Online Shoppers Don’t Always Care About Faster Delivery

Analyzing online customer data may reveal that other delivery attributes matter more than how quickly an order is received.

Reading Time: 7 min 



An MIT SMR initiative exploring how technology is reshaping the practice of management.
More in this series

Image courtesy of Jing Jing Tsong/

The COVID-19 pandemic changed how we shop. Soon after the onset of the global pandemic, retailers reported a rapid shift from in-store to online purchases. Online buying accounted for 18% of worldwide retail sales in 2020, up from about half that in 2018. Grocery shoppers stampeded to the web: As of May 2020, 41% of U.S. online grocery shoppers were first-time users of such services. The expectation is that the online shift will persist well beyond the pandemic across most retail subsectors.

Adapting to these changes is no small feat, and doing so profitably — well, that’s even harder. In response to the shift in consumer purchasing behavior since the beginning of the pandemic, many of the retail executives with whom we work are rethinking their existing omnichannel strategies. Our conversations with the e-commerce director of a leading European grocery retailer offered evidence of how difficult it is to shift operations to support the online shopper. This director was reexamining where to locate inventory and whether to own or outsource home-delivery capabilities in order to minimize the time between the placement of an online order and its receipt by the customer at home. Each of the proposed changes aimed at shortening the time between order and delivery was going to require investment. He wanted to make the appropriate choice.

We asked this director to question his long-held assumption that speed was the most important online delivery factor. How important is it for a customer to receive the goods fast? Isn’t it plausible that some customers prioritize being able to choose the delivery day over speed, and that others might be willing to forgo speed in return for a precise delivery-time window?

Obtaining the answers to these questions is an essential first step toward designing the optimal delivery network for online orders. To succeed, this director needed to know more about customers’ delivery preferences. Our research into retailers that offer attended home deliveries (in which a customer is present to receive their goods) suggests that focusing on speed alone may cause a company to overlook opportunities to acquire and retain online customers who prefer delivery attributes other than speed. In fact, we found that customers exhibit differences in their willingness to pay for delivery speed, delivery precision, and delivery day choice. Thus, retailers that can deliver not only on their product promise but also on the delivery attributes that matter most to customers have a competitive advantage. These are the retailers that will convert shoppers new to the channel into loyal, frequent customers.

Identifying Customer Preferences

Today’s retailers have a tremendous amount of data that can yield important insights about customer preferences. To assess how online customers value different attributes, such as delivery speed, delivery precision, and delivery day choice, we analyzed the preferences of online customers of a grocery chain across hundreds of thousands of purchasing instances. For each instance, we knew what the customer ordered. More important, we knew the available home-delivery time slots that were shown to the customer and the time slot the customer selected.

By analyzing customers’ choices, we discovered that online shoppers do not focus solely on speed. Instead, delivery attributes such as time slot precision (the duration of the delivery time window) and day choice (the availability of time slots across days of the week) also matter. In fact, we observed many situations in which customers preferred precision or flexibility over speed. We found that a customer is, on average, willing to wait 10.8 hours longer for a delivery if the delivery window is one hour shorter, and will wait an additional 7.5 hours longer if the delivery can be received on a preferred day of the week, all else being equal. The data also showed that customers tend to prefer to receive orders at the end of the week rather than on weekends.

Customer loyalty and basket sizes are also factors in shoppers’ delivery preferences. We found that customers who exhibit different levels of loyalty to the retailer exhibit different delivery preferences. For example, repeat customers are willing to pay more for the same delivery attributes compared with other shoppers. Moreover, customers with very large baskets are willing to pay double the delivery fee to improve delivery-window precision by one hour. Leveraging customers’ willingness to pay for the three core home delivery attributes of speed, precision, and day choice will require omnichannel retailers to consider preferences beyond simply delivery speed.

While these findings emerged from research we conducted in partnership with a grocery chain offering attended home delivery, they are likely applicable to other retailers offering that service, such as sellers of furniture, home goods, appliances, and other durable goods. However, the approach we developed for understanding the preferences of customers is relevant to any retailer rethinking its distribution networks. Amazon, Target, and Walmart are three among many assessing objectives other than solely speed and seeking to determine the “right” logistics needed to tailor delivery to the customer — a more nuanced strategy.

