COVID-19 & the Performance of Last Mile Delivery Networks: Much Ado About Nothing?
co-authored with Arun Rao
In my previous post on changes in the retail landscape post COVID-19, I made several predictions about how retail would evolve going forward. A key prediction of mine was that consumers would begin to value reliability of shipments in e-commerce over speed of delivery. The trigger event being COVID-19, which led to an unprecedented surge in e-commerce in many product categories such as groceries, at-home entertainment, in-home fitness, and others.
Some of this surge was driven by consumer panic over potential shutdowns, particularly for essential products. But the closure of brick and mortar stores meant that for many non-essential items, going online was the only option that consumers could avail of. The most startling example here was the closure of all Macy’s stores and the furloughing of their store employees, while macys.com remained open.
Delivery networks were built to handle a more gradual increase in e-commerce order volumes, with some additional capacity to handle swings and holiday seasons.
This sudden, overnight shift to e-commerce further stressed delivery networks resulting in even longer lead times for product deliveries. As I highlighted in that post, the turning point was Amazon’s announcement that they would no longer adhere to the 1-day or 2-day delivery commitment of Prime:
Journey of an e-commerce package
To understand this, let’s dive into this topic in further detail and see what happens when a consumer clicks on the “Buy” button on an e-commerce site:
1. The journey of an e-commerce shipment begins with a “click” by the consumer to purchase an item. This kicks of a sequence of events processed through multiple IT systems.
2. The order management system (OMS) processes the order and directs it to the warehouse that it needs to be fulfilled from.
3. This order allocation gets recorded in the warehouse management system (WMS) which then directs a warehouse worker to pick and pack the particular product that was ordered.
4. At the end of the pick and pack process, order details are fed into a transportation management system (TMS) which generates a carrier (USPS, FedEx etc.) label that is then attached to the package.
5. When the carrier picks up the package, the “click to ship” step ends.
6. The package then enters the “ship to deliver” step. The package stays with the carrier through the entirety of this step until it is delivered to the end consumer.
In e-commerce, the delivery time for a package to a consumer is also known as “click to deliver” time which is the sum of “click to ship” and “ship to deliver” time.
Several factors affect these individual components of time and hence total delivery time:
- “Click to ship” time is often affected by warehouse congestion which is a function of the resources available in the warehouse: labor and equipment.
- During times when demand surges, inventory outages can delay the start of “pick and pack” slowing down “click to ship” time.
- Similarly, “ship to deliver” time is affected by several factors. Unlike full truckload networks, which travel point to point and hence have relatively stable transit times, parcel networks necessarily involve consolidation and sorting facilities that can affect the total transit time.
- Surging demand can affect the performance of this network depending on resources available just as it can affect time spent in the warehouse during the “click to ship” phase.
How did parcel networks respond to COVID-19?
January is the month when sanity begins to return to parcel carriers (and truck transportation in general) after an intense holiday season. This January was no different, with talk of a new virus in central China a seemingly distant affair. Then, on January 20, the first COVID-19 case was confirmed in Washington State.
Just like the coronavirus itself, this news went viral throughout the nation. Whether this resulted in “social distancing” or not (which became policy in several states only weeks later), consumer stockpiling accelerated through February.
Our research shows that transit times for parcel carriers like FedEx and UPS were significantly impacted in February. The cause seems pretty straightforward:
They were experiencing “holiday shipping” levels in a month where this was not expected.
Resources available in terms of number of trucks on the road, labor and equipment in consolidation / sorting centers were not able to keep pace with the demand. As each state began to declare “stay at home” all through March, one could safely assume that e-commerce would skyrocket and parcel shipping would be under more stress, creating further delays.
Interestingly, the data we reviewed from several shippers reveal that this did not happen! Transit times went back to pre-COVID-19 levels and for premium services like 1-day and 2-day shipping, they seemed to be better than “business as usual”.
This means that the major carriers adapted very quickly to the changing dynamics of the market by adding adequate resources. That they can ramp up is not surprising, given their ability to handle holiday season, but the fact they managed to do so with little notice is commendable.
It also makes sense that premium services were better than usual. These services are typically multimodal and use Air for part of the journey and Ground for the first and/or the Last Mile delivery. Given the significant decline in passenger air traffic, and the increase in e-commerce, air cargo capacity has been rapidly expanding.
In summary, this seems to be remarkable recovery story. However, there is a human cost to this story, where delivery workers have to work every day like it’s holiday season. Something the large carriers have to watch out for.
Why are my packages still delayed?
If parcel delivery improved in March, why are packages still taking a long time? To answer this question, let’s go back to the e-commerce package journey: Remember that consumer delivery time is a combination of “click to ship” and “ship to deliver” times.
In order to quickly deliver a product, e-commerce companies like Amazon predictively stock inventory near the consumer. This allows them to quickly turnaround and deliver it through cost-effective Ground shipping within a day or two (shorter journey equals faster “ship-to-deliver” times).
However, when consumers started panic-buying in February and March, that demand surge began to deplete buffer inventory in the supply chain — mostly in the warehouses and DCs located near consumers. These products now had to either be fulfilled from DCs further away from consumers, leading to longer travel times because (although these travel times are perfectly appropriate for the distance travelled) or the DC had to wait for fresh inventory to arrive from overseas or a larger first or middle mile DC.
The latter leads to an increase in the “click to ship” time leading to a delay in delivery even with smaller “ship to deliver” times. This may be a tougher problem to resolve for retailers as they have to manage inventory better across a wide variety of products.
How does gig economy last mile grocery delivery compare?
Gig economy last mile grocery delivery services whether Instacart, Shipt, or Amazon Whole Foods delivery, are all having a torrid time. Over the past few weeks, it has become impossible to find a time slot to get basic groceries delivered, even though you can easily get to the store and get the same item.
This is clearly a “ship to deliver” problem driven by a lack of adequate fulfillment workers (shoppers), drivers and delivery vehicles.
A friend of mine was enthusiastic enough (or desperate enough) to check these hours for 24 hours and figure out the exact time each delivery service opened its slots up! These slots were usually gone within half an hour of that opening time. Clearly, these delivery services do not have the same experience and ramp-up capacity as FedEx or UPS. But they are changing. Instacart is on course to hiring a record number of shoppers / delivery workers. But they are also facing challenges such as strikes from those very workers.
Capacity expansion is definitely on the horizon though as new players get into grocery delivery. One of my earlier retail predictions was that food delivery services would expand into grocery delivery.
This is certainly happening, with Uber Eats announcing its partnership in Europe, and DoorDash announcing it’s partnership with convenience stores in the US.
We are entering a brave new world of last mile delivery based on the resilience shown by traditional parcel carriers and the capacity expansion of gig economy delivery services. Now, if only we could solve the inventory problem! Stay tuned…
