A senior executive from a large global manufacturer came to speak in my MBA class this week, and discussed the major shifts in global business that was occurring due to the emergence of omni-channel capabilities in the supply chain. His insights and presence were most appreciated by everyone, especially given the snowfall we were experiencing in Raleigh last night!
He began with a brief history, noting that for many years legacy businesses had a single channel, and then multi-channels came along – which included multiple touch points in the channel. Retailers channel knowledge existed in technical and functional silos – people had separate silos of activity in this model. This led to Omni-Channel sales. This view of the supply chain is focused primarily on how we get inventory to the customer. This is the major crux of the problem. He pointed out that the customer doesn’t care which division owns the inventory – they just want the product whenever and wherever they desire. Retailers have established a single view of the customer, but are still operating largely in functional silos.
This is rapidly changing. For instance, Amazon knows that if you order something from a website, and it comes in three packages, you would expect it to be one transaction. But if Amazon tells you when the packages will be delivered, on which day, the customer now has a predictive delivery schedule, and doesn’t care if they come in three packages even though it is one transaction. Some people even want to be told what two hour window the delivery will fall into.
The Omni-Channel capability is associated with getting product to the customer in any possible fashion, that is convenient for the customer. This is largely driven by the emergence of e-commerce, and in the process, is turning traditional business models on their heads. To give you an idea of what this is doing, the classic retail metrics, “same store sales” and “sales per square foot”, are no longer being reported by Macy’s department stores. Why? Because Macy’s is now fulfilling orders from stores….so store sales and inventory at stores don’t matter. And so they aren’t reporting it. Think about this scenario: you walk into a Macy’s, find a popular shirt you like, and they just sold out of the last small (your size). But the sales person tells you, “I can find you a small at the store in Phoenix Arizona, and will have them ship it to you free of charge…you should get it in two days.” This causes additional problems, however. The sales person who wants that commission is happy – but the one in Phoenix who has to ship what is perhaps the last small shirt in the store, loses on that potential sale. And this causes all kinds of mischievous sales person behaviors!
Deloitte has developed an Omni-Channel framework that discusses the many different capabilities required to launch a good omni-channel business. Some of the issues impacted include manufacturing, catalog management, BI and analytics, pricing, order management, customer service, fulfillment and transportation, and many, many others. A few of these were discussed in our class last night by this executive.
Catalog management – Apparel manufacturers sell to department stores, sporting goods and specialty shops, Macy’s, Dillard’s, Penny’s, Herrods, and many other chains. In many cases their channels are competing with one another. Online sales allow consumers to compare prices between channels, however, the information about that product is uploaded by a “catalog manager”, who provides the data associated with that particular product. Too often, the information about the very same product, appearing on different websites, may differ significantly! For example, one may say machine washable, the other hand wash, one says made in USA, the other Made in Vietnam – for the same product! Standardizing product descriptions is a real headache for web content describing products. The non-profit organization GS1 is trying to figure out how to standardize all this data. For example, a simple description like a “boot cut jean” may mean many different things to different manufacturers! This becomes very complicated, because some jean manufacturers have their own particular fashion issues that they attach meaning to that they consider unique and proprietary! Even the way companies measure the inseam on a jean can be different. Companies are trying to work together to define even standard sizes like SM in men’s vs. SM in a children’s department – and tables that define what a SMALL really means are required to drive this type of standardization!
Checkout – Many companies are starting to use a mobile computing device for checkout. This works well in the Apple store, but can be complicated for apparel. There may be a hangar, and there may be a box or a tag that goes with it, as well as other issues. For many reasons, mobile checkout in apparel doesn’t work as well, and have to think about it differently. And security is an issue – the Target data breach spilled information on as many 70 million customers to thieves. These criminals dressed as technicians and put in card skimmers for every card reader in Target stores!
Order Management Order management is about fulfillment of orders. Amazon is now bigger than Walmart if you look at the total volume of transactions, and so Order management is a big issue. Amazon has a fee based structure where they get X% of the cost of stuff they sell – but they never own the inventory. If I make 5% of $50 – my annual revenue is $2.5. Walmart’s revenue is on the $50 side of the equation – so if you grossed up Amazon’s revenue – they are bigger – even though they don’t make as much money.
Ali Baba is a different market in China – they are what’s known as a “marketplace” that takes demand and gives it to vendors who fulfills it on their behalf. This graying of the boundaries between order taking and order management is creating a lot of complexity and confusion. Who is responsible for fulfilling the transaction – and how do I capture demand and hand it to somebody? To have this capability, involves having the technology integration to make it look like my event…not somebody else’s event.
