Interview Transcript

Disclaimer: This interview is for informational purposes only and should not be relied upon as a basis for investment decisions. In Practise is an independent publisher and all opinions expressed by guests are solely their own opinions and do not reflect the opinion of In Practise.

Just to give you some background, I've been following Wayfair since they went public, and the entire strategy seemed interesting on paper. They raised capital to vertically integrate because the customer experience isn't ideal for selling big and bulky items online, so they developed their own delivery network and CastleGate warehouse network.

However, I still can't determine if this has constituted a meaningful competitive advantage for the company. I've broken this down into three parts. I took everything that's Asia consolidation plus ocean freight, that's one leg. The middle mile, which is the second leg, being the drayage from the port to the warehouse or consolidation centers. And finally the third leg, from the warehouse or consolidation center to the end customer, which is the last mile.

I dismissed the first leg because I don't see how Wayfair would have a competitive advantage in negotiating ocean freight compared to Amazon or other big-box retailers. If there is a competitive advantage, it should be between the middle mile and last mile. Since you are, if I understand correctly, a middle mile expert, I wanted to explore how they operate there to determine if they have a real advantage and if the capital they deployed since 2014 was worthwhile. That's the background.

So I'd like to teleport you back to your job at Wayfair and ask how it worked. Let's say you receive a 40-foot-high cube at a harbor full of 18 cubic feet mattresses. How does the drayage process work?

Part of why I joined Wayfair in 2019 was for a similar reason. They were just beginning to move containers on their own ocean carrier contracts into CastleGate facilities. From my conversations before joining them in 2019, Amazon was still very self-serving, and no one was really providing these logistics capabilities outside of their own bubble. Wayfair's vision was to figure it out for big and bulky items for their own purposes and then sell those capabilities externally, to big and bulky suppliers who want to sell freight through other channels.

This way, they could capture some revenue from sales that their sellers push through other channels. They could even extend beyond that and become a digital freight forwarder in the marketplace, competing with companies like Flexport if they could figure it out on the tech side and create an efficient product for their internal customers.

That was a compelling value proposition that attracted me. I aimed to figure out how to take those capabilities that serve the CastleGate facilities and apply them to the dropship warehouses, providing freight services for goods that may not even be sold on Wayfair. This was a major draw for me. It's an interesting value proposition of building a business within a business to serve not just Wayfair but also the freight market in general.

They could offer competitive ocean rates that sellers couldn't get on their own. Wayfair is in the same league as Amazon, Home Depot, or Overstock in securing competitive ocean rates. They can get comparable rates to those other big retailers. However, they would excel with the smaller sellers who are much smaller than Wayfair in aggregate. They can upcharge the ocean freight and sell it to a Wayfair supplier, creating a win-win situation. The Wayfair supplier gets a better rate than they could on their own, and Wayfair makes a margin on the container.

The value proposition is there if they can create a winning product and point it externally. The same applies to the warehouses. If they can make the CastleGate facilities multi-channel fulfillment centers, they could take in products and ship them, even if sold on Overstock or other sales channels.

Do you mean storing goods in CastleGate even if they weren't sold on Wayfair?

Correct. If they could tell a supplier, "We'll be your 3PL. We don't care if you sell it on Wayfair or not. We'll fulfill and ship that product for you," it would generate more interest from suppliers to store products in the CastleGate network. When I was there, it was very much a Wayfair-only fulfillment center, and suppliers were sometimes hesitant to store too much product because it would be captive to the Wayfair sales channel.

You mentioned that Wayfair could manage to get the same kind of ocean freight rates as Amazon or Home Depot. Is that true? I mean, Amazon is much larger than Wayfair if you consider the entire company, same with Home Depot. How would they secure similar rates?

In ocean freight, Wayfair's size puts them at the equivalent of a top 20 importer. The volume they bring in means the rates they get are similar to those of Walmart, Amazon, Home Depot, or Target. These big four might get slightly cheaper rates, but it's not a significant step change, like 10% or 20% lower than what the next tier of shippers would get.

