From:
Bruce Kushnick
New Networks Institute (NNI)
26 Broadway, Suite 400
(212) 837-7867
New York, NY 10004
WWW.newnetworks.com
RELEASE DATE: OCTOBER 24, 1996
NEW REPORT REMOVES HYPE FROM ONLINE STATISTICS:
AND THE WEB GETS SMALLER.
New York, New York - Were there 9 million or 30 million people online in 1995? Why is there a 345% difference between "Hours used a month" among different surveys? Does 30% of all online subscriptions and traffic come from New Media industry "Noise?"
"Inter-NOT: Online & Internet Statistics Reality Check '96", is the first independent report and statistics primer that delivers a rational, objective, detailed approach to online statistics. It explains the differences between all 25 major studies, including the newest research from Yankelovich Partners, Intelliquest, Nielsen Media Research/Commercenet, FIND/SVP, O'Reilly & Assoc., Fairfield Research, INTECO, SRI International, Jupiter Communications, Georgia Tech, Project 2000, Survey.Net, Boardwatch, and Network Wizard's computer counting.
The report also offers a new form of market segmentation called "Layer Analysis" and provides 135 proprietary "Man on the Street" interviews. This report is the first of the three-report Inter-NOT Series.
"Statistical research for the online industry has been more like an amateur bake-off than an exercise in scientific analysis. Major flaws in the surveys, from lack of market segmentation and poor survey design to seat-of-the-pants market sizing, have created over-hyped statistics that are not useful for market planning," stated Bruce Kushnick, President, New Networks Institute (NNI).
Highlights of Findings
Total number of users - There never were 25 million people online, not yesterday, not today. After cross-referencing the surveys and adjusting for statistical problems, NNI found only 15 million users in America, about 10% of households only 6% of the population. Everything else is hype.
The primary problems include:
- Definitions: Every survey counts different items and no two studies actually compared the same things. These include:
- Number of users - "Households", "Users", "People with Access"
- Geography covered - "USA Only", "US and Canada", "Worldwide"
- Network used - Everything from all networks to just the "Internet"
- Random surveys can over-count users: Many firms, such as Yankelovich (which found 43 million people online), have included anything online from people only using email to even those who tried but no longer use online services. Others, such as by Nielsen Media and Intelliquest, overcount by not taking into account households without telephones, now estimated at 6-8% of all US households.
"There are more people who do not have telephone service in America than there are people online," states Kushnick.
Random surveys face an uphill climb. NNI's proprietary interviews found that many consumers do not know the difference between terms such as "Internet", "Net", "World Wide Web", "America Online", or even "CD-ROM".
- Self-Selecting surveys skew data NNI found that in almost all cases, self-selecting surveys (surveys where the person decides to fill out a survey online) show a larger male online population, with better computer equipment, who are very heavy online users. The Jupiter/Yahoo study had a 345% difference in the hours spent online compared to an average of random surveys.
- Online subscriber numbers are filled with puffery Numbers presented by the online companies, such as AOL or CompuServe have various forms of puffery, almost always inflating the total number of subscribers. For example, companies have included overseas operations, free trials, or even double-count households.
"During the course of writing this report, NNI found that many online services, including AOL, are billing customers for supposedly 'free trials'. This could account for as much as 5% of all online subscriptions. NNI has filed a formal complaint with the Federal Trade Commission to investigate this problem," Kushnick stated.
- New Media Noise? Industry members confuse the surveys.Over 900,000 people are in some way involved with the online industry (counting family members online). On the average, NNI found that New Media Industry members have 3-5 accounts per person. Some insiders have 15 accounts or more. Meanwhile, as far as network traffic and hours logged, New Media industry people are constantly online, every day, visiting web sites religiously. NNI believes that they account for only 6% of the online population but 30% of all traffic.
Other Significant Trends:
- Market is gradually shiftingThere is a definite shift from the "Phase 1 Users", such as educators and computer programmers, to less technical, less educated, and more women. However, the shift is still not at all bringing online equality as most users still live in households with 75%-100% higher incomes than national averages.
- Massive customer turnover (Churn)According to various sources, approximately 35%-50% of those who try commercial services leave, sometimes going to another service. Today there is no accurate data about how many people have tried and no longer use online services. NNI estimates that is represents at 30%- 40% of total online users.
- Online consumer users may have peakedNumerous studies indicate that over the last 12 months there has been a definite slowing of the industry, in terms of new users. With the extremely high turn-over rates, unless the last quarter of 1996 brings in large amounts of new customers who stay longer than 6 months, the overall industry will have either peaked or go into a slow decline, especially in the consumer segments.
- What's missing? Market SegmentationNNI found that almost all studies do not have a robust market segmentation scheme, and therefore are limited for market planning.
According to Kushnick, "The most important statistics for market planning are not total number of people online, but rather the number of users in a company's "Layer" or market segment. But most people always quote the larger numbers. It's like saying that anyone with a television set will watch PBS programming."
NNI's Layer Analysis shows that each layer must be treated with different business plans and initiatives because they all have separate users, growth rates, and revenue and profit potentials …and many layers are not revenue producing, but are useful applications. NNI has identified 12 major layers which include: "Academics and Universities", "Kids" (different from schools), "Personal & Entertainment" (sports, financial info., chat rooms", etc.), "Financial Services", "Adult", "Computerist" and the "New Media Industry".
For more information, please contact Bruce Kushnick at 212-837-7867 or at www.newnetworks.com. For sales information, please contact Interactive Publishing Alert at HTTP://www.netcreations.com/ipa/