CNBC.com recently published an article titled “Online Shopping officially overtakes brick-and-mortar for the first time ever.” By Kate Rooney. Obviously, I was intrigued enough to read the article which quoted data from the Department of Commerce and analysis from Bespoke investment Group. They looked at the penetration of sales to determine that for the first time ever, online sales were more prominent than traditional brick-and-mortar retail, as illustrated in the graph below. At first glance, this made perfect sense. We have all been hearing for years that traditional retail is dead and that an apocalypse is imminent for stores. The time is at hand!
Then I noticed that the total penetration of what the article defined as “bricks” was essentially the same as it had been in 1992. With the meteoric rise of non-store sales, this made me question where the market share was coming from, prompting me to visit the Commerce Department website and downloading the supporting data to see for myself.
I was disappointed to discover, once I started to look at the data, that the analysis was flawed. There can be no debate that online shopping is growing at an astronomical rate. There can also be no argument that this is largely at the expense of traditional online retailers. However, the assertion that online (“clicks”) is now bigger than traditional in-store shopping (“bricks”) is just wrong.
The Department of Commerce classifies its data from a voluntary survey every month of statistically representative companies that are anonymized and classified according to the nature of their business and represented by NAICS codes. In the article, the author equates “bricks” to NAICS code 452, which is defined as General Merchandise. This code essentially represents department stores, but the author attributes it to the entirety of traditional store retail in asserting that online shopping is bigger than “traditional” retail. It ignores the fact that there are four other distinct codes that are also explicitly “traditional” retail:
|442||Furniture and home furnishings stores|
|443||Electronics and appliance stores|
|448||Clothing and clothing access. stores|
|451||Sporting goods, hobby, musical instrument, and book stores|
|452||General merchandise stores|
Collectively, these codes, along with a portion of code 453 (Misc. Stores Retailers) that represents office supplies (sub-code 4532) make up a classification known as GAFO. GAFO stands for General Merchandise, Apparel and Accessories, Furniture and Other Sales and the definition for this designation is officially “firms which specialize in department store types of merchandise.” In other words, this is the official designation for what would otherwise be known as “traditional retail” or “bricks”. When comparing this classification to NAICS code 454 (Non-store Retailers) we see a different picture:
It is important to note that the overall trend is that non-store retail is taking share from traditional GAFO-defined retail, and the trend has been accelerating since 2016. It is not true, however, that online is bigger than brick-and-mortar; “clicks” as defined in the article are still barely half the size of traditional retail.
It is also interesting to see where the share loss has been concentrated in recent years:
The Office Supplies, Electronics and Sporting Goods sectors lost share at double-digit rates since January of 2016. All other sectors included in the GAFO category also lost share over this time period, but at more moderate rates.
The other element that rang falsely in the article was the use of NAICS code 454 as a proxy for online sales. While purely online retailers like Amazon are a threat and putting pressure on retail, code 454 also includes sub codes 4542 (Vending Machine Operators) and 4543 (Direct Selling Establishments). It also omits the fact that all “traditional” retailers today have a significant online presence of their own. Therefore, the Department of Commerce publishes a separate quarterly report on eCommerce.
Looking at the latest quarterly report from the Department of Commerce and their full definition of “Retail” and “E-commerce” we get a different picture:
Again, e-commerce growth is outpacing traditional “Stores” growth. However, the point here is that “stores” are still growing and have been since the “crash” of 2008. The difference between the misinterpretation of the NAICS codified data and the “official” e-commerce data can be attributed to the fact that e-commerce and traditional stores are extremely difficult to separate. No retailer today can be classified as purely “stores” or purely “online”. Even Amazon is recognizing the importance of brick-and-mortar, although not yet in a financially material way.
There are some definite implications from the exercise of examining the data:
- Stores are still relevant. As seen in the data from the e-commerce quarterly reports, stores are still growing. The stores that are failing are the ones who were late to understand the changing role of physical stores and evolve their value proposition. It is no longer enough to house a breadth or depth of products in the store. No physical venue can ever match the internet. Therefore malls, and especially their anchor tenants, are failing. The value proposition of convenience and selection is no longer valid in the consumer mind.
- E-commerce growth will continue to outpace store growth. The rapid growth in penetration of e-commerce as a percentage of total retail will continue. Retailers should have a hard look at what products can be “digitized” and try to get ahead of the curve. What began with books, music and movies will continue. Retailers can either resist and become increasingly irrelevant or get ahead of the game by beating the competition to the punch.
- A new approach to physical stores is necessary. The difficulty is balancing the shift from current model, which still has relevance to a large (though decreasing) portion of the population, to the future model. For every Best Buy that is successfully making the transition, there is a Sears that is not. Sears is a tragic example of vision not being adequate. Eddie Lampert accurately prophesized the e-commerce shift years before the market yet did not successfully translate that foreknowledge into action. This Illustrates that in order to successfully navigate, retailers must intimately understand their customer, and more importantly the trends impacting their customers.
The author of the CNBC article is negligent in their definition of both the numerator (“bricks”) and denominator (“clicks”) in their comparison. Their assertion is, at best, several years premature. At worst it is a gross misinterpretation of the facts to match a catchy headline. However, we must not throw out the baby with the bathwater. The message remains valid. E-commerce is increasingly important. Successful retailers must be relentless in applying this lesson to transform into a truly “omni-channel” experience for their customers. “Which channel is most important?” is not the question. “How can I best satisfy the various needs of my customers?” is.