So you're saying that once you reach this top 20 group, the difference in rates isn't significantly different?

Yes, there is a difference, but it's not significant enough to be a huge competitive advantage, especially if the intention is to add a margin to that freight and sell it as an NVO to a much smaller shipper.

How large is the difference between the number one and the number 20 in the top 20 importers?

I would estimate it at about $100 less per container that a company like Walmart might get on ocean freight to the West Coast, maybe a bit more to the East Coast, compared to a mid-sized shipper.

And $100 out of how much?

Out of $2,000, or $3,000.

$3,000 or $2,000.

Yes, ocean rates can be quite volatile, depending on the market. There are certainly advantages to size, but the advantage diminishes.

So when a container arrives, whether at an East or West Coast harbor, how do you decide which carrier to work with?

The drayage market is highly fragmented and specialized within trucking. Wayfair never wanted its own fleet of drayage trucks or drivers. We issued an RFP and secured third-party drayage providers to handle the size, scale, and variability of our volume. We outsourced this function but managed and oversaw operations, deciding whether to use trucker A or B. Rates and capabilities might differ, but external parties owned the assets, employed the drivers, and picked up the containers, regardless of their destination.

Can you name a couple of the big companies you were working with on the drayage side of things?

IMC is a big one, along with Cargomatic, and Schneider. There are also some regionally specific carriers we used in certain markets. For example, in Los Angeles, we worked with a company called Dray Alliance and another called HUDD, which is actually a Maersk subsidiary that does a lot of drayage for us.

On what basis did you decide to go with one carrier over another?

Are you asking about an RFP-type selection process or something else?

How did you choose one carrier versus another? Was it based on the size of the items being shipped, the price quoted, the cost, or the speed of delivery to the warehouse?

For any given port or warehouse, we typically had about three carriers supporting that lane, with a lane being the port and warehouse pair. We allocated our volume across those carriers based on their available capacity. Depending on market fluctuations and their other customers, they might have fewer truckers available one week than the next. We managed their capacity, the volume we anticipated, and the arrival cadence of that volume. For instance, having 100 containers arrive on the same ship at a port is more challenging to handle within a few days than if those 100 containers are spread out as 20 per day over five different ships. We assessed arrival patterns, available capacity, and also considered empty containers at the warehouse that needed to be returned to the port. We sought efficiencies by having a trucker complete a full round trip, taking a full load to the warehouse and returning an empty to the port to minimize costs. This was a continual daily operation to evaluate these variables and make the best decisions based on cost and reliability.

So you basically had the option of three carriers on each lane.

Yes, roughly. Some lanes had fewer, some had more. For dropship warehouses, we often dedicated just one trucker because the volume was so low that having two truckers would have been excessive.

When you say dedicated warehouses, do you mean supplier warehouses?

Yes, a supplier warehouse might receive two containers a month, something like that. You don't need more than one trucker to support that volume.

How would pricing work for that? How would you price those and bill the supplier?

Drayage is all point-to-point pricing, not mileage-based or zone-based like some trucking segments. It's from the Port of New Jersey to the warehouse address. We include this in our RFP, specifying the lanes we're running and the volume in those lanes. Each trucker calculates their own numbers, often depending on their location relative to these points. They provide a rate, and we select one, locking it in at, say, $500 per container. This includes pickup from the port, drop-off at the warehouse, and returning the empty container to the port. They give us a base rate for that round trip. On top of that fixed base rate, we add surcharges. The fuel surcharge is typically 30% to 35%, varying with the fuel market. There are daily charges for the chassis, around $30 per day. Sometimes there's storage if the trucker holds the container overnight. Additional fees may apply for one-way trips or taking the container to a different location before the warehouse. The core is the fixed base rate.

It doesn't depend on the density or weight of the freight. It's just on a per container basis.

