Many of the optimizations and enhancements that online retailers make to prepare for the peak season are useful year-round. Here are 6 ways that tech has changed holiday shopping.
1. Ship From/To Store
Being able to fulfill and ship orders from your warehouse is normal - but what happens when the order volume suddenly jumps up by a factor of 10? Or a 100? Or 10,000?
Pop-up distribution centers are certainly an option, assuming you have a few months lead time to get them established and stocked. Another option is to ship items from existing stores.
There are several advantages of treating stores as distribution centers:
- Faster turnaround time
- Faster delivery time, especially if the store is physically closer to the customer
- Better margin, because you are able to sell products while they’re still in-season/-style
- Improved inventory turns - items that don’t sell well in one store can be shipped out faster
- Reduced shipping costs, in many cases
Doing this requires the order-routing and -fulfillment software systems to treat stores as warehouses, and select where to ship items from based on multiple factors instead of just last-in-first-out or simple proximity. In most cases, the decision reduces to which option is cheaper/better, but quantifying cheaper and better is non-trivial!
The flipside of this is allowing customers to choose to ship their online orders to a specific (presumably nearby) store, for pick-up. When appropriate, this lets the system fulfill the order from the store where the items will be picked up, reducing shipping costs to zero.
The engineering effort and software required to do all of this can be substantial, but it is well worth the effort. It also has the advantage of continuing to work after the holidays, allowing for order optimizations and happier customers throughout the year.
Not Just for the Holidays Any More
This is an important point: all of the enhancements listed in this article provide strategic advantages to the retailer year-round, not just during peak seasons. The additional flexibility means that bikinis not sold in the brief Vermont summer can be shipped to California customers in the fall as needed, reducing bulk transportation costs and improved turns and margins.
2. Not Shipping from Stores is Good, Too
Fine-tuning the order-fulfillment mechanisms during peak times can also include not shipping some or all items from stores, so the stores can focus on fulfilling in-store purchases.
Ordinarily, each store is treated as a mini-warehouse, which means that employees must fulfill orders routed from the e-commerce site and also from customers in the store. When Black Friday hits, this can cause the employees to become overloaded, which detracts from the in-store customer experience.
The solution is simple: hide stores from the e-commerce system during peak times. A variant of this would be to limit the quantities available to e-commerce orders from certain stores.
The engineering behind this solution is not so simple - but again, it is well worth the effort to preserve the in-store customer experience (and prevent staff burnout).
Note that there can be two levels of this: making items invisible to the e-commerce site does not prevent orders from being routed to stores; also making items invisible to the fulfillment-routing service prevents orders from being routed to stores. A combination of both “levers” can be used to find the appropriate balance.
3. Automate IT peak preparations
The last thing you want to happen on peak days is a systems failure. To prevent this, automate the IT peak preparations to support stores, warehouses, and e-commerce portals/apps well in advance of the peak days. This lets you prepare and test in advance of The Day, and also ensures that IT employees are not working insane hours during holidays.
Make a List and Test IT Twice
There are multiple critical factors to ensure peak-time operational fluidity. Make a list and test it twice, especially IT functions such as provisioning, scale-up/scale-down, fulfillment visibility and order routing, et al. Creating a test environment to simulate peak times is a valuable investment; this allows product owners to create expected/extreme scenarios to see what will happen with different balances of e-commerce and store loads. This kind of simulation (well in advance of The Day) can help determine where pop-up warehouses might be useful, for example.
While it may not be practical to actually load the production servers with projected traffic, get as close to this as you can before The Day. Remember the e-commerce Disasters of Christmas Past!
4. Split Order Fulfillment
With the basic fulfillment-from-store and inventory “load balancing” mechanisms in place, more sophisticated strategies can be employed to further improve system performance, customer service, and margins. For example, small “accessories” may be more economical to ship from stores than warehouses, even if this means splitting the customer order into multiple shipments. In other cases, it may be better to ship everything from a single store.
It is difficult, if not impossible, to determine what is best for a particular order unless and until you have these more-sophisticated systems in place. Once you have them, multiple opportunities to improve performance become possible.
5. Real-Time Inventory Accuracy
The ultimate goal of any inventory/warehouse management system is to have 100% accurate inventory counts, locations, and availability at all times. In practice, this can be difficult to achieve.
Older, monolithic inventory-management systems may have been designed in the days of batch-processing, which causes inventory accuracy lags. Even with more modern relational database systems, peak times may cause severe database-table contention, resulting in failed transactions and overall processing bottlenecks.
It is not unusual for the “truth” about inventory to be spread across multiple disparate systems; knowing exactly how much inventory you had of a particular item at a certain store at any given moment can be difficult in this case. The only realistic remedy for this is investing in more modern architectures, approaches, and development methods - well in advance of Black Friday.
Again, the investment is well worth the effort and expense - more, happier customers, less-stressed employees, greater sales volume, and higher margins.
For retailers, there’s no better holiday gift than that!
6. Platforms, not Applications
The key to achieving these remarkable results is realizing that IT retailing consists of a collection of related services - a platform - not a bag of disparate applications. Moving to a service/platform model is a key enabler for achieving these kinds of optimizations, and also provides a basis for ongoing improvements and optimizations that no collection of monolithic applications can touch.
This has become easier with the rise of service-oriented architectures (SOA) and especially microservice architectures, but be warned: taking a monolithic enterprise application and splitting it into services is no sleigh ride. It takes brave planning, fierce decoupling, and a precise development process to turn an set of disparate applications into a coherent service platform.
Conclusion
These are just some of the ways in which tech has improved the peak-season shopping experience, for customers and companies. The list is far from exhaustive, and the precise changes and enhancements vary with each company and style of commerce.
It can be done, but not by Santa Claus - and not the night before!