Calculating online sales is not as easy as it looks
By Kurt Peters
How much retail takes place online? The Census Bureau of the U.S. Department of Commerce reports that in the first half of this year, online sales totaled $24.4 billion. That number varies by $3.4 billion, or 16%, from the $21 billion that comScore Networks Inc. reported.
And what proportion of all shopping does online shopping represent? According to the Commerce Department, online sales were 1.5% of all retail sales excluding food service in the first three quarters of this year, more than double the 0.7% that the Commerce Department calculated in Q4 1999, the first time it started breaking out e-commerce numbers for retailing.
Impressive numbers—except they don’t tell the whole story. For one thing, the percentage that shows e-commerce as a proportion of all retail sales understates e-commerce’s influence significantly. And the online retail sales that both the Commerce Department and comScore report don’t include what could be considered retail sales at eBay.com.
EBay began life as a garage sale site for consumers to unload stuff they didn’t want any more. But it didn’t take long for smart entrepreneurs to figure out they could buy wholesale and sell it on eBay for a profit—and make a living at it. Others quickly learned they could do the same with surplus or liquidated merchandise.
In 2003, eBay had estimated gross merchandise sales, as it calls the sum of all sales and auctions at its site, of $23.2 billion, compared to $14.8 billion in 2002. “If they were a retailer, eBay would be growing faster than anyone else,” says Scott Wingo, president of ChannelAdvisor Corp., which provides software and systems to help companies sell on eBay.
While many people still think of eBay as an online flea market, the channel has been transformed in the past three years as eBay has made an effort to become attractive to brand name manufacturers and retailers looking for another outlet for merchandise or looking for a liquidation channel for returned or outmoded goods.
Consequently, the amount of sales that take place on eBay that could be considered retail is significant. EBay itself claims not to know how much of the sales at eBay.com are substitutes for purchases at retail outlets. But as part of keeping in touch with the pulse of its potential market, ChannelAdvisor periodically surveys sellers on eBay to determine who are the ones simply getting rid of junk from their attics and who are actually making a living at selling on eBay. The answer: Quite a few are making a living at it.
Wingo breaks down 2003 eBay sales as follows: Gross merchandise sales reached about $23.2 billion. Of that amount, about a third, or $7.6 billion, were overseas sales. Another roughly $4.3 billion were automotive sales. Assuming that those sales do not represent retail sales that could have occurred elsewhere, that leaves $11.25 billion in merchandise sales. ChannelAdvisor estimates that sellers made a living in 2003 from the sale of $3.85 billion worth of merchandise.
Wingo estimates that sellers who account for the $3.85 billion in gross merchandise sales take in at least $20,000 a month. With a net margin of 10%, that translates into income of $2,000 a month.
The bottom line: Based on ChannelAdvisor’s surveys, the Commerce Department and comScore are missing $4 billion a year in online retail sales. Most researchers are projecting online sales this year of $55 billion, excluding event tickets and travel. But the real number could be closer to $59 billion. If sellers who work part time at selling on eBay are taken into account, the number goes even higher. It’s simply a matter of where to draw the line.
Wingo says he sees no end to the growth of eBay. And in fact, he expects the proportion of eBay sales that represents retail to continue to grow. In just two years, that proportion has grown from 20% of gross merchandise sales excluding international and autos, to 40%; from $417 million in Q1 of 2002 to $1.14 billion in Q4 of 2003. “The mix of sellers is changing from the garage sale and small businesses to the big guys,” he says. “And the big guys will become a huge factor.”
The right proportion
Just adding eBay retail sales to the Q3 online retail numbers would hike the proportion of all retail sales that take place online, from the 1.5% that the Commerce Department reports to 1.65%. But even that doesn’t tell the whole story.
In figuring the percentages, the Commerce Department takes the total retail sales number minus food service, which it rightly determines is not likely to move to the Internet—$2.5 trillion for the first nine months of 2003—and divides it by the e-retail number, $37.68 billion for that period, to come up with a proportion of 1.51%. But even taking out food service sales, the Commerce Department is still counting many offline sales that are unlikely to take place online. Those Commerce Department categories include automobile and other motor vehicle dealers, gasoline stations and fuel dealers. (The Commerce Department collects automobile online sales but reports them in the automotive category, not e-commerce.)
Subtracting those categories from the total leaves $1.65 trillion in retail sales that could plausibly take place over the Internet for the first nine months of 2003. The proportion of online sales shoots up to 2.28%. Adding eBay retail sales of approximately $2.71 billion in the first nine months of 2003 takes the proportion to 2.38%. So, online retail sales as a proportion of retail sales that could take place online could be 60% higher than the proportion that the Commerce Department reports.
The proportion changes, however, when it is calculated on the basis of the comScore numbers. ComScore reports online retail sales for the first half of the year at $21 billion. That number is 1.95% of $1.07 trillion in first half retail sales exclusive of food service, autos, fuel and gas station sales. Adding eBay’s $1.74 billion to comScore’s $21 billion gives a proportion of online sales to total sales of 2.11%.
The difference in numbers between the Commerce Department and comScore is the result of differing methodologies. The Commerce Department surveys 11,000 retailers to determine their e-commerce sales while comScore monitors the online behavior of a representative cross-section of the 1.5 million online consumers who have given comScore explicit permission to monitor their behavior. But each also includes somewhat different categories from the other. Both researchers include buying at Quixtar.com, the online arm that supports Amway, and other such sites.
Interestingly, the Commerce Department and comScore benchmark their numbers against each other. When the Commerce Department comes out with its numbers two and a half months after comScore, comScore often points out that its numbers are not a whole lot different from the Commerce Department’s, the benefit of subscribing to comScore being that a customer will get information that it can act on earlier than if it waits for the Commerce Department. ComScore subscribers also get weekly data vs. quarterly from the CommerceDepartment and category data. At the same time, a source familiar with the Commerce Department’s methodology says researchers there take comfort in knowing that their numbers are not far off from comScore’s.
A shift in behavior
Other researchers report e-commerce numbers as well. BizRate.com Inc., a shopping comparison site and quality monitoring company, says consumers spent $24.45 billion online in the first half of 2003, almost identical to the Commerce Department’s figure. BizRate bases its numbers on the shopping activity at the 2,000+ retail sites where it has contracts to monitor shoppers’ activity. It adds to those numbers the publicly reported numbers for sites it does not monitor, then uses a formula to fill in the rest.
While the raw numbers are interesting, what makes them compelling is to see how they reflect a shift in consumers’ buying behavior. The rise of e-commerce has been accompanied by a decline in catalog sales. From 2000 to 2001, the Commerce Department reports that e-commerce sales were up by $6.24 billion while catalog sales were down by $7 billion. From 2001 to 2002, e-commerce sales were up $8.87 billion and catalog sales down $1.63 billion. The shrinkage in catalog sales accelerated last year, with e-commerce sales up an estimated $12.2 billion and catalog sales down an estimated $5.9 billion. E-commerce sales as a proportion of remote sales—that is, combined e-commerce/mail-order sales—grew from 25.7% in 2000 to 31.6% in 2001, 37.2% in 2002 and 45.3% in 2003.
In addition, the number of catalog shoppers has declined the past two years by 20 million, while the number of at-home online purchasers has risen nearly 28 million, reports the American Interactive Consumer Survey from The Dieringer Research Group.
Because the growth in e-commerce sales outstripped the shrink in catalog sales, another conclusion to be drawn from those numbers is that store shopping is shifting to the web. The magnitude was only a matter of $5.9 billion in 2003, so no one believes any more—as many did in 1999 and 2000—that store retailing is in danger. Nonetheless, the shift underscores the importance to retailers of having a multi-channel strategy. “There’s nothing to suggest that that momentum will decelerate,” says Jim Okamura, partner with retailing consultants J.C. Williams Group. “If you’re a national retailer or a regional retailer of any size, consumers are expecting you to have multiple channels that they can buy from, not just browse in.”
A big impact
Furthermore, the reports of dollar volume of online sales ignore the by now well known phenomenon of the web influencing offline purchases. Dieringer’s American Interactive Consumer Survey reports that the web influences $1.50 in offline sales for every $1 spent online. In that case, the web is responsible for $110 million in offline retail sales. When Dieringer wraps travel sales into the equation, it estimates that online research affects $137.6 billion in offline sales, meaning the total effect of online shopping is $230 billion in retail and travel sales.
Tom Miller, senior consultant with Dieringer, says that ratio has held steady
over the nearly 10 years that Dieringer has conducted its study. “Anybody doing
hard numbers like comScore is creating the right benchmark for online sales,”
Miller says. “But if all you’re doing is counting clicks, you’re missing something
with a big impact on retail.”
kurt@verticalwebmedia.com
