The Art of Retail II

It’s so easy to take retail for granted. Whether it’s opening up Facebook Marketplace, going to Costco, or ordering online at Amazon, it’s so easy to buy things today. Buying things is quite magical. We can buy anything from new pants to thousand-dollar Herman Miller chairs with pieces of paper and bits of plastic.

Reality, however, often has more detail than first imagined. In part I, I dived into the larger macro trends within retail, particularly the dance between unlocking new supply and increasing potential demand. In part II, we’re talking specifically about supply. Supply is a subject that I have a deep history with, so much that I’m currently founding a startup in the space (long shot but does anyone know truck drivers in the US? would love to talk with them!). As a result, when I think of retail, looking at it from the supply angle feels instinctive to me. But looking online, I found very few down-to-earth articles about supply within retail. So I’ve chronicled the impacts of the supply chain on retailers in this post and explored the long-term impacts of having a good supply chain.


In life and in retail, demand is how companies get big, supply is how companies stay alive.

Demand in retail is simple. The tried and true formula for generating demand is cheapest goods + widest selection + best environment.

Walmart is the biggest discount/super-center chain around because it wins on finding the cheapest goods, Amazon wins on widest selection, and the plethora of other retailers from Target to Best Buy to Home Depot provide the best environment for their niche. And over the years, retailers have been able to boil this demand formula down to an encompassing variable, shelf space. More shelf space means more inventory and wider selection, which in turn means higher overall sales velocity, which means better terms with suppliers, which then means cheaper goods to the end consumer. As long as the retailers are efficient, the company with more shelf space will consume exponentially more demand and profits. In essence, retailing is a game best played by giants.

On the other side of this equation is supply, the ability for a retailer to support their demand without going broke. The prerequisite for cheap prices and wide selections is a supply chain that can find suppliers, squeeze margins, and deliver without delays. In general, this part of retailing is left to MBAs, industry experts, and a whole lot of late nights. For retailers, and the rest of the world, supply chains are a necessary evil. Until we invent teleportation, we need trucks, trains, ships, and planes to deliver goods from one location to another. Thankfully, for the past few centuries, we’ve figured out for the most part how to get the right goods to the right place at the right time.

It’s hard to be excited about truck warehouse idle time decreasing from 4 hours to 1 hour and much easier to focus on flashy sales tactics like Prime Day or Walmart. And so, everyone has focused their attention on the demand for the past few decades.

These days, supply is finally getting its moment in the light. With shortages hitting everything from cars to iPhones to even toilet paper, we’re finally looking at our aging supply chain infrastructure. However, despite that, much of the current conversation about supply chains is terribly wrong. The narrative shifts from shoppers hoarding hand sanitizer to Ever Given stuck the Suez canal to manufacturers cutting back on orders and closing factories in Asia to infuriating bureaucratic policies stifling supply chains. While these are very real events, the true illness in supply is the focus on costs at the expense of everything else.


From a business 101 standpoint, higher supply chain costs mean higher prices in stores. And higher prices mean fewer shoppers. As a result, almost every supply chain optimizes to reduce costs through finding cheaper suppliers, more efficient schedules, and higher levels of automation. In fact, it’s been an arms race between retailers for the past few decades to reduce prices. Everything from material to workmanship to worker benefits was sacrificed in favor of price. And shoppers rewarded this behavior. Walmart’s incredibly low prices meant droves of shoppers flocked to their stores from miles around and paid little heed to the lower quality, worse customer service, or the sticker that said made in China.

Much of the decrease in costs has also been from vertical integration. Walmart handles everything from load pickup, storage at warehouses, delivery to stores, and the eventual sales process. Amazon takes this model and transports it online with the Fulfilled by Amazon service, they also go further and even produce the goods themselves with the Amazon Basics line.

To reduce further costs, retailers (and shoppers) often go with the cheapest suppliers. Taking advantage of labor arbitrage, these suppliers cluster in a handful of geographical locations, in particular China and Asian countries. From there, items are shipped in bulk to the US, and retailers distribute through a massive trucking network, and store, pack, deliver, stock the goods.

In isolation, a tightly integrated supply chain isn’t necessarily a bad thing. Companies increase profits and consumers benefit from lower prices. But problems arise when one component of this tightly choreographed dance fails and brings down everything with it. At the micro-level, companies with tighter supply chains fall into the classic innovator’s dilemma and die with each new paradigm of retail.

