Retail: What’s in store for e-commerce operations

Depending on a business’s market, size and geography, the concept of omni-channel fulfillment can mean any number of things. However you slice it, the idea of enabling consumers to buy anything, anywhere, at any time is daunting, particularly when you provide choices about where and how they’d like to receive their purchases.

As the linear supply chain transitions into a supply web, fulfillment functions can now be found in distribution centers, stores and even manufacturing facilities. The blending of best practices from each world has created a new battleground for competitive advantages.

As a result, the store’s role in omni-channel retail fulfillment is undergoing rapid change. The notion that e-commerce will soon make brick-and-mortar stores obsolete has largely been debunked, but businesses have had varying levels of success turning their retail footprint into a strategic advantage. Many have borrowed technologies and processes from warehousing and distribution operations, and some have seen inventory and fulfillment systems fully integrated with those inside the four walls. But according to Kim Baudry, market development director for Dematic, few have achieved the ultimate vision of omni-channel customer service.

“Walk into the store of someone who says they embrace omni-channel,” Baudry suggests. “Ask an associate if you can get a product shipped from another store to your home if it is not available on the shelf. At best, they might be able to look it up online and ship it from the warehouse.”

Just as distribution operations are compelled to overcome a pallet-based legacy to deliver smaller and more complex orders, retail stores and back rooms have their own set of inherited limitations. Substantial change is required in both places, and the businesses leading the way have learned that the DC and store need to be on the same page if it’s going to work.

“Two years ago people thought, ‘we’re going to get by,’ even though everything was indicating they would need to develop ‘from anywhere, to anywhere’ capabilities,” Baudry says. “Now it’s more serious, and many are realizing they have to do it.”

Filling the wishes of store and DC
The development of in-store fulfillment has taken multiple paths, says Bryan Jensen, vice president at St. Onge, an engineering and logistics consulting firm. He identifies three of the more prominent concepts, beginning with the idea that store inventory can be used to support demand for “tail” SKUs. These SKUs include odd sizes, last season’s colors, and other things that are still in stores but not in the DC.

“The question is how to give an e-commerce customer access to an odd product at a store in, say, Manhattan,” Jensen explains. “The old strategy was to collect the items, ship them back to the DC and resend them to the stores where they will actually sell. This costs enormous amounts of money. But if I make the item available on the Internet, I can satisfy customer demand in Florida.”

Shipping from a store can also soften the impact of single-line, direct-to-consumer orders on distribution operations accustomed to larger orders bound for stores. Chris Arnold, vice president of operations and solutions development for Intelligrated, is careful not to overstate the volume likely to ship from stores. “But even 100 orders across a network of stores is a big number for some retailers,” he says. “It can serve as a sort of pressure valve for peak times in the DC.”

Coordination between the DC and store might also take the form of store- or aisle-ready shipments from the DC, which could include items bound for in-store pickup or ship-from-store e-commerce orders. “Depending on how you batch, you can get some good efficiencies by picking to a tote container that clearly identifies an e-commerce order for a store associate,” Arnold says. “A DC picker can then fill a larger order made up of several single-line orders, and it will feel just like a standard store replenishment order.”

This leads into Jensen’s second observation, which involves the desire to position inventory as close as possible to the customer. The traditional “push” model is characterized by DCs sending inventory in bulk to stores, who then deal with managing and selling it. The omni-channel approach suggests a “pull” model, which would instead ship only the inventory most needed at each store.

Bruce Stubbs, director of industry marketing for Honeywell Scanning and Mobility, says these models are not necessarily at odds with one another. “The goal is not to overthrow push with pull,” he says. “We’re not going to manufacture and ship an item in the same day just because a consumer demands it. But if the DC knows the true demand from a store, it can actually help preserve a push approach at the DC based on the reality of inventory.”

Instead of scrambling to react to stock-outs or overwhelming a store with excess inventory, a partnership between the store and DC can result in just-in-time allocation. Jensen says some have overcome the challenge of travel time between the two by establishing hub stores. If 100,000 or more SKUs are available in the DC, perhaps only 30,000 are available in each store. But a hub store will have certain quantities of nearly all SKUs. A form of this has been around for some time, Jensen says, but having a hub store handle store replenishment and direct-to-consumer fulfillment can alleviate omni-channel pressures on the store and the DC.

Keith Phillips, president and CEO of Voxware, says this satellite DC might serve as an enhanced back room for a single store or provide three or four deliveries per day to each store within a certain radius. Smaller trucks on smaller routes might ferry items to the back room of a store, which in turn might also resemble a miniature DC. This translates into having somewhere between an eight- and 12-hour inventory supply for that store and customer pickup for online orders, he says. But it’s not as simple as trucks whizzing back and forth four times a day. Most back rooms have been specifically designed to support just that one store.

“Retailers have been very good at that for decades,” Phillips says. “But now the same space has to supply this store, that store and direct-to-consumer. And, you can’t just make an assumption that you will sell the same number or type of items in the store as you used to before e-commerce shoppers entered the mix.”