Correct. If it's an overweight container, special roads and a special chassis are needed, incurring extra charges as a surcharge on top of the base rate. A Hazmat container, rare for Wayfair, incurs extra costs due to special permitting or a driver with specific credentials.

But everything that's not overweight is priced on a container basis.

Yes.

And it doesn't depend on what's inside the container, as long as it's not overweight.

Correct. Even if you ship a 20-foot container instead of a 40-foot one, the trucker charges the same because the work is the same regardless of container size.

That's helpful. So, you get a quote from the trucker. What would be the average markup to the customer, including fuel and storage surcharges? For example, if the trucker quotes $100 for the container from the port to the warehouse, how much is the average markup? Can you break this down between fuel, storage, etc.?

Let me see if I can come up with a good example for you. I'm drawing from my current experience here. For instance, I'm in Savannah, which is comparable to where Wayfair's warehouse is located in the Savannah region. The base plus the fuel surcharge versus everything else, like landed, is roughly 1/3. So, it's like a 30% to 35% markup with all the additional fees associated with the move.

So, a 35% markup that you would bill the customer?

Yes, if it's going to a dropship warehouse, that would be passed through to the supplier. However, it's not a profit markup; it's a markup of just surcharges on top of that.

What I'm trying to understand is the difference between everything that the drayage carrier bills you as Wayfair and what you would bill a supplier.

We experimented with this a bit. We typically added about 10%. Sometimes we tried to add a little more, sometimes we negotiated it down, but it was challenging. We found it difficult to upcharge drayage and pass that through to suppliers. They would often say they could find a cheaper rate elsewhere.

They would accuse us of overcharging because they could find cheaper drayage elsewhere. Our value proposition was that we handled the ocean and drayage together as one, taking all of that off their plate. Usually, they saved enough on the ocean that a slightly more expensive drayage price was acceptable. We aimed to make the whole package a beneficial value proposition for them.

How would this 10% markup work? So, you would take the total cost that the carrier gives you with all the surcharges, whether it's fuel, chassis, storage, and add a 10% markup on top of this on average?

We typically add the 10% to the base rate.

What do you mean?

Let's say a hypothetical base rate is $200, and then the fuel surcharges might be $50 or $75, and other fees might be another $50 or $75. We would only add the 10% to that $200 base rate. Then we would send that straight through.

You would only make money by marking up the base rate, but all the other surcharges and extra fees would be passed along to the customer without an additional surcharge.

Yes, because it's very transparent. When we bill them, we have to provide line item details on some of these things. Adding a markup on a surcharge sometimes looked bad and, in other cases, was not allowed by the FMC.

And how does this compare with others, like Amazon for example?

I don't know much about Amazon's trade program, so I can't speak to that. I would hesitate to say that Amazon is able to get competitive rates on drayage compared to anyone, really. There is some benefit that accrues to a large importer because there's volume and consistency that are attractive to a trucker.

However, as a large or even mid-size importer, you can't rely on a million independent owner-operators to support your freight. You need tier one trucking providers with the service levels, technological capabilities, and capacity to support your volume, which usually comes at a premium.

While you might get a good discount on that premium rate, you're still using an IMC, Schneider, or Cargomatic, some of the largest drayage providers in the country, instead of Joe Schmo's Trucking, who just happens to have the port credentials and will get your container for next to nothing.

We would hear from suppliers who say they know someone who could do it cheaper, but that type of trucker can't fit in our portfolio as a large importer. There's a challenge to being competitive in the drayage market because your requirements become much more difficult to fulfill.

Out of 100 suppliers who would ship containers using Wayfair's ocean freight program, how many of them would decide to detach from you when the container arrives at the port and find their own drayage company?

We mandated that they could not. We sold it as a package. They needed to do both ocean and drayage to be in the program. Part of the reason for that was the handoff became a huge space for errors and issues.

Why?

