When working professionals were shifting to home offices a year ago as COVID-19 was spreading, comedians and pundits predicted that people would no longer need bras and pants that aren’t stretchy. Instead customers would make room for pajamas and athleisure wear.
Only one part of their forecast came true. Sales of pajamas more than doubled in 2020 compared with 2019, the fastest growth among clothing categories, while sales of suits and dresses fell. In contrast to popular opinion, sales of bras and pants did not exhibit major changes compared to 2019, but how they were purchased shifted significantly as more consumers than ever started shopping online.
As vaccines fuel hope about life after COVID-19, retailers are trying to learn more about the pandemic shopper. Are they different from the existing online users? Do they buy differently? Will they stick around? How crucial will they be for profit margins?
The stakes are high. Without a doubt, bringing more customers into online channels offers a great upside. Not only do they contribute to sales today, but they may continue to buy online even after the pandemic. However, if they have a bad experience, they may return their products, which is costly for retailers, and they may not buy from the retailer again. They might also be costlier to serve given their limited experience with online shopping.
Making sense of apparel buying
To learn about new online shoppers who emerged during the pandemic, we focused on the apparel industry, which reached global sales of $1.9 trillion in 2019. Even before the pandemic, roughly 28 percent of global apparel transactions were happening online. At the same time, many brands had closed their physical stores in the so-called “retail apocalypse.” The pandemic accelerated both of these trends for the apparel industry, with J. Crew and Century 21 Stores filing for bankruptcy, while ecommerce sales saw double-digit growth.
We set out to compare consumers who already shopped online before the pandemic with those who started buying products online only after the pandemic. True Fit, a data-driven personalization platform for fashion retailers, provided us with an aggregated dataset of online apparel sales from a variety of anonymized well-known fashion retailers in the United States and the United Kingdom in 2019 and 2020.
The data included anonymized sales of more than 63 million users who made more than 170 million transactions that resulted in approximately 480 million products sold for several billions of dollars. We compared each week of 2020 to the same week in 2019 to factor in seasonal changes in consumption.
We found that roughly 33 percent of US shoppers and 20 percent of those in the UK had not purchased from online retailers before the pandemic. In March, that figure decreased slightly to 32 percent and 18 percent, respectively, as housebound people started turning to the internet for purchases. By April, immediately after the first severe lockdown, the number of new online customers had spiked, as new shoppers represented 38 percent of shoppers in the US and 23 percent in the UK.
Over time, the fraction of new online customers in the US and UK returned to pre-pandemic levels, increasing slightly in October. But the shift to online shopping happened even as lockdown measures dampened sales (both the quantity of products sold and dollars spent) in the early weeks of lockdowns. In the UK, sales remained lower than in 2019 throughout October. In the US, sales recovered more quickly, and within a month of the lockdown measures, the quantity of sales increased compared to 2019. Within another month, so did spending despite prices that were lower in 2020.
People still need pants
We also looked at the average basket size among new ecommerce customers and found that it was similar to that of existing customers, those who bought online before the pandemic. That suggests that new customers turned to online channels to cover similar needs as existing users.
We further examined basket composition, investigating whether new consumers bought different product categories compared to existing consumers. With the exception of pants, including jeans and trousers, there were no significant differences among the popular product categories of new and existing online consumers. New customers were less likely to purchase pants through a website compared with existing customers, suggesting that categories other than pants might have prompted them to buy online.
Another notable trend: In the US and UK, kids apparel accounted for a larger fraction of sales than adult clothing in 2020. This makes sense: Kids needed new sizes, even in a pandemic.
Will new online shoppers stick around?
It’s too early to say whether new ecommerce customers will stay online or move back to traditional brick-and-mortar stores when restrictions disappear. However, we examined their loyalty patterns by looking at the repeat purchase rate of these new customers compared with those of existing users. To do that, we calculated the portion of consumers who bought again within 60 days of their first purchase. We then tracked buying rates after their second purchase, and so forth.
It’s a well-known phenomenon that repurchase rates after the first transaction are lower than repurchase rates after several transactions. In other words, customers who make a few purchases are more likely to buy again than those who only made one purchase.
The interesting question here: Are repurchase rates for these new online shoppers lower than those of other customers acquired before the pandemic? If, as some retailers fear, these new online customers are less loyal compared with existing customers, we would see lower repurchase rates both now and in the future.