But before we dive deeper into how fragile supply chains are, let’s first look at the broader strategies employed by each company. While on the surface, they all use trucks and have warehouses, retailers approach supply from radically different perspectives.

Walmart has a hub and spoke model where goods flow from distribution centers (hubs) to individual stores (spokes). Their supply chain starts with Walmart merchants building direct relationships with manufacturers or distributors and ordering goods in bulk. Suppliers then access Walmart’svendor-managed inventory system and can adjust production relative to their sell-through rates while Walmart provides the transportation for vendors to ship goods to Walmart distribution centers with Walmart’s own private fleet. From there, Walmart will arrange delivery to individual stores based on their own internal data. While efficient, the tradeoff of this approach is that Walmart is forced to work with vendors that have the capability to provide enough supply for at least one distribution center which roughly translates to 90-100 stores of demand. They lose the flexibility to be opportunistic at a store level.

Costco, on the other hand, runs every store as a warehouse. They load through back of house and unpack goods early each morning. This saves costs both on item storage and on-premise logistics. The natural result of this model has been cheap goods, big bundles, high inventory turn rates. More importantly, it also gives Costco a unique purchasing ability where each store has the flexibility to stock niche, local items without requiring massive supply-side guarantees from the vendor. However, it reduces the margins on each item slightly compared to Walmart.

Target uses a pick and pack system where shipments from fulfillment centers are often multiple less than truckload crammed into one trailer. And the results of this are higher priced goods, lower inventory turn, and smaller store footprints.

Amazon looks a bit closer to Walmart where they staff massive fulfillment centers just outside of major cities and have literal droves of delivery vans that fan out across cities. In lower densities areas, they employ third parties like UPS to deliver goods, increasing delivery costs but decreasing operational scope.

These massively different supply strategies stem from the consumers that these companies service. Walmart started with Walton’s discovery that discounting by half would lead to at least 3x in sales. As a result, Walmart began on a multi-decade strategy of continuously undercutting competitors on pricing. Their supply strategy gradually formed into the hub and spoke model to accommodate the high volume of goods moving through each store. And they came to where they are today by simply out executing their competitors in the discount/superstore arena.

But while strong and wildly profitable, Walmart’s supply strategy isn’t without its tradeoffs.

With a supply chain hyper-optimized for massive volume and variety, Walmart’s demand strategy becomes fundamentally incompatible with higher-margin businesses. For example, if Walmart suddenly decided to convert their stores to Whole Foods style higher-end products, their supply chain would be hemorrhaging money from inefficiencies and idle time. Obviously, Walmart isn’t going to shift its strategy so drastically. However, taken to a less extensive scale, this explains why Walmart lags so far behind Costco in the wholesaling retail model and behind Amazon in online shopping.

Let’s now return to our original point. A hyper-optimized supply chain loses the flexibility for change. I’ll dive into the intricacies of change with the example of Walmart, a retailer whose supply chains have been heralded as textbook perfect for most of the past few decades and as a result, severely hindered their early e-commerce efforts.

First and most obviously, e-commerce doesn’t solve the last mile challenge which is basically from the warehouse/store to a customer’s doorstep. For current Walmart stores, their customers solve the last mile challenge. Every time I drive to a Walmart, I’m saving them significant amounts of time that would have otherwise been used to stock, pack, and deliver the goods. To even get started in e-commerce, Walmart would need to build out a distribution system where they can ship goods to doorsteps. Today, Walmart solves this by partnering with an existing logistics company, in particular FedEx and USPS, to fulfill that capability.

Second, let’s talk about warehouse management. Before the order can be delivered, Walmart needs to stock and package the order. There are two options here. In the first option, Walmart can attribute the associated cost and revenue to the local store. The store can either have dedicated employees picking out items from shelves, Instacart style, or have some employees in the back picking from unpacked bulk loads. Both are rather messy solutions given that stores today are optimized for the shopper experience. However, the model makes sense long-term given that Walmart can leverage the proximity of their stores to consumers and minimize delivery costs. The other option here is to run packaging at the warehouse level. This makes sense in that Walmart can now offer a wider selection online, with the assurance that the package is actually in stock and delivered within a reasonable timeframe. The downside here is that Walmart now needs to staff exponentially more workers per warehouse and figure out safety and operations logistics while at the same time, paying for higher delivery costs. Already, this e-commerce stuff is a headache.