If they have their own trucker and we are providing the ocean service, it becomes our responsibility to communicate with their trucker, ensuring they get arrival and customs clearance information and all the particulars needed to get that container. This becomes a burden for whoever's handling the ocean piece. We didn't want to work with a hundred random supplier-chosen truckers. We wanted to work with a handful of vetted, high-quality truckers with whom we had relationships and technological integrations to minimize errors.

When something goes wrong in a handoff and the container gets stuck at the port, it can incur expensive demurrage fees that add up quickly. Then everyone starts pointing fingers; suppliers at us, us at the trucker, and the trucker at the supplier. No one wants to own those fees. All that debate goes away if we at Wayfair handle both legs. We tell them not to worry, don't get involved. We'll make sure it gets there, and if there are fees, that's on us.

That makes sense. If they want to use Wayfair's ocean freight service, they must also use Wayfair's drayage. What if you have to dray containers to a CastleGate facility? Would it be cheaper because you would be sending more containers?

Yes, the truckers would give us better rates to our own facility than to a supplier's facility because there's more density there. They can keep their truckers busy going back and forth with full and empty containers.

How much better would the discount be, approximately?

I would say probably 5% to 10% cheaper. It's not a huge markup unless it's a particularly difficult situation. For example, we had a supplier in downtown Philadelphia with no parking spaces, which resulted in a much higher rate than a warehouse next door would have.

If a supplier wants to store goods in CastleGate, you would charge them for the drayage, right?

Yes.

That becomes interesting because if you charge them for the drayage, a 10% premium on the base, and you get a discount from the carrier.

Yes, there was a different pricing model for coming into the CastleGate network compared to their own dropship warehouse.

What was the difference?

Since the CastleGate network is specific to goods sold on Wayfair, there was high assurance that all the product in that container is Wayfair product. From a pricing standpoint, there was a willingness to charge the freight either flat or even subsidize it a little to encourage goods into the Wayfair network. When shipping goods into the dropship network, we don't know how much will ultimately be sold on Wayfair. So, there was a strict policy not to subsidize freight going into the dropship network because we didn't want to subsidize competitor sales.

Not only would you not subsidize it, you would even mark up the base.

Correct. We are marking up the base because it's meant to be a revenue stream for Wayfair to provide this freight service for sellers into their dropship warehouses.

Anecdotally, from my conversations with suppliers, about 10% to 20% of the goods moved into their dropship facilities ended up being Wayfair product. A significant amount was sold elsewhere.

Let's take an example here. I'm a supplier, and I have two containers. One is going from the harbor to my own warehouse using the Wayfair drayage service, and the other one is going to CastleGate. Let's assume the base rate is one hundred. If it goes to my own warehouse, as a supplier, you're going to mark this up by 10%, so it will cost me 110 plus all the surcharges that you will pass along at no markup. What if I send the container to CastleGate? Compared to this 110 of base plus markup, how much would it cost me on average if you decide to subsidize it because I'm sending everything to CastleGate?

It would probably be pretty close to just the flat one hundred with the surcharges, in most cases, unless they were a strategic supplier or there was some negotiating push on the Wayfair merchandising side. They might say, for example, if we drop our price by 5%, they'll give us an extra 5% volume into the Wayfair network. Conversations like that, which are outside of my purview, would happen on the merchandising side to encourage higher penetration into the CastleGate network with strategic suppliers. It wasn't available to everybody. There's a list of strategic suppliers that received more attention from Wayfair because of their product mix.

So there is an incentive for the supplier there.

Yes, there is an incentive to move into the CastleGate warehouse. It's important to note the context of when I was there, which was during Covid and the post-Covid times. During that period, manufacturing was difficult, so the supply of goods was challenging, and sales for home goods were sometimes sky-high. Everyone was trying to hedge their bets and be a little bit selfish at the same time. Suppliers wanted more control in their own warehouses to give them optionality on sales channels, while Wayfair wanted more inventory in their network for selection and cost-effectiveness within the CastleGate network. There was a strategic push and pull where Wayfair was constantly trying to encourage more inventory into CastleGate, while suppliers had the opposite perspective during those challenging months and years.