To make sure the new and existing consumers, and their market conditions, were as similar as possible, we compared repurchase rates between consumers who started buying online right before the pandemic (between January 1 and March 14) and those who started buying online only after the pandemic.
For pandemic consumers, we created two cohorts of new online shoppers. The first cohort contained those who joined during the first wave, between March 15 and June 15. The second cohort comprised those who joined later, between June 16 and August 15.
As expected, repurchase rates increased as customers continued buying. Most important, we found no evidence that new ecommerce shoppers were less loyal than existing customers. In fact, we observed quite the opposite: Customers who started buying online during the most severe lockdown conditions exhibited higher repeat purchase rates compared with those who joined before lockdowns or after the lockdowns were lifted. This pattern was robust across both the US and UK.
Now that we know that these pandemic consumers could stick around, it is crucial to investigate whether they have negative effects on companies. Namely, do they yield more product returns, which are notoriously higher online than in the physical world?
What about product returns?
Product returns are too common in online channels. They are often attributed to the inability to inspect items before purchase or to generous return policies that attract consumers to retailers’ websites. Companies typically have to balance the popularity of these policies in a competitive market with the high costs of managing returns, including shipping and restocking fees that cut into margins.
Given the importance of this behavior on a company’s profitability, it’s natural to ask whether these new online shoppers will increase product returns.
Based on their lack of familiarity with online apparel shopping, retailers might expect new customers to return items more often compared with existing customers. That would also mean that retailers might need to spend more effort helping new consumers find the right products. On the other hand, these new customers might be less familiar with the return process and take more time to find the right product to avoid it.
To shed light on this question, we calculated product return rates—defined as the percentage of sales returned within 30 days of purchase—for new and existing customers.
Return rates of products purchased in March and April were lower than subsequent months, reflecting the lockdown reality or the fact that customers were only buying based on necessity. As the pandemic evolved, product return rates increased for both new and existing customers in the US and UK. Perhaps customers learned that they could return products, started buying things they didn’t need, or visited newly reopened physical stores to handle returns. It’s difficult to isolate any of these factors.
Importantly, and maybe surprisingly, we found that new online customers in both geographies had lower return rates than those of existing customers, even as time passed. This finding is particularly promising for e-retailers. It suggests that new online shoppers might be intrinsically different from past customers in a key way: how they view returns.
We even went a step further and probed three other factors that might explain lower return rates.
First, we wondered if new online customers simply weren’t aware of a brand’s return process yet. Perhaps these customers’ return rates would eventually become similar to those of existing customers. However, when we calculated the return rates for new customers’ first and repeat purchases, we found few differences between them, suggesting that such learning is unlikely to explain the difference.
Second, we asked whether new online shoppers purchased fewer return-worthy products than existing customers. If that was the case, the lower return rates were merely driven by the types of products they bought, not their behavior. But when we calculated return rates across all product categories, comparing new and existing customers, new customers still had lower return rates.
Lastly, could it be that these new online shoppers were the brand’s loyal brick-and-mortar customers? Perhaps they had no choice but to shop online when the pandemic hit? If that were the case, these shoppers could actually be higher quality customers who would eventually revert to in-person browsing.
Without visibility into these consumers’ previous offline purchases, we couldn’t rule out this explanation, though we do see similarities in basket sizes between new and existing online shoppers. That suggests that the new shoppers aren’t “higher quality” in terms of basket size.
The takeaway for apparel retailers
With the COVID-19 pandemic and lockdown measures, a segment of new online customers has emerged in the apparel industry. These customers buy similar products to existing cohorts, with similar basket sizes and average spending. Importantly, they have relatively high repurchase rates and seem to return products less frequently than existing online users.
While the pandemic is still playing out, this preliminary evidence suggests that these new customers are turning out to be valuable consumers for online retailers. Companies should use the data they collect from digital interactions to learn about these new customers: How do the new e-commerce customers compare to the pre-pandemic ones? Are they buying the same products at the same rates and prices? Are they returning items in the same rates? What is important for these new customers?
Companies should use these insights to deepen their relationships with these new customers—and more newcomers on the way—and delight them with good service and product selection.
About the Authors
Ayelet Israeli is an associate professor at Harvard Business School. Eva Ascarza is the Jakurski Family Associate Professor of Business Administration. Laura Castrillo is the former head of analytics for Europe, the Middle East, and Africa at True Fit.
[Image: iStockphoto/cybrain]
How has the pandemic changed your buying habits?
Share your thoughts in the comments below.