When Walmart first launched its online segment, they opted to go down a third route. They carved out space from their store-serving warehouses (these are generally Walmart Supercenters) and employed third-party operators like Ingram to handle the load. The only other distribution method was a single online-focused distribution center in Carrollton, Georgia. As you can imagine, this didn’t help in their fight against Amazon. For reference, at around the same time period, Amazon had 40 U.S. warehouses within 35 miles of major cities.

Third, let’s talk about inventory management. Walmart has an incredibly sophisticated inventory management system where suppliers can view sell-through rates at stores and prepare production accordingly. An online store actually integrates extremely well here. The problem however arises for individual stores. For example, if an item is low in supply, should the in-store shoppers be prioritized, leaving e-commerce customers waiting on delays? Or should the e-commerce orders get fulfilled first, potentially diluting consumer sentiment of in-store shopping at Walmart? Or, if Walmart packages at the distribution center layer, how do workers know to input fulfilled sales into the system? But of all the problems before, this thankfully is only a small headache. With the current strategy of high inventory turnover, Walmart already has the infrastructure in place for shifting inventory management to accommodate online orders.

Fourth, let’s talk about the customer experience. In particular, returns. By establishing themselves as middlemen between customers and suppliers, Walmart and other retailers exercise an incredible amount of power over suppliers. Generally, when an item is returned, Walmart issues a chargeback to the supplier for Walmart’s cost. In other words, Walmart is paying nothing for the item. Further, Walmart might even add a processing fee to be paid by the supplier for Walmart’s troubles and also resells the returned merchandise to bulk distributors and take a profit there. It’s a great model. Unfortunately, this model works out horribly for online orders. With free return shipping, Walmart now needs to pay for expensive last-mile shipping again. On the other hand, offering only in-store returns risks the customer relationship. In sum, a crappy deal all around.

Now that we’ve finished listing out the larger problems, let’s talk numbers. On average, the logistics cost behind each item in a physical Walmart store is 3% of price. The logistics cost for online orders was and still remains around roughly 15% of price. As an executive at Walmart, it then becomes easy to dismiss online commerce and deprioritize online experiences in favor of offline growth. And as you probably know, that wasn’t a great decision.

Obviously, pivoting business models when you’re at the size of Walmart is always a tough game, and moving into online sales involves much more than just supply. But also true is the fact that Walmart wasted one of their best chances to dominate online retail, primarily due to a self-inflicted wound from their supply chains.

And that brings me to the less expressed tenet of supply chains today, everything is as everything was. We’re still using the same systems that were put in place decades ago. And as a result, supply chains contain a bizarre mix of futuristic robots and antiquated relics. The innovations have mostly occurred on a micro-scale like automatic picking robots, drones for delivery, and multi-directional conveyor belts. While useful, they’ve only improved the warehouse/packing component of the supply chain. There’s still the transportation, delivery, and stocking components that have looked exactly the same since the 80s. Actually, the trucks are shinier today. But overall, supply leaders have massaged existing systems to support new paradigms rather than build new supply chains for new consumption mediums.

In fact, this is precisely the premise for the innovator’s dilemma. A startup takes a lower-margin business that gets written off as niche and low potential. The startup is able to build and figure out the kinks without competition for a few years as the space develops. With everything in place, the startup can now rocket-ship upwards and incumbents struggle to rival the infrastructure of the startup.

In retail, there are quite a few spaces that fit the bill. The major one that we’re in today is online shopping,=; just a decade ago, no one trusted the internet enough to put in their credit card information. But on the horizon, we have efforts like Live Shopping where creators marathon through products and convince viewers to purchase them. We also have entirely new models of distribution with customizable vending machines like Amazon lockers and concierge boxes. Or even drones and delivery robots that ship goods to your doorstep, no human required.

In every single one of these areas, traditional retailers have been incredibly slow to react and in most cases, it’s not because the technology isn’t there, or they don’t want to, it’s because their supply chains are holding them back. It’s never been a worse time to be a retailer, it’s also never been a better time to be a retailer.

(Shameless plug again for my startup: Excited about making trucking more efficient? Message me!)