Anticipatory Shipping- The Game Changer in E-Commerce

Published by MBA Skool Team, Published on August 03, 2014

Remember the telepath Professor Xavier from X-men? E-commerce industries would go lengths and breadths to hire him so that they could understand and predict consumer demands. Under the pressure of the demands and whims of online consumers to deliver the goods in the shortest time, the online e-commerce companies are turning into fortune-tellers. Forbes recently reported that Amazon has obtained a patent for what it calls “anticipatory shipping” — a proactive system of delivering products to customers before they place an order.

Image Courtesy:, Stuart Miles

Moreover, it further reports how Google is trying to hurt Amazon where it has more expertise: in the business of selling goods and delivering them to customers. Google , with the help of its optimizing and scheduling software Google Now, is partnering with retailers to gain access to thousands of consumers through its venture named Google Shopping Express. The question still arises as to how fruitful these 'futuristic trends' can be in achieving their main goal - getting an order to the customer faster.

This article explores the anticipatory shipping model in depth. Further, it explores whether e-commerce can be revolutionised using this predictive technology.

The Model

The anticipatory model in simple terms predicts what users would likely buy, when they would buy it, and where they would need it. At the outset, this is based on the technology of predictive analytics. With the advent of big data, this level of maturity in the end-to-end sales process could tremendously boost the sales of a retailer - especially an online retailer- while bringing down the cost ultimately resulting in overall profit. The model uses many big data solution techniques to make sense of the large amount of customer purchase data such as a customers' previous orders, product searches, wish lists, shopping-cart contents and returns.

Most important of those techniques is collaborative filtering engine (CFE).Wikipedia defines Collaborative filtering as the process of filtering for information or patterns using techniques involving collaboration among multiple agents, viewpoints and data sources. Simply explained, it tries to generate filters such as “customer who bought this, also bought that” or “customers who looked at this item, also looked at these other items.” For example, the amount of time the mouse click hovers around the item on the site or the number of clicks may show that the consumer is interested in the product. However, clicks are not always an indicator that the user is looking for a product, and the click data can signal “noise” more than valuable engagement with a product. Yet, eventually with enough click path data, a collaborative machine learning filter will eventually find patterns and associations from which it can create associated groupings.

Once the eventual associations are recognised, the products will be packaged in one or more items as a package for eventual shipment to a delivery address, selecting a destination geographical area to which to ship the package, and shipping the package to the destination geographical area without completely specifying the delivery address at the time of shipment. Further, they will include complete delivery address for the package while the package is in transit. Thus, the product is closer to the consumer and hence the delivery time is least.

It is worth pointing out that a similar technology used in the above model has been successfully used by Bakers Shoes in collaboration with Descartes by using a Distribution Centre (DC) bypass system. Generally, it is common practice for a manufacturer to ship goods directly to a ’s domestic DCs, where goods are stored until being shipped to individual stores. In the DC bypass system, they remove one link out of the supply chain i.e. directly ship goods from manufacturer to the stores by having a 'floating warehouse'.

FIG. 1 block diagram illustrating a shipping network

Fig2. A block diagram illustrating a system configured to implement speculative shipping and late-select addressing.


Can this model change the game?


SWOT Analysis

A few strengths and weakness given in the SWOT analysis are discussed below:


Lower Costs

Since basis of this model is predictive analytics, if the retailer can harness the purchase data with clever algorithms to predict purchase behaviour down to an individual basis, they stand the chance of further reducing costs to their customers. A recent survey conducted by Boston Consulting Group supports low cost deliveries - it found that 74% of the 1500 US consumers surveyed prioritize free delivery over receiving their items the same day. By predicting the flow of packages till the point of delivery, the cost of storing inventory significantly goes down. According to Stitch Labs, a software tool to help manage inventory, inventory costs could be as much as 45%- 80% of the business costs. Hence, by tackling the inventory costs it would result in lower costs and hence lower prices could be offered to the consumers. This in turn will result in increased revenues and higher productivity.

Delivery Time

As explained before, the idea of pre-shipping to an intermediate location, based upon a forecast, but before the absolute demand is known, is not new. For years, retail apparel manufacturers have been practicing DC Bypass system. Typically, it involves a retailer shipping goods from Asia to cross docks in California. And while the goods reach the docks, the retailer is monitoring the what is being bought at the stores and when the goods reach the docks they can route the goods to the stores optimally. In Fig.1 , the process is shown to start at a DC that is used for central stocking of selected SKUs, specifically lower velocity items. The advantage with anticipatory shipping from a central location is the longer transit time gives the local demand time to build up. This would also allow greater use of slower modes of delivery that save money on freight. This model could certainly ensure that the “last mile” the product travels is short and is just as efficient as the “first mile”.

Customer Loyalty

Anticipatory shipping could give certain customers discounts—or even outright gifts—on products that customers received but don’t want. This idea is clearly proved in a sociology experiment conducted by Haisley, E., & Loewenstein, G. (2011) - 'The impact of gifts on deposit balances and customer satisfaction was examined in a longitudinal field experiment conducted at a commercial bank. Gifts increased deposit balances, survey response rates, and customer satisfaction compared to the no-gift control.' The retailer could use its data on household price sensitivity, and supply and demand, to decide which products and which households qualify for such deals. The crux of the idea is that if any retailer who gives freebies is going to make friends quickly and hence enjoy customer loyalty.


Big Data Algorithms

According to Harvard Business Review, companies have not been able to harvest the data to enhance their understanding. This is the reason why investments in big data have failed to pay off. Very few companies know how to exploit the data already embedded in their core operating systems.

High Returns

A study by the IBI research institute shows that 40 percent of customers already take returns into account when purchasing online. The eCommerce industry has already suffered from high return rates. Moreover, there is a possibility that with anticipatory shipping the big data algorithm may fail resulting in costly returns.

Cost of local nodes

This model will require a lot of local distribution nodes where the predicted products would be stored before they are delivered to the consumer. Hence, if not outsourced , the retailer would need to build up a local infrastructure to carry the anticipated shippings.

Fake Consumers

Another possibility could be where a consumer deliberately behaves as to receive a freebie. Predictive marketing specialist Paul Gibson thinks it “may train a consumer to browse and not buy, thinking they will get a freebie. And indeed if one doesn't then arrive, they may feel the retailer doesn't appreciate them." However, such a possibility seems to relate to the lesser percentage of the population


This model has incited a lot of industry excitement and only time could tell whether it will be effective in bringing down its overall costs. The fact that Amazon is pushing this phenomenon in a Minority Report kind-of-way, except that it's retail instead of crime, justifies a look at this futuristic trend.


This article has been authored by Pratik Nandekar from NMIMS Mumbai








Views expressed in the article are personal. The articles are for educational & academic purpose only, and have been uploaded by the MBA Skool Team.

If you are interested in writing articles for us, Submit Here

Share this Page on:
Facebook ShareTweetShare on Linkedin