Prediction: Amazon Predictive by 2021
In Feb 2014, I read an article in the WSJ about Amazon patenting anticipatory shipping. Basically if you put something in your cart and are progressing towards buying it, they will start to ship it to you. From the article:
Amazon.com wants to get you that package fast, really fast — before you’ve even ordered it.
The Seattle retailer late last year was granted a patent for what it calls “anticipatory shipping,” a method to start delivering packages even before customers have clicked “buy.” The technique could cut down on delivery time and cost and help ensure consumers don’t turn to brick-and-mortar retailers.
That got me thinking. Why wait until you put it in your cart?
Amazon knows a lot about you. They know what you buy. What you look at. What you put on your wish lists. How often something is on your wish list before you buy it.
They could start shipping you things before you even buy them. If you don’t like them, ship them back.
The thought hit me so hard that I wrote down this prediction:
Amazon is going to ship you stuff they think that you were going to buy. If you like it, keep it, if not ship it back. — February 22, 2014 at 1:48 AM
Why Predictive Ship?
Amazon could easily start doing tests on this. It could start small. They pick an item that is maybe over stocked at their warehouse. Write an algorithm to pick the 1,000 people most likely to want that item the most. Factor in their address, their buying patterns. Look at how long the item has been on their wish list. Then, ship it to them. The consumer gets an email saying:
Surprise 🛳! We think you are really going to like what is in the box that is coming tomorrow. If you don’t, just slap the return sticker on the box, tape it back up and send it back to us!
If the customer doesn’t ship it back in 15 days, Amazon bills them.
Amazon then carefully tracks and measures how many people do or don’t send it back. They take those learnings and put them back into their algorithm. They do this a couple times and they could get a pretty good at picking whom to send what to.
The Math Works Out
I am going to make up a fictional product with made up costs and margins. Depending on the price of the item, the margins, and the return rate, this could easily be profitable.
Imagine a $25 item. The item costs Amazon $5, and they already ship it for free so shipping costs them $7.50. Amazon earns roughly 50% margins on this item.
$25 — $5 cost of goods — $7.50 shipping = $12.50 profit.
If they ship this item out to 1,000 people and no one returns it, which means all the consumers chose to buy it, Amazon earns: $12.50*1000 = $12,500
If everyone ships it back, that means they paid to shipped out, didn’t earn anything, then paid for you to ship it back. They lose: 1000 people *2 shipping trips*$7.50 shipping = $-15,000
The break even point is at 54.60%. If 54.60% or 546 consumers keep the item, Amazon makes a whooping $15. If one less consumer ships it back, for 54.50% keep rate, they lose $12.50. So if the Amazon algorithm can beat %54 they make money every time.
Of course this is a made up example with made up margins and costs. But Amazon could, and maybe already has calculated this out. They can take their real costs and margins, put them into a similar equation and know exactly what % their algorithm has to hit.
Other think this might happen as well. Found this tweet from Matthew Panzarino from 2013.
So I am going to go on the record and say
Amazon is going to start predictively shipping items to consumers that they can keep and pay for, or ship back, by the end of 2021.
Tweet for the record.
Amazon is already close
Amazon already started down this path by shipping samples.
Do I need to be a Prime member to receive samples? No, you don't have to be a Prime member to receive samples. Anyone…
Amazon's latest advertising play involves free samples delivered right to your door
Scott Olson | Getty Images Amazon is sending shoppers free samples of products - curated to their tastes - as part of…
If they are already shipping a box to you, might as well throw something else you might enjoy as well.
Free samples now, paid products later?