Retail has experienced more change over the past five years than in the prior 50. Indeed, the pace of change accelerated throughout the COVID-19 pandemic as retailers adapted to changes in consumption, channel shifts,1 and rising customer expectations around speed and convenience. In fact, the race to shorten click-to-customer cycle time is arguably the single greatest influence on the shape of future omnichannel supply chains. Needless to say, the bar continues to rise for retail and direct-to-consumer brands.
How much does speed matter? Our research shows that when delivery times are too long, almost half of omnichannel consumers will shop elsewhere. As for how long is too long, we’ve found that more than 90 percent of US online shoppers expect free two- to three-day shipping. As retail supply chains accelerate, US consumers largely remain unwilling to pay for speed. McKinsey research shows approximately one in five US consumers will accept a marginal increase in shipping fees for faster shipping than standard free-delivery options. Given the high and rising costs of omnichannel order fulfillment, roughly 10 to 20 percent of sales in omnichannel retail, retailers are faced with tough decisions as they work toward improving delivery speeds profitably. Should they continue to build, should they partner, or can technology help unlock value in the speed equation where infrastructure and operations fall short?
The challenges of accelerated delivery
Retailers know speed matters: we estimate that Amazon’s free-delivery offering has accelerated more than 75 percent since the early 2000s, from more than eight days to two-day shipping by 2015—with select markets offering one-day delivery by 2019. Amazon continues to be a catalyst across retail, setting a high bar for direct-to-consumer delivery. In our experience, other retailers have closely followed this path. McKinsey’s recent survey of chief supply-chain officers found the pace will continue to accelerate over the next two years. We found roughly 75 percent of apparel, hard goods, and specialty retailers intend to build out network capabilities that offer two-day or faster delivery, and 42 percent are aiming for one-day click-to-customer lead times by 2022.
As delivery times compress, the detailed physics of the supply chain becomes increasingly important. Simply put, seconds count. Most fulfillment operations need time to pick and pack deliveries—by itself, that process takes an average of four to eight hours, though best-in-class omnichannel operations can fulfill orders within two hours of customer purchase. Once picked, parcel carriers then must pick up shipments from the distribution center, which often influences order cutoff times—the latest time a retailer can accept an order to meet the promised delivery time. Once a package is in the parcel network, traveling the final mile to the customer can take an additional day or more. Bringing it all together, one- or even two-day shipping requires tight cycle times and great execution across multiple parties in the supply chain.
To combat these challenges at least partially, most omnichannel retailers already use their stores for fulfillment or pickup. There are clear benefits to using stores, for example, enabling greater overall inventory productivity, quickening speed to customer, and avoiding markdowns. While these benefits can be meaningful, challenges still must be overcome:
- Inventory accuracy. Stores generally have lower inventory-accuracy rates (70 to 90 percent) than distribution centers typically enjoy (more than 99.5 percent).
- SKU complexity. When the online assortment includes channel exclusives, endless aisles, and even third-party products, minimizing margin-eroding split shipments across the network becomes challenging.
- Demand forecasting. Positioning inventory across distribution centers, various store types, and market fulfillment centers remains a struggle for most retailers; in fact, of all the levers to help retailers solve for speed to customer, accurate demand forecasting and distributed-inventory placement may have the greatest impact outside of network changes.
- Picking costs. While there are exceptions, for a majority of retailers the cost of in-store picking is much higher—typically 1.5 to 2 times higher on a cost-per-pick basis—than picking at distribution and fulfillment centers.
- Execution quality. Stores weren’t designed with fulfillment in mind, nor are they necessarily staffed or equipped with the technology to do so at scale. Particularly during peak times, it’s hard for most stores to manage exceptions, ensure accurate picks, and tightly control cycle times to customers—all of which are important to a great customer-delivery experience.
So what’s the next move for retailers? How do they overcome these challenges and provide faster fulfillment and better overall customer experience? Answers will vary, and it’s important to remember that beyond speed, other omnichannel conveniences such as curbside, returns, and buy online, pickup in store all play a significant role in differentiating the omnichannel fulfillment value proposition.
Solving the speed dilemma
Despite the many challenges of providing ever-faster omnichannel order fulfillment, we see four key characteristics that define fast and efficient fulfillment models.
Understand where speed matters
As customers expect faster delivery speeds, retailers can create greater impact from a segmented approach in shaping their delivery-speed promise. Retailers’ fulfillment engines can deliver to different segments at different speeds.
As customers expect faster delivery speeds, retailers can create greater impact from a segmented approach in shaping their delivery-speed promise.
In our research, we have found same-day delivery does not create value uniformly across the country. For instance, in working with one specialty retailer, we found roughly 20 US cities have densities that would typically justify the investments to enable same-day or next-day fulfillment. Consumers living in major cities such as Chicago, Los Angeles, New York, and San Francisco are likely to expect faster delivery than consumers in a smaller market. Indeed, consumers’ age, location, and economic disposition affect their expectations and willingness to pay for convenience and speed. And consumer expectations can further vary by retail segment:
- food and grocery—less than one day
- beauty—less than one day
- apparel—less than two days
- home décor—less than two days
- electronics—less than two days
- general merchandise—two days or less
Grocery and convenience retailers, for example, are solving for different customer needs than most other segments. In the next three to five years, we believe table-stakes delivery times will generally be fewer than two days across all major categories—and it would be hard to go back.