Precision Instead of Speed

Executing a strategy dedicated to speed alone can be expensive, particularly for grocery retailers, whose margins are notoriously thin. Retailers competing on speed tend to redesign their networks to bring inventory closer to the customers. This entails using stores as fulfillment centers or acquiring additional warehouse space closer to consumers. Some retailers invest in automatic picking systems to reduce the time required to pick and pack orders, whereas others expand their delivery fleets to shorten routes and decrease transportation lead times. Alternatively, they may partner with delivery providers. Embedded in each of these approaches is the belief that delivery speed drives value.

These moves may indirectly enable the retailer to improve its delivery precision and offer customers more choice across delivery days. But there is a less costly option worth considering: analyzing operational data about delivery patterns. For example, retailers could improve their routing capabilities by capturing and utilizing data about typical traffic patterns or tracking service times at customer delivery points. Based on these analyses, retailers can offer more precise time slots to their delivery customers.

In addition, retailers using third-party logistics providers could review their contracts and use their operational data to negotiate delivery services that better meet customers’ needs (such as requiring that they schedule services throughout the evening and weekend). These efforts are less expensive than those inventory and warehousing investments designed to improve mainly speed.

Committing to a speedy delivery seemed particularly important during the early COVID-19 lockdowns, when consumers were panic-buying. Nonetheless, many retailers were unable to offer consumers even a one-week delivery lead time, let alone a same-day option. This recent experience — and the popular press that accompanied these challenges — resulted in even more retailers focusing on speed as the crucial delivery attribute. We caution retailers, however, that this focus on speed may be too myopic once the pandemic ends. Why? As many consumers emerge from the pandemic, they will be more attuned to the trade-offs between speed and their availability to receive a delivery.

Interpreting Delivery Preferences

Instead of competing on delivery speed alone, retailers can assess the preferences of their customers and their corresponding willingness to pay for multiple delivery attributes, such as speed, precision, and flexibility. We have four recommendations for those who are in charge of designing their business’s omnichannel strategy.

1. Invest in data and analytics infrastructure. This includes tools that collect and manage data on customer behavior, such as how they navigate through a site. It is also essential to track purchases across both online and offline channels. That, in turn, often requires changes in data management to give cross-functional teams access to data from both channels.

2. Collect and analyze customer-specific time-slot selection data. Retailers just beginning this effort can focus on basic descriptive analytics that highlight what attributes customers prefer, on average, given the options chosen by customers at the moment of checkout. More advanced teams can deploy discrete choice models, as we have done, to understand willingness-to-pay characteristics across customer segments and various order types.

3. Understand what delivery attributes drive loyalty and repeat purchases. This requires the retailer to have some capability in predictive analytics. Questions of interest include whether different customer segments prefer one delivery attribute over another.

4. Work across teams to roll out new delivery strategies. To capitalize on their newfound understanding of customer preferences, operations and marketing teams must collaborate closely. Operations can design the fulfillment strategy that best meets customers’ needs. Marketing can promote the improved service and communicate how it differs from competitors’ delivery options.

Retailers need to support this new data-driven approach to decision-making by hiring the right talent, including operating and marketing professionals who are analytic and strategic systems thinkers.

In our experience, businesses that were founded as online retailers find this easier to execute than legacy retailers do, but the shift is still possible. Successful retailers tend to have an analytical champion who is willing to develop the capability internally to build the data collection infrastructure and attract the needed talent.

Retailers spend an inordinate amount of time focusing on personalizing their product assortments to meet customers’ needs. Focusing on meeting their delivery preferences could be the key differentiator in the extremely competitive retail business. Analytically minded retailers can craft delivery time slots that are unique to each customer based on revealed preferences. Based on our research, we strongly encourage retailers to rethink their operations to optimize not only on speed but also on the most appropriate combination of speed, precision, and flexibility.



An MIT SMR initiative exploring how technology is reshaping the practice of management.
More in this series

Reprint #:


More Like This

Add a comment

You must to post a comment.

First time here? Sign up for a free account: Comment on articles and get access to many more articles.