In terms of the average time it would take for me as a supplier to get the container from the harbor to my warehouse versus it arriving at CastleGate, would it arrive at CastleGate quicker?

Once containers arrive at a port, they have four free days before they start to incur fees. As soon as it docks, you'll arrange for a trucker to pick it up before that four-day window expires. Typically, the warehouses are about 40 to 60 miles away from a port, which is the typical range for drayage. It's just a few-hour trip for a trucker to pick up the container and drop it off. It's not a multi-day journey.

Did you see a difference in the type of goods for which the supplier preferred to store them in their own warehouse versus sending them to CastleGate in terms of the size or type of goods?

I don't think so. At Wayfair, they tended to have a large parcel facility and a small parcel facility and were willing to take hot tubs or gazebos, some pretty large items, into their network. My understanding was that it was much more driven by the sales characteristics of each product. Wayfair wanted to prioritize higher-turn, higher-margin, or higher-damage-rate products into the CastleGate network to maximize the cost-effectiveness of their buildings.

But it's also largely the supplier's decision to decide which type of goods they want to store in their own warehouses. Did you see a tendency for suppliers to prefer certain storage channels for their goods?

No, I really couldn't. Aside from knowing what types of items Wayfair wanted, I mostly just saw containers moving. I didn't typically know or draw patterns between what's in the box, so to speak.

When it comes to a supplier's warehouse, we intentionally kept an arm's length from some of the product details. If we tried to get too close, they would be resistant to using our services because they saw Wayfair somewhat as a threat. They were afraid of being "Amazoned," if you will, if they revealed too much information about their goods or factory addresses.

I think I have a good sense of how containers go from harbors to either the CastleGate warehouse or a supplier warehouse. But regarding the middle mile, the journey from a CastleGate facility to a delivery station, if I understood correctly, Wayfair would manage the middle mile for 90% of large parcel volume.

I don't know much about that. I don't know much about the middle mile network, but you are correct in that Wayfair did have a fleet of trucks they used to run goods into the delivery network. However, I wasn't involved in that part of the logistics puzzle.

Basically, your involvement stopped once the container reached either the CastleGate warehouse or the supplier warehouse.

Yes, exactly.

How would Wayfair decide which CastleGate warehouse to place a good into when it arrives at the port?

I wasn't deeply involved in this process, so I don't know the specifics of the logic. However, there was a regional fulfillment strategy aimed at being as close to the end customer as possible with the stocking strategy. Some logic looked at sales patterns regionally. For instance, in the winter, we wouldn't send patio sets to the Northeast as frequently as to the Los Angeles region. There was logic that considered who was buying what and when throughout the year. Freight recommendations would then be made to ship specific collections of SKUs to designated destinations.

If a supplier used Wayfair for drayage services, you would have relationships with carriers, correct? If I had to handle it on my own without Wayfair, would it be more expensive for me to manage drayage?

It could be comparable, or it might even be cheaper. As a small shipper, you might not need the sophistication that Wayfair requires for their drayage provider. You could potentially go to a brokerage, put your load out there, and get a very cheap rate or a one-off drayage from someone you might not need to contact again.

Did you notice a difference in damage rates when using less sophisticated carriers versus tier-one carriers for drayage?

Typically, no. The drayage leg is usually not where damage occurs. It's also challenging to determine if and when damage happens during that stage because the drayage leg is a short over-the-road move at the end of the cycle. The container is sealed, and you would know if the seal is broken before it reaches the warehouse. Generally, goods are packed to withstand a journey from a factory in China to the US. Once the container is sealed at the factory, it is trucked to a port in China, moves through the port onto a vessel, crosses the ocean to the US, and is then trucked to the warehouse. If any damage occurs, it's difficult to pinpoint at which stage of the journey it happened.

So, if I send a container and want to split it between CastleGate and my own warehouse, how would you handle that? For example, if I ship a container full of mattresses from Asia, but I want half to go to CastleGate and the other half to my own warehouse, what would I do?

