John Lewis is a leading retailer in the United Kingdom. The chain includes 29 full-line department stores, 8 At Homes (small format stores), and an online/catalogue business. The company differentiates itself on customer service. When a customer orders a bed and a washing machine, for example, John Lewis not only delivers the items, it also offers to take away and dispose the old bed and washing machine.
A customer can order items in a store and then decide to take some items home and have the others delivered. Or a customer can order items online, pick up some of the items at a local John Lewis or Waitrose shop, and have the other items delivered.
To enable this seamless customer experience John Lewis added a “dynamic booking system” (DBS) to its ecommerce site. Andy Gay, a systems manager at John Lewis, was kind enough to speak to me about this project.
The dynamic booking system has some interesting characteristics. The delivery options are based on a real time, constraint-based, continuous routing optimization engine from Descartes (an ARC client). This engine runs behind the scenes and gives customers feasible schedule options within 7-15 seconds. If the customer wants the delivery sometime in the next week, and the goods they bought cost more than 30 pounds, and the customer is willing to take delivery in a six-hour time slot, then the delivery is free. If they want a four-hour time slot, they can elect to pay more. If they prefer a two- hour slot, then the price is higher. And if they want a two hour slot on a Saturday, a very desirable time slot, then they would pay even more. And the highest price is for an expedited delivery that would occur within the next two days.
Ecommerce Site Delivery Options are Powered by Descartes Routing Application
In short, the DBS gives customers delivery options, provides an additional revenue stream for John Lewis, and improves John Lewis’s on-time delivery metrics. The routing engine also has reduced the average miles traveled per delivery, as measured from the company’s outbound service center to the customer’s home, from 7.6 to 6.7 miles.
John Lewis takes its corporate social responsibility seriously. The company has committed itself to reduce carbon emissions by 15 percent within ten years. This project helps the company meet that commitment; in fact, the desire to be “green” was one of the project drivers.
The project was completed last year and it has been a smashing success. Service revenues associated with assembling products at a customer’s home have increased significantly over those pre-DBS. One reason for this is that in the past if in-store shoppers wanted a service associated with a home delivery, it was a burdensome paper-based process. With the online DBS delivery engine, coordinating home deliveries with services is much easier.
The ability to accomplish this, however, did not come easily. DBS was a multiyear project. The routing engine is based on complex optimization, and yet the solution had to scale to provide quick delivery options to large numbers of customers. The routing engine also needed to be integrated to several applications.
But perhaps even tougher, John Lewis needed to custom build an “administration” layer with logic to support the routing. While the routing engine can do the delivery optimization, the admin subsystem needed the cube of every product so that the routing engine would know what could fit on a truck. The company needed to know whether a product could be carried by one man or would require two, whether an installer would be needed as part of the delivery, and if so, what kind of installer.
The company also had to do stopwatch studies. To give feasible schedules, the company needed to know not just how long it would take the truck to get to a customer’s home, but a whole series of other time standard driven parameters. Was this a delivery to a house or the seventh floor of an apartment building? How long should it take to get a washing machine to the seventh floor? What should the time standard be for installing a washing machine? How long should it take to tote away an old washing machine? I’ve written about using time standards in a warehouse. That is tough enough. What John Lewis did was an order of magnitude more difficult.
Finally, the company’s delivery team needed to transition into being more than drivers, but rather serve as an extension of the retail selling team and view themselves as critical to delighting the customer. That, obviously, is a very big change in culture.
In conclusion, John Lewis can provide a much more unified and seamless experience across retail channels then the vast majority of retailers. But getting there was not easy.