Bring the back room to the forefront
A store’s back room is not typically known for its rigid discipline and organization. The square footage can be very small and as items are received, they are rarely validated like they would be in a warehouse. An item might be in stock, but associates can’t know where it could be. This causes problems for customers in store and online.

Jensen says a back room does not need to be loaded with picking inventory, but what inventory is there should be pickable. Solutions like RF and voice can boost visibility while also supporting processes for packaging and labeling. In many cases, Jensen says the pack stations in a back room could be a carbon copy of those in the DC.

The store’s back room will benefit from any improvements in labels, packaging, bar coding and tracking tools already coming from the DC. Still, many have yet to deploy RF guns, so the idea of leapfrogging an entire generation of data capture technology to adopt voice and real-time capabilities for e-commerce can seem intimidating. Nevertheless, it’s a challenge many retailers are eager to tackle.

“There is executive management interest to know what is in the back room,” Dematic’s Baudry says. “They know it will impact the DC, so it’s a natural extension to break down the silos, and we’re seeing more buy-in from COOs and CFOs to unify inventory.”

That said, the inventory accuracy challenge extends beyond the back room: Imagine if consumers were allowed to wander around inside a DC picking whatever they wanted. Items on the retail floor are constantly in flux, which can complicate efforts to make that inventory available online. A sale is a sale, but a central question for ship-from-store fulfillment is how to simultaneously manage the experience of the in-store and online consumer.

Arnold offers the example of a retailer who went to great lengths to camouflage the process of picking from shelves to satisfy e-commerce orders. “These ‘secret shoppers’ were associates who looked like any other customer,” he recalls. “They picked items to an ordinary shopping cart and brought them to the back room to validate, consolidate, pack and ship.”

As the methodologies of the back room become more similar to those in the DC, Arnold says he has even seen some rudimentary sortation technologies developed. The result is a kind of “outsourcing” of fulfillment functions. “Because you have solutions in the back room to pick, pack and ship, the DC doesn’t need to do it for every item,” he says. “The DC can pick for multiple stores, send a tote to a store for sortation, and the burden of processing them for either e-commerce or pickup at other store locations has been removed from the DC.”

Distributing order management
“From anywhere, to anywhere” is heavily dependent on information technology systems. Traceability is key, as the movement of every product is collected in as close to real time as possible. Stubbs emphasizes the importance of capturing the handoffs and chain of custody between suppliers and the DC, the DC and the store, the store and the truck, and the truck to the customer. “This will remain a very software-driven exercise,” he says. “Research suggests about 23% of businesses in western Europe and 27% in the United States are still using paper in some major processes. As we look at technologies to drive these processes faster and more accurately, paper will not even be an option going forward.”

Various “lite” versions of warehouse management system (WMS) might be used to manage the mini-DCs of the back room, he says. It can be as simple as subtracting inventory from the WMS as recorded from point of sale (POS) systems and adding what’s received in the back room. The store’s “lite” WMS might also relay the information to the DC to inform inventory upstream.

To support true distributed order management (DOM), systems must work together at a much higher level, looking across all platforms and all orders to determine the best place to direct an order for fulfillment. Chuck Fuerst, director of product strategy for HighJump Software, says this challenges a lot of store systems that were not necessarily designed for that sort of connectivity. Although it’s possible for the store to leverage a subset of the functionality of a warehouse’s WMS, Fuerst says it is more common to see a separate instance of a WMS in each place—even if the solution is from the same supplier.

“Early on, you need to establish goals and a focus,” Fuerst says. “Whether you are solely focused on cost to serve or never losing a sale, you will make different decisions. If the e-commerce engine is a priority, you might emphasize those orders as opposed to the storefront and accept lack of inventory on the shelf.”

In any case, Jensen says that no one should make a habit of transferring inventory between fulfillment locations. “Basing store deliveries on POS data and ensuring replenishment rises to meet that is obviously better than loading up an eastern location with something that doesn’t end up selling and needs to move west,” he says. “The cost of inventory is not a cost; the cost is in excess inventory. Once you understand the penalty of excess inventory, that’s where you can use that money differently by investing in solutions to move and position inventory economically.”

Fuerst estimates a 20% savings in inventory costs is not unreasonable and will prevent product relocation or markdowns. The e-commerce factor might also boost sales if a two-item order can be shipped in two parcels from separate locations, but a customer should be fully informed of what’s happening. Even if it’s cheap and on-time, if the customer doesn’t expect two packages it could impact their experience, Fuerst says.

“The most important thing is to say what you will do and do what you say,” Fuerst adds. “It’s an exciting space to be in, and a fairly significant trend. It’s been a while since I have seen something drive so much change in this industry, and we’re still on the front end of it. I don’t think anybody feels they have it figured out.”

Companies mentioned in this article
Dematic, dematic.com
HighJump Software, highjump.com
Honeywell Scanning and Mobility, honeywellaidc.com
Intelligrated, intelligrated.com
St. Onge Company, stonge.com
Voxware, voxware.com