| Register/Subscribe Home |
|
|
ONLINE FEATURES
Book Reviews
BW Video
Columnists
Interactive Gallery
Newsletters
Past Covers
Philanthropy
Podcasts
Special Reports
BLOGS
Auto Beat
Bangalore Tigers
Blogspotting
Brand New Day
Byte of the Apple
Economics Unbound
Eye on Asia
Fine On Media
Green Biz
Hot Property
Investing Insights
Management IQ
NEXT: Innovation
NussbaumOnDesign
Tech Beat
Working Parents
TECHNOLOGY
J.D. Power Ratings
Product Reviews
Tech Stats
Wildstrom: Tech Maven
AUTOS
Home Page
Auto Reviews
Classic Cars
Car Care & Safety
Hybrids
INNOVATION
& DESIGN Home Page Architecture Brand Equity Auto Design Game Room SMALLBIZ Smart Answers Success Stories Today's Tip INVESTING Investing: Europe Annual Reports BW 50 S&P Picks & Pans Stock Screeners Free S&P Stock Report SCOREBOARDS Hot Growth 100 Mutual Funds Info Tech 100 S&P 500 B-SCHOOLS Undergrad Programs MBA Blogs MBA Profiles MBA Rankings Who's Hiring Grads | |
JANUARY 9, 2002 PERSPECTIVE By Heather Green E-Tailing's Statistical Black Hole Thanks to conflicting methodologies, it's impossible to divine anything concrete about how much consumers are really spending online
How's this for confusing? On Dec. 26, Yahoo! announced that its users spent $10.3 billion online in the fourth quarter. Seems straightforward until you triangulate that information with the forecasts of most experienced market researchers, including Nielsen//Netratings and Jupiter Media Metrix, which are estimating holiday sales at between $10 billion and $11 billion. So, is the takeaway that Yahoo! users did most of the shopping this holiday? Not according to America Online. On Jan. 2, AOL announced that its members also purchased about $11 billion in the fourth quarter. (Holiday sales are considered to start in mid October, so the period encompasses most of the fourth quarter.) DIG DOWN DEEP. Does that mean that the researchers are way off and that people actually spent more than $20 billion online this season? No. It just means we have entered the black hole of e-commerce estimates. To get a clearer picture of what the numbers mean, you have to dig down and compare methodologies. This is worth doing for a few reasons: to demonstrate the numbers' fuzziness and the importance of triangulating data from different sources. And to ground the hype before it takes off. Granted, e-tailing is still a pretty young industry with an ever-changing landscape. But online merchants don't get off the hook that easily. They complicate matters by publishing numbers without making it clear how they came up with their results. For example, Yahoo!'s Dec. 26 announcement states that sales at its online virtual mall rose 86% this year. But that 86% doesn't correlate to the $10.3 billion, and that fact isn't readily apparent. And since Yahoo! won't release sales figures for 2000, it's impossible to tell what the 86% really means. It gets even murkier when you look at how the inscrutable figure are arrived at. Since e-tailer and portals can't directly track what each of the millions of online shoppers is doing and spending, they make estimates based on panels and polls, which are generally constructed so that the mix of members reflects the overall population. Nielsen//Netratings and Jupiter Media Metrix each have panels of thousands of online users who allow the services to track what sites they visit. Using that data, the two companies project estimates of what sites the total online population is visiting. DUAL USERS. But the established traffic-tracking panels can't show how much, what, or even whether people are buying, so it makes them indicators, not true shopping monitors. To make sales forecasts, Nielsen//Netratings combines monthly polls of how much people say they spend with traffic trends. Clearly, data extrapolated from most panels have problems. What about polls? Companies like Ascolese Associates and ACNielsen survey groups of consumers, generally around 1,000 people, and ask them what they did online, how much they bought online, and what they intend to spend. AOL and Yahoo! used these two polling services to come up with their respective $11 billion and $10.3 billion estimates. Those results have a few problems, too. First, both AOL and Yahoo! include anything that people who use their services bought online. So, if someone is an AOL user but bought something at Amazon's service, that goes into AOL's hopper. Another problem is that the overlap in numbers isn't immediately apparent. An AOL subscriber could also use Yahoo!, and therefore think of themselves as a Yahoo! user as well. That means AOL's $11 billion and Yahoo's $10.3 billion can't be added together. I THINK I SHOPPED. Nor can they be taken as definitive totals of how much people spent just at those services' virtual malls. Yahoo! and AOL argue that they're interested in mapping the overall shopping behavior of people who use their services, not just what kind of shopping consumers do at their virtual malls. O.K., but then what do the figures really show? Finally, many surveys based on people's recollections have their own problems. Respondents don't always accurately remember what they bought and when. Even if they do, they don't necessarily want to reveal this information for sensitive categories, such as cosmetics or alcohol, notes Mohanbir Sawhney, a professor of technology and e-commerce at Northwestern's University's Kellogg School of Management. So, is e-commerce research doomed to be inaccurate -- or at least unclear? Quite the opposite. The promise of online research is that it will eventually be able to track and parse people's specific actions, as opposed to just what sites they visit or just what people say they do and buy. One promising newcomer attempting to do that is ComScore Networks. Launched in September, 2000, it uses technology that tracks actual purchasing data of a panel of about 1.5 million online consumers. Sophisticated monitoring services could potentially provide much more detailed information about the entire e-tail universe. "MORE GRANULAR." Says Sawhney: "Market research about e-commerce is a field in its infancy, but we will see better tools and better methodologies that will make it far superior to the techniques used in the offline world. The data are far more granular in the online world, once the initial hurdles have been overcome." That's encouraging, and I look forward to the day when press releases contain clear, meaningful data. Meantime, we're stuck with the byzantine and confusing.  Green covers e-business and the Internet for BusinessWeek Get BusinessWeek directly on your desktop with our RSS feeds.  Add BusinessWeek news to your Web site with our headline feed. Click to buy an e-print or reprint of a BusinessWeek or BusinessWeek Online story or video. To subscribe online to BusinessWeek magazine, please click here. Learn more, go to the BusinessWeekOnline home page  | | JANUARY |
 Copyright 2002 , by The McGraw-Hill Companies Inc. All rights reserved.Terms of Use | Privacy Policy |
Printer-Friendly Version
