Seidman's Online Insider - Vol. 4, Issue 41

============================================================================
               Seidman's Online Insider - Vol. 4, Issue 41
    Visit the Online Insider on the Web < http://www.onlineinsider.com >
============================================================================

Copyright (C) 1997 Robert Seidman.  All rights reserved. May be
reproduced in any medium for noncommercial purposes as long as
attribution is given.

IN THIS ISSUE

- Editor's Note
- What's Happening in Insider Talk?
- More Fun With Numbers
- CompuServe's Numbers
- Stock Watch
- Subscription Info

Editor's Note
=============

There will be no newsletter next week, but we'll be back with one on
Dec. 7.

Last week I wrote about a new site, http://www.websitegarage.com, that
offers a free "tune-up" for Web sites.  The site will evaluate how well
any specific page (URL) fares, ranging from poor to excellent in five
categories.  I was amused when someone wrote me to advise that Web Site
Garage's own site only ranked "Good" on an overall basis.  It ranked
POOR in Web site popularity, which is especially amusing since one of
the things it claims it will do (for a price) is increase the popularity
of the site.  So, I forwarded the note to the address of the "Head
Mechanic," per the business card sent with the press materials, and the
mail bounced with a user-unknown message.  I still think it's a great
idea, but the execution isn't all that impressive so far.

For those of you who missed my appearance on the Silicon Alley
Reporter, hosted by Jason Calacanis, you can hear it (or much to
my horror, SEE it) at < http://www.pseudo.com >.  Just select Silicon
Alley Reporter from the pull-down menu at the top and you'll figure
the rest out.  You'll also hear Ted Leonsis call-in and talk about
AOL's future.  Ted claimed that he reads the newsletter every Sunday
night or, on Mondays when I'm lazy and don't send it out until
then.  I didn't get the chance to tell Ted that usually when he
gets it on Monday, it isnt because I was busy or lazy.  Most of those
times, I still sent it out on Sunday -- it's just that AOL didn't post
it to his e-mail account until Monday!

You can also hear Fred Wilson's (one of the partners of Flatiron Partners,
< http://www.flatironpartners.com >) thoughts on when the newsletter has
been good and NOT so good.  Of course,  Fred thinks every issue should
contain a "More Fun With Numbers".  So Fred, this issue is for you!


What's Happening in Insider Talk?
=================================

The discussions in "Insider Talk" are beginning to take off.  It's very
difficult to pick a "post of the week" this week because there are many
that I thought were very, very good.  So I'm calling it a three-way tie!

Former Prodigy in-house attorney William C. Schneck gives his views on
what some are calling CDA II ( a follow-up to the defeated
Communications Decency Act) at:
< http://www.wellengaged.com/engaged/seidman.cgi?c=policy&t=10&q=8 >.

Is the Web the greatest thing to come along in the 20th century?  Gerard
Van der Leun doesn't think so (and as much as I like the Web, I
wholeheartedly agree!):
< http://www.wellengaged.com/engaged/seidman.cgi?c=web&t=6&q=26 >.

The Mining Co.'s Scott Germaise becomes the first repeat "post of the
week" winner, with his thoughts on whether the online experience is
solitary or interactive.  The question doesn't seem that important to me
(though it does have wide-ranging impact in terms of how you might
"program" your service), but I think Germaise's conclusion that there's
no one right answer is the right answer to many of today's raging
debates over what works and what doesn't on the Internet.  Read it for
yourself at:
< http://www.wellengaged.com/engaged/seidman.cgi?c=online&t=21&q=18>.

For a list of all that's going on, please visit:
< http://www.onlineinsider.com/html/insider_talk.html >.


More Fun With Numbers
=====================

Put on the theme song from the movie "Deliverance" -- it's DUELING
surveys.

But first, a couple of questions:

How many adults use the Internet in the United States?

A. 39.7 percent
B. 27 percent
C. Both of the above
D. Neither of the above.

>From where do most people in the United States access the Internet?

A. Home
B. Somewhere besides home
C. Both of the above

I don't honestly know the answer to these questions in light of results
from IntelliQuest and Chilton Research, because IntelliQuest says 27
percent of people age 16 or older access the Net and the Chilton survey
says 39.7 percent of people age 18 or older access the Net.

While IntelliQuest reported that 42 percent, or more than 23 million, of
the overall online population access from more than one location, more
people (66 percent of online/Internet users) said they access from home
than from work (46 percent of online/Internet users).  (The numbers
don't add up to 100 percent because of those pesky 23 million-plus who
access from more than one location.)

Now, the Chilton Research numbers are tricky, because they list nothing
but percentages.  Chilton's survey indicated that 23.1 percent of the
survey population (including those who DON'T access the Internet) access
from home, whereas 29.1 percent access from somewhere besides home (with
20.1 percent of the survey population accessing from work).

If I have to pick which numbers I like better, I'll definitely go with
IntelliQuest.  You can find out more about its study at
< http://www.intelliquest.com >.  I couldn't find anything for Chilton
Research other than the press release that I got from
www.prnewswire.com.

One final note on the Chilton Research survey -- it said 45 percent of
consumers own PCs and 39 percent access the Internet.  Though it's true
the survey also claims most people don't access from home, it's
interesting to note that the gap between the people who access and the
people who have PCs is pretty narrow.  Granted, Chilton said most people
don't access from home, but most of the people with PCs have access of
some sort whether it's from their own PCs or not.

*More Data Than You Probably Want!*

Now, on to IntelliQuest.  I had to think on this some.  Can I buy 56
million adult (16 or older) U.S. users that access the Internet or an
Online Service according to Intelliquest?  Yes, I can, because
IntelliQuest defines a user as anyone who used the Internet and/or
online service at least once during a three-month period.  That doesn't
mean that I think there are 56 million people who either accessed
through their jobs or their own accounts.  I'm guessing that number
would be lower (I'd say maybe as much as 20 percent lower).  Why?
Because if you showed your mom and dad the Web on your account, they
could legitimately say they used the Web.  I don't doubt a lot of that
sort of usage occurs.

But while I can buy into big numbers, I still am concerned that they
create unrealistic expectations.  People don't look deep.  They hear 56
MILLION ADULTS and think, "We ought to be able to make money on this!"
I believe there is a growth trend, and that is good!  I believe more
people are accessing for more time.  I believe the Internet is becoming
more important.  I believe all of these things.  But I don't believe
it's as big as this study would suggest.  And I believe that a layer or
two into the study, IntelliQuest may have some very valuable demographic
data.  But the top-level results it released are meaningless and
sometimes wrong.

Where the IntelliQuest survey proved itself pretty useless to me was
where it said the 56 million users average 9.8 hours of use per week!
This is almost 550 MILLION hours per week, or 2.4 BILLION hours a
month.  That's pure fantasy.  I think someday it will happen, but
there's no way 56 million adults are using the Internet for a total of
2.4 billion hours a month.  If you believe such a level exists now and
plan a business model around those numbers, please, just send me a check
for $10,000 and forget about it.  You'll save a lot of money by doing
that instead of funding a business based on the rosy picture painted in
the study.  Okay, perhaps I am being harsh. I do take into consideration
that some people who are connected at work all week might have said
they are connected more than 40 hours a week.  If that's the case,
the numbers are perhaps accurate, but meaningless in terms of what
they mean from a potential business perspective.  If I leave my TV
at home on 168 hours a week, but I am only home 70 hours a week, and
I am only awake 14 of those 70 hours and I only WATCH the TV 2 out
of 14 hours, which numbers are important?  The answer to that question,
I suppose, depends on who you are.  Are you the network, local and
cable television companies, or are you the advertiser?

Now, I'm not going to fault IntelliQuest; I think all this speaks more
to people's inability to answer questions correctly on a survey.  What's
the incentive to be accurate?  I'm not suggesting that people lie when
they take surveys, but people will say anything.  It also depends on how
the questions are asked.  We could speculate on these reasons all we
want.  But here's something we don't have to speculate on.  The
conclusion is wrong.  Not just a little bit, but very, very, very wrong.

*Real Numbers vs. Fantasy*

Given the choice between a survey and real data, I'll take real data.

America Online is responsible for 9 million U.S.  accounts. It could
well be responsible for (and probably is) far more adult users (since
accounts can have multiple users), but let's stick with 9 million so we
err on the conservative side.  AOL represents, by IntelliQuest's
estimate, about 16 percent of the overall users.  We also learned
something about CompuServe's base from its recent quarterly results.
Here's what we know, on the basis of numbers from the companies
themselves:

                                       Avg. Hours/Month       Avg.
                                        Per Acct**            Hours/Week
Service                                                       Per Acct***
=======                              ===============        =============
America Online                          21                            5
CompuServe (CSi)                         4