As the digital economy booms, consumers have ever greater choice in what they can buy, when and how. This can pose problems for fulfillment centers like ours. In this post, our CEO, James Hyde, shows how a little academic thinking can solve help these problems.
I recently met with Professor Roger Maull, Academic Director of the Initiative for the Digital Economy at the University of Exeter (INDEX) in the UK.
Roger and his team of around 10 academics are researching how new technologies impact diverse areas of the economy. These include how wearables can predict stress and disease; how personal data is shared, used and ultimately owned; and how the Internet of Things will one day predict when to order you toilet roll.
One aspect of the digital economy that Roger’s team is particularly interested in is variety and its effect on business productivity.
The digital economy means more variety
As eCommerce booms, we find more and more choice in what we can buy, when we can buy it and how it’s delivered. They say that variety is the spice of life, but it’s also a killer when it comes to business efficiency – and this is the issue that Roger and I discussed.
The pick and pack of online orders, which we do lots of at James and James, may seem like a simple process. But when you try to optimize the work of a hundred people, plus thousands of orders, storage locations and products, in an ever-moving fulfillment warehouse, you quickly come across many problems. These, including the age-old Travelling Salesman Problem, have been debated by academics for decades.
More variety means more complexity
Roger’s colleague, Dr Phil Godsiff, uses a new analogy – Campari Soda – to explain this problem of variety.
Willy (who obviously lives in Italy) goes to the same cafe each morning at 11:00 and orders a Campari Soda. The cafe owner opens up at 11:00, makes it for him and everyone is happy. But there are five ways in which this situation could change. Each puts different demands on the cafe owner, if he is to keep Willy happy:
- Time: Willy turns up at 10:00, when the cafe is normally closed
- Volume: Willy orders 100 Campari Sodas all at the same time
- Request: Willy starts ordering a different drink every day
- Effort: Willy turns up with his own recipe and offers to make it
- Preference: Willy decides that today, he’d like a little less ice
These five types of variance – time, volume, request, effort and preference – all need managing. And if more than one of them changes at the same time, this only compounds the difficulty – Willy turning up at 10:00 and ordering 100 different drinks would require some serious investment by the cafe owner.
More complexity means a pick and pack challenge
The level of variation in each type can be very scientifically stated using the following scale of options: 0, 1, 2, some, lots and everything. This is explained to me on the small, but perfectly formed, self-adhesive chalkboard you can see in the background above.
We can use James and James – and the orders we receive from online stores – to provide a real-world example:
- Time: In a fulfillment center, orders turn up all day long, but luckily the courier collections are at the same time every day. So there are some time requirement differences between orders.
- Volume: Any number of orders could turn up each day depending on factors like the weather (on a hot day, consumers tend to shop online less). We must get all of the orders packed before the end of the day, so there is lots of volume variance to manage.
- Request: It goes without saying that almost every order is unique. So the set of possible products and locations that must be picked from is pretty much everything.
- Effort: All single item orders are roughly the same effort, but multi-item orders take significantly longer. We use three different packing processes to account for this variety: orders of just 1 item; orders of 2 to some items; and orders of lots of items.
- Preference: Our clients often have specific requests: always pack this product in a box; use a red box for these orders; include a leaflet with these products; or add this paperwork to this wholesale order. Every client has their own preferences and to cater to everything would involve a complexity of almost infinite options – everything.
Less variety means more efficiency
Ashby’s Law of Requisite Variety says (in a nutshell) that any system must be at least as complex as the number of varieties it has to manage. This basically means that, if you want a process that can cope with hugely varied work, you need to design a system that is hugely complex.
And here comes a business choice. You can either have a business that does everything your clients ask for, is very manual, and so either expensive or very prone to error. Or you can have a business that offers a limited set of options, is more automated, and so is consistently productive and accurate. You cannot have both.
At James and James, we’ve taken the second option – we believe that it’s better to be excellent at the variety you choose to offer, than to be mediocre at everything. For example, we focus on working with direct-to-consumer brands that sell small, fast-moving products. This reduces variance in request and effort. We enable clients to customize shipping labels and packing slips, rather than all of their packaging. This reduces variance in preference.
This approach has helped us to create a detailed order fulfillment process, which has both flexibility and strict boundaries, to ensure absolute quality and efficiency at every stage.
It also means that we can manage the process through our own software platform. This takes into account all of the above variances, to calculate the actions required, between us receiving an order from an online store and dispatching it to a customer. Nothing is left to chance and human error is removed.