Smarter Supply Chain for Retailing
Supply Chain matters more than ever in retailing. No matter how far you extend your definition of supply chain (we include assortment, allocations and forecasting, as well as logistics in ours), the strategy to get merchandise to a place for sale—digital and physical—and then ultimately into the hands of the customer, have never mattered more.
For most retailers, Omni has driven up the cost of logistics. Once relegated to the ‘back of the business’ in B2B palletized shipments, now it is front and center of the corporation, and the buyer’s mind-set, with the variety of shopping experiences and channels.
As we discussed in chapter one, retailers are struggling with their Omni-strategies on several fronts. In this section, we want to discuss the focus and solutions, and then deal with these major Omni-challenges.
The Warehouse Is the Store, and The Store Is the Warehouse
One of the big challenges in Omni is the smaller orders of ecommerce. The ‘one each’ model drives a completely different warehouse model, from facilities management, equipment, to personnel. Big questions plague the retailer such as:
- to have one national mega-warehouse or lots of centers around the country;
- to mix mode the warehouse—in order words, pallet/carton level shipments to stores within the same fulfillment center for the ecommerce operation; and
- how to leverage store geographic presence—in other words, can I make the store a warehouse, too?
Of course you want to hear the answer. But there is no one right answer. Since our inception at ChainLink, our key principle for supply chain design is that markets determine supply chains. So, for example, the category of product and the expectation of the customer is crucial in deciding a distribution strategy. Apparel can easily be shipped in 24 hours or less from one central warehouse. But promising overnight delivery for furniture—that requires regional centers.
Another example, if you are getting into that fast or same-day delivery (which makes sense to grocery), being near the customer is key, no matter what the category is. Here, leverage of the store as warehouse has to be part of the thinking, since same-day is a huge commitment of resources. And at scale, your warehouse costs will sky-rocket. Retailers need to consider the impact to cost and profitability as they add each level of service and capabilities into their business model.
In reality, it is time to dust off and update the distribution/multi-echelon planning model as part of the analysis here. These kinds of investments really require a deep understanding of the business model, the customer and the competition. A piloting must take place to see how to effectively manage in these new environments before any national strategy can be put in place, even when leveraging established third-party service organizations.
Part of the model may also include leveraging suppliers’ warehouses and locations. Retailers are often requesting drop-ship options from the manufacturer directly to the consumer. So suppliers are also going through the same analysis of their models. (Suppliers are just as driven by Omni-challenges as retailers. For more on suppliers (manufacturers) and Omni-channel, read this blog series.) The supplier may need more stock in ‘sales-ready’ inventory, which will change their pick/pack/shipping operations or warehouse design to support consumer-oriented orders and also include parcel expertise (and software) to the mix.
From an allocations perspective, retailers are, rather than pushing the whole collection or seasonal offering to the stores, holding back at the DC for ecommerce orders as well as hedging their bets on where demand will materialize, and reduce shipments between stores to support that demand.1 Allocation is no longer a static fixed plan, but must be responsive to actual sales. Richer rule sets and integration to analyze more frequent data are required, looking at a variety of criteria to understand when to allocate inventory. Rules can be chain-wide and/or by category, seasons, promotions, store locations or formats, demographic attributes, and so on.
Allocations Need to Include Logistics
Customers don’t buy unless they ‘see’ merchandise. But the cost of cross shipping between stores and warehouse is pretty high. (In this blog post we follow a shirt through the supply chain and show the costs of each move.) In the past, merchandising and store managers thought more about getting the right presentation at the location for initial sales, and then later dealt with all the subsequent changes to in-store stocking (due to promotions, weather, greater or lesser than expected sales. They thought more about moving merchandise off the shelf to get it sold rather than the costs of all those moves.
With web sales, there is a tug of war between needing to justify the store’s existence vs. the low cost of fulfillment from a center warehouse. So just as in the past, micro-merchandising and allocations analysis is about having the right quantifies by location. Now that model needs to include a third dimension: the cost and methods around shipping within the retailer’s operations. Whether pooling, rebalancing merchandise, or ship to store for customer hoped-for pickup, these logistics costs are part of the cost of goods sold, and they eat up profitability. This requires more agility out of retailers to master these moves, smartly, whether fast fashion or a more traditional model. There is much more to contemplate; you can read some ideas here.
Selling the Shirt, Not Losing the Shirt, on Home Delivery
We have to learn quickly what others are doing right—and wrong—if we want to keep up and not lose a ton of money in the home delivery game. The key here, again, is understanding your customers and what they are ultimately willing to pay for—or not. Blanketly offering all shipping options may just not make sense.
Here is a good resource for sharing ideas and case studies that I encourage you to read. And in ChainLink’s the brief, read about Profitable Fulfillment.2 Retailers like Home Depot, Woolworth in Australia, Crate and Barrel, Rakuten in Japan and John Lewis (my poster child for smarter home delivery strategies), are featured within the above case study links. Whether you’re a large or small retailer, and/or a logistician that serves retailers, you can learn so much from these papers and case studies.
New Solutions – Continuous Platforms
Today we hear the term ‘platform’ a lot. Turns out, there are many meanings. But the concept for Omni-retail/fulfillment is one where applications no longer quadrant off from one another—logistics from ordering, or forecasting from warehouse distribution—but rather work in concert from planning through execution.
The new fulfillment platform not only shares information from task to task (here we mean tasks such as locating an item, determining its price, determining shipping preferences, optimizing the time/cost of transportation, and so on; or evaluating demand for an item at a location and then analyzing the cost to move it to that location and then determining if that is the most profitable decision, and so many more). The tasks and data need be orchestrated for the optimal outcomes in a continuous fashion, since ecommerce applications are always on—24×7. Each order changes the picture—the schedule and routes in transportation, the pick list in the warehouse, and so on. Scale and density or the lack therefore, change the ability and cost to execute. An optimal path for fulfillment of goods to a customer may be different from day to day. Retailers need to learn to orchestrate this continuous analysis in order to achieve the best economic game plan.
This can’t be done with past software. Traditional optimization models are not designed for an always-on model. In a continuous process, always-on retail, there really is no end—from planning through execution. An always-on approach recognizes that change is always coming—new orders, new customer requirements, disruptions on the road that impact the current schedule, or shorter or longer service calls.
Continuous platform also means making your whole network one virtual warehouse. So all the stores, DCs and in-transit inventory is visible. In this way the best fulfillment strategy based on time, cost and so on can be generated.
Smarter inventory management is really required here, too. Of course, you are ultimately exposing your inventory system to your customers, so accuracy is crucial. RFID, barcoding and a modern distributed warehouse fulfillment system is required. RFID seems to be one of those technologies that is light years ahead as far as its contribution to inventory accuracy. The platform needs perpetual inventory updates, possibly in real-time to ensure that stocked items can be sold and promised items do exist.
There is no segmentation, then, from the cool online site your customers are seeing vs. the back warehouse, or, possibly even your suppliers’ stock.
This continuous flow of data can also provide a much better picture of demand. Think of it. Demand has to be understood in that always-on mode, too. Analysis of actual sales and the cost of shipments by location can give retailers a better handle on future stocking strategies. Though catalogue selection offerings may be large, not all items need to be stocked at all locations—or at all. Demand at shelf/location level is not a common forecasting approach by many retailers. But with techniques like flow casting or daily analysis of demand, a much better inventory approach can be developed on what to stock—and what not to stock.
Firms like JDA, Descartes, GT Nexus, One Network, Logility, and Tools Group, and others have spent millions on architectural changes to respond to this new reality. Buyers of technology can learn a lot from these firms and their customers on the most advanced ideas today.
Supply Chain management is going through radical changes as it keeps pace—and sometimes leads—the business strategies of retailing today. In an always-on retail world, there is an assumption that there will be an immediate response. That means that all the systems and processes have to be responsive—and accurate. The days of semi-annual cycle counting are over. Inventory has to be accurate up to the second—by location—in order to accurately promise and fulfill. No overnight delivery will occur if the promised inventory cannot be located. Stuffing channels and over stocking is not an answer today. Nor will having warehouses all over the country be affordable. Data mastery must be part of the solution. But data is a servant of a smart process, one that is in sync with the needs of the market.
To view other articles from this issue of the brief, click here.