball to decide who will need these products before they even know it themselves.

The success of its model will partly come down to how customers feel about surrendering to a "buyer power" as a trade-off for convenience and instant gratification. While I'd like to think We The People would resist an Orwellian future, I know I've already become desensitized to and gladly accept the customized coupons the Stop & Shop register spits out at me when I swipe my rewards card (how does it know I wanted $1 off Fage and not Chobani?!).

As described in the New York Times, Target (NYSE:TGT) once tasked its Guest Marketing Analytics department to tap into the power of predictive analytics and identify soon-to-be new parents (the "holy grail" of retail customers). The team used customer data to pinpoint women in their second trimester of pregnancy and then mailed them baby product ads and coupons.

Target's research, however, found that some pregnant women felt "spied on" and "upset." "If we send someone a catalog and say, 'Congratulations on your first child!' and they've never told us they're pregnant, that's going to make some people uncomfortable," statistician Andrew Pole told writer Charles Duhigg. "[Even] if you're following the law, you can do things where people get queasy."

The more explicit questions seem to revolve around the Amazon system's logistics. Will third-party carriers even have the ability to sync their own delivery modules up with Amazon's to execute this sophisticated-sounding plan? How much loss will Amazon incur, especially initially, if it miscalculates potential interest in product? Can customers opt out of having their personal information scoured for insights? And, as Mike Flacy of Digital Trends points out, "abuse of the system may eventually occur … assuming customers figure out how to trick the predictive model into shipping items in advance, then refusing to pay for or return the item after it arrives."

CNET's Chris Matyszczyk likens Amazon's anticipatory shipping to a "perfect lover … who anticipates your needs rather than reacts to them." Matyszczyk then speculates about how Theodore Twombly, Joaquin Phoenix's introverted character in Her, would likely embrace this model. I'd like to think Amazon's proposal can work to consumer benefit (and it would be awesome if it does), but I keep thinking of another perspective on Twombly: a man who paid a heavy price for becoming too dependent on a seductively intuitive operating system that catered to all of his needs.

Jenn Gidman is an editor and writer who has whipped content into shape for MSN, AOL,, Reader's Digest, and Scholastic. You can find her on Twitter at @jenngidman.
< Previous
  • 1
Next >
No positions in stocks mentioned.
Featured Videos