As noted earlier, order fulfillment is now using retail stores as warehouses. If we run out of shirts, the sales person can check on the system to see if it is in a store, call up the sales person at that store, and ask them to send it to the customer. The store is a node in the network. The problem is that stores don’t do order fulfillment very well. Sales people like to be working with customers, not shipping stuff in boxes. Also I need the intelligence in the network to know where inventory is. But how do I incentive people to fulfill an order from a store if the origin of the order is an online website? Hmmmm…
Capacity for order fulfillment is a massive question in the omnichannel environment. On December 22, 2013, everyone launched promotions, and people clicked on orders that were supposed to ship and be delivered by Christmas. We all know what happened – the UPS network failed. In 2014, UPS bulked up on capacity and people, anticipating the huge Christmas rush. However, the promotions didn’t show up as promised. Everyone predicted Christmas would break UPS again, but the promotions didn’t come, and UPS had added so much capacity in their network, that their costs soared and they missed their profit and earning projections, which Wall Street dinged them severely for! A similar metaphor is: Do you build a church that can fit people for Easter Sunday or for a regular church day? (Easter Sunday is when everyone who doesn’t normally show up for church shows up!) UPS built a network for Easter Sunday, and no one showed up….
Inventory and Warehouse data
If you are into sustainability, check out Patagonia’s website and look up footprint chronicles. Patagonia has made a commitment to be more environmentally conscious, and have carbon footprint on shoeboxes. If you look for an item on their website, the instant they are out of stock they tell you where to go buy it – from Zappos or Travel Country for instance. Essentially they are giving up the sale and handing it to Zappos, who can tell you the exact number of items in inventory, and shows this on the Patagonia web catalog! To enable this Patagonia is getting real-time inventory data from Zappos and Patagonia is essentiailly giving away a sale to another channel….so a natural question is – will customers go over to Zappos first next time?
Promotions by their very nature drive huge variability in sales – up to 25X of a standard sales rate during a promotion. Companies cannot provide the same service level at a 25X sales level as they can at an X level of sales. So the Easter Sunday question comes up again - will I build a network to service a 25X promotion? UPS tried to do it and got killed. As supply chain professionals ponder how to design their network, omni channel and internet shopping is changing every prior rules and algorithm on network design ever known to man! If you buy things online, you might see a summer sale, back to school sale Private Sale, Thanksgiving Event, Holiday Sale, etc. So if you buy something, and you call up the retailer and ask for a lower price when it goes on sale a week later …the cost of return and re-fulfillment is huge – the retailer in most cases will simply give you the money (the difference) in the sale price. Most people won’t go back and to the same store and ask for the refund, but omni-channel retailing allows you to simply sit at home, pick up the phone, and get the refund. Which sort of defeats the purpose of promotions as a sales incentive tool..
Malltail is a company in Korea. Because tariffs are so high in Korea, this individual built an entire business on people in Korea ordering anything in the US, and having it shipped to an address in the US. This service will then repackage the product, and send it directly to Korea. Believe it or not, Samsung TV’s are cheaper to buy on Amazon and have shipped to Korea – even though they are made in Korea. But personal import exemption on duty is so high, and certain states in the US don’t charge sales tax that there is an entire business based on avoiding regulatory fines and sending product to people in Korea who buy TV’s and clothes.
RFID One of the big promises of RFID is that it improves stock accuracy and reduces item degradation. RFID tags are still expensive, up to 10 cents each, and so it is still a dubious investment. But they do hold the promise of being able to provide more efficient inventory tracking and auditing capabilities in a store or location, and thus can make people more confident in systems that show that an item is indeed on the shelf somewhere in the location. It is a little known fact that Macy’s has $1B of inventory that has only 1 item in a number of stores. But people don’t have confidence that this inventory actually exists, because of “item degradation”, or lack of confidence in the data. Macy’s can unlock this inventory and make it saleable with RFID – otherwise they would have no confidence on having that inventory on the shelf.
Inventory Integration with Wholesale Partners – If Samsung is selling a television, and show that it is in-stock at one of a number of local Best Buy stores, then what is the relationship that transpires? Will Samsung will get a kickback from them for the sale? To enable this requires visibility that Best Buy made the sale because of a reference coming from Samsung’s website – but how does this occur?
The ultimate outcome for Omnichannel sales is the integration and merging of the Brand (the Nirvana of this view) where it is about the entire brand experience, not a channel within a brand. Retailers leverage their single view of the customer in coordinated and strategic ways. The integration that is happening between channels in the market is incredible – and it has all happened in the last two years. And many of the problems have still to be worked out, as you can tell by this short set of descriptions….more to come!