We were not able to facilitate that.

What's the solution for me as a supplier?

What we could do is split it within CastleGate, sending half to the West Coast and half to the East Coast, but both would end up at a CastleGate facility. While I was there, we didn't progress enough to split a container between CastleGate and dropship networks due to pricing complexities. If they use a cheap rate to get into CastleGate and then decide to send some to their own warehouses, it involves a different rate sheet. There's some complexity to it.

So, I have to choose if the container goes to CastleGate or my own warehouse. I couldn't split it between two warehouses.

It depends. Some suppliers have warehouses close to each other, and you could use the same trucker to stop off at one warehouse, unload half the container, and then take it to another warehouse to unload the rest. But if we wanted to split the container and send it to two different geographies, we weren't able to do that type of move, which we call LCL or less than container load.

I guess it wouldn't make sense because the trucker would have an empty truck for the second leg and would charge extra.

Yes, they would charge extra for that. It would be doable if the trucker could make a short run to different warehouses on the same journey. But that was a future evolution we hadn't built out yet when I was there, as it adds a lot of complexity to the operation. We were trying to keep it simple.

When, for example, Wayfair receives tens of containers per day going to different CastleGate warehouses, how do you manage that? Let me give you a tangible example. You have one container full of mattresses and another full of coffee tables from two different suppliers, and you want to dispatch these to different CastleGate warehouses. How would you handle this at the harbor, considering you mentioned you wouldn't touch the containers before?

When the containers are arranged at the origin, they have a destination attached to them in the paperwork. We know before the container is packed at the factory that it's going to a specific destination. We track it with a container number all the way to its destination.

I see. So the mix of products in the container is determined before it even gets into the ocean?

Yes, there's a manifest listing the products, and the supplier provides this to a customs broker for the commercial invoice and necessary documentation. This is processed while the container is on the water. We know when customs clearance is received, the container number, and its destination. We then inform the trucker, who picks up the specific container and takes it to the designated destination.

So, at the harbor in the US, a container is never opened, correct? Even if it goes to different CastleGate warehouses, the same container is never split between two warehouses. It is assembled back in Asia and has a specific destination at a CastleGate warehouse.

Yes, that is primarily true. We established two facilities, one in Los Angeles and one in Savannah, for a third-party deconsolidation process. We could send containers to these centers, open them, and perform cross-docking. They would distribute goods to multiple destinations. This was a CastleGate-specific operation and did not involve supplier warehouses at the time. It allowed us to send a full container to Savannah and then split it, sending trucks to New Jersey, Florida, and Kentucky, for example.

But the cross-docking was only for goods going to CastleGate, correct? You would never cross-dock containers going directly to a supplier warehouse?

Correct, at the time we launched it, that was true. There were aspirations to eventually reach that point, but I never saw it happen. Since I left, I've heard rumors that the deconsolidation process has been shut down, so I'm not sure if it still exists.

It has been shut down for CastleGate, really.

Yes, yes.

Why?

It was a bit of a pilot to see if we could better support this regional stocking strategy in a cost-effective way by funneling goods through the deconsolidation center instead of sending full containers to every single region. When I was there, we were learning and did it for about a year before I left, trying to see if it was cost-effective. I guess it wasn't, or it was challenging enough that they decided not to continue.

The alternative is to send it to a warehouse and use Wayfair labor to break the container apart. Then they could potentially use the middle mile network to get it to the correct warehouse. This adds complexity and throughput to the middle mile network, but if they can support it, it could be a more cost-effective alternative.

Either that or have containers that are assembled at the origin.

Part of it is what you're talking about, and part of it was smaller factories not shipping a whole container worth of product regularly. We could combine them and send them at once. But those decisions need to be made months before a container arrives in the US, like one to two months. If we can get the container to the US before deciding the destination, theoretically, we might make a better stocking decision because we're closer to the endpoint. So there are some trade-offs there.