For retailers still navigating segmented speed capabilities, A/B testing is one way to glean these insights. For instance, one large US home-goods retailer took a two-pronged approach to solving the targeted speed problem. First, it set out to determine which items customers most wanted to receive quickly so it could deploy that inventory in advance closer to customers. Second, the retailer knew speed mattered but didn’t want costs to balloon from offering superfast shipping to customers who didn’t demand or expect it. Next, the retailer ran speed tests similar to an A/B test in that it showed different delivery-speed promises made to customers on the same product pages. It then tracked conversion as well as customer satisfaction, loyalty scores, and repurchase behavior. Over the course of about three months the retailer ran these tests and made two key changes: it started optimizing inventory placement across the network to enable faster delivery for specific products, and it started offering different, targeted speed promises for the same item to different customer groups based on geography, past behavior, and demographics. By adopting these changes, the retailer was able to achieve a nearly 10 percent improvement in its online conversion rate.
Invest selectively in network expansion
We estimate that retailers can provide two-day delivery to 80 percent of the US population by using approximately three distribution centers across their network. However, offering next-day delivery to 80 percent of the United States would require more than eight distribution centers—even more if retailers want to solve for lower parcel expense and density. Indeed, next- or same-day parcel delivery can cost retailers more than $15 per package, which is not tenable for the majority of retailers. At the same time, opening at-scale distribution centers—which often requires more than $100 million in capital for each distribution center—is likewise not feasible for many retailers. And for most retailers that are interested in growth, keeping up with the scale of Amazon and Walmart distribution networks is likely not the most prudent way to allocate resources. Indeed, if Amazon’s logistics unit were a separate company, we estimate it would be the fifth-largest third-party logistics company in the world; few, if any, retailers have the resources to compete at that scale.
Urban- or market-fulfillment-center strategies, such as dark stores and dedicated fulfillment locations, have emerged across retail networks. These are typically smaller-format operations—often less than 50,000 square feet, but sometimes smaller than 10,000 square feet. The capital costs are generally $5 million to $15 million, a fraction of the cost of standing up a new distribution center. Despite higher rent and labor costs for these locations, these costs are often offset by reduced costs of last-mile delivery, which can be substantial at up to 20 percent. Such approaches are already being explored across segments including grocery, home improvement, apparel, and consumer electronics. Faster and lower-cost shipments aren’t the only ways retailers can create value—in our experience, providing rapid in-market replenishment for nearby stores has allowed reductions in inventory of up to 20 percent by consolidating the forward supply of stock spread across stores.
Increase productivity with analytics and automation
Inventory analytics (to better position inventory in key omnichannel markets) and robotics and automation in the supply chain are two key levers to reducing time and cost to customer.
Refining inventory analytics. Retailers are expanding their fulfillment networks to include more complex distribution nodes, such as large multimarket distribution centers, urban fulfillment centers, stores, and dark stores. As they expand their networks, the choices and tradeoffs around distributed inventory—for example, balancing the forward weeks of inventory in a store versus allocating inventory to a local market fulfillment center—have become more complex. Traditional systems for allocating and replenishing inventory fall short in identifying the tradeoffs between breadth and depth of inventory across a cluster of in-market nodes. Indeed, many retailers see the shortfall of their current systems as a technical liability. Bespoke analytics and tools can help solve these issues as retailers test, learn, and adapt more sophisticated omnichannel inventory strategies. This is a critical area of focus for retailers pushing the boundaries of fast and profitable delivery to customers.
Adopting automation technology and robotics. Robotics and automated systems in distribution centers and stores can improve the speed and accuracy of omnichannel order fulfillment processes.2 The rapid growth of sophisticated solution integrators and “co-botic” solutions (robots that collaborate with humans) continues to accelerate, with numerous at-scale implementations emerging in the market. Moreover, many robotic and fulfillment solution providers have innovated different economic models, such as robotics as a service (RaaS), that reduce the upfront capital burden and allow more scalable, variable cost models to grow with the business. This can be equally attractive for companies with limited automation budgets and larger enterprises needing solutions that seasonally scale.
Solve last-mile challenges
Even the fastest fulfillment operation is still at the mercy of the speed and quality of final-mile delivery. For most, this means national parcel carriers remain critical partners in enabling advantages around speed. But the parcel market continues to pose headwinds, such as escalating surcharges (which are likely to remain standard practice) and strict, enforced capacity agreements. These pressures have created headlines and headaches for retailers of all sizes. In response, retailers must develop strategies that allow them to meet peak demand but do so at a reasonable cost. There are viable alternatives to traditional, national contracts:
- Regional carriers. Including regional carriers is becoming an imperative to a fast, resilient, and cost-effective last mile. It takes real effort for retailers to find, set up, and work day to day with a set of regional carriers to consistently meet high quality standards and transparency for customers; however, the recent market conditions have created opportunity for selectively “fragmenting” the parcel carrier base that serves across a retailer’s network.
- New delivery services. Using gig and platform services allows quick and scalable last-mile delivery options for retailers without taking on the burden of navigating new processes and fixed costs. The costs for these options can be high, though the benefits of greater conversion can offset the incremental expense. For example, Sephora and Best Buy have partnered with Instacart (traditionally viewed as a grocery-first business) to enable same-day fulfillment. Likewise, American Eagle Outfitters has used its partnership with ShopRunner to offer its customers free same-day fulfillment.
While autonomous vehicles and drones are the center of media attention, and do have long-term potential in select segments of the value chain, we see the maturity and capability of these solutions as longer term and therefore less relevant for most retailers’ last mile in the near future.
How operations can support a customer value proposition beyond speed
The North Star of a great omnichannel strategy is removing friction from the parts of the fulfillment process that matter most to customers.
For most retail supply-chain leaders, creating a faster fulfillment network is rightfully top of mind. A recent McKinsey survey of consumers shows that five of the top nine factors driving customer value in omnichannel retail are related to logistics (Exhibit 1). But it is important to remember fulfillment-network improvement is but one among several capabilities a retailer must build to remain competitive in omnichannel retail. Indeed, the North Star of a great omnichannel strategy is removing friction from the parts of the fulfillment process that matter most to customers. Below are a few of the avenues retailers can explore with operations support.
Ease of returns
Return rates continue to increase in omnichannel, with an average online return rate of 20 to 30 percent in the United States. Returns have become an important part of the omnichannel journey for customers. A majority of customers will not buy online if they don’t find the return policy satisfactory—that is, free returns with adequate time to evaluate their purchase, typically 30 days or longer. But returns are expensive for retailers. In addition to forward and reverse logistics and processing cost, retailers face low net-recovery rates—particularly in fashion or high-damage categories, where older returned products must be marked down and therefore sell for less the second time than their original price. It is important for retailers to consider returns as part of their omnichannel value proposition and optimize cost drivers relative to their customer value proposition.3
Scheduled delivery and consistency
Some categories lend themselves better to scheduled delivery than speed. For example, customers may prefer to select a specific window to receive bulky products such as furniture, appliances, large electronics, and certain home-improvement products rather than to receive them as quickly as possible. High-value products and fresh items in grocery that aren’t necessarily bulky may also lend themselves to these capabilities. Clear communication, predictability, and narrow delivery windows are essential to get this right. Retailers can also consider dedicating themselves to ensuring deliveries are made within the promised time period—regardless of the speed of the delivery. While occasional disruptions are inevitable, some retailers are beginning to test and correlate the impact of a delivery-consistency promise on conversion, and the impact of a missed delivery window on net promoter scores and repeat purchases.
Buy online, pick up in store and curbside pickup
Store-based pickup options have experienced tremendous tailwinds and innovations throughout the pandemic (Exhibit 2). A July 2020 poll of 50 retail executives indicates that store-based pickup offerings grew about threefold from mid-2019 to mid-2020. Some retailers report year-over-year growth of more than 200 percent in their own store-based pickup offerings. Moreover, McKinsey’s consumer insight research shows that about 60 percent of consumers plan to continue to use this option after the pandemic. These relatively high adoption rates are in part due to the convenient, free, and fast offering on this capability, as best-in-class retailers have orders ready for pickup in under two hours. This can be a compelling alternative to fast ship-to-home delivery.
Lockers and pickup and drop-off points
Lockers and pickup and drop-off points have somewhat limited penetration at less than 5 percent, but momentum is growing thanks to retailers such as Amazon and Walmart and parcel companies such as FedEx and UPS. Generally, this is a convenience that will likely be concentrated in food and grocery, and in urban areas where people lack a convenient or safe place to leave a package. Lockers have further benefits in that they reduce redelivery rates, which frees up already constrained capacity in parcel-delivery networks.
While these capabilities are not yet mainstream, US retailers can learn from higher-penetrated markets such as Europe, where a variety of nongrocers are using them. For instance, a farm-supply store uses lockers to provide 24/7 availability of critical parts outside of stores.
More and more, customers expect their retailers to share their values and be committed to improving the planet. Retailers can appeal to such customers in a number of ways. For instance, many retailers are going public with their sustainability commitments through 2050, often with the aim to be carbon neutral. Retailers can engage customers in these goals through fulfillment strategies such as allowing customers to opt in on the sustainable choice—whether it be choosing fewer packages or slower shipping. And retailers can provide incentives for doing so—for instance, by offering discounts on those orders or future purchase incentives.
The evidence is clear: customers expect faster delivery, and there’s limited willingness to pay for it. While there are real costs associated with enabling faster delivery, the cost curve can be shifted with a combination of strategies involving network expansion, technological capabilities, and partnerships. And a host of options beyond merely quick fulfillment can further help meet customer expectations. For information on meeting customer expectations during the upcoming holiday season, see sidebar “Navigating the upcoming holiday season.”
Pursuing such strategies will require both a mindset and operating-model shift among retailers. But by making these changes, and making them well, retailers can profitably provide customers the assortment, availability, and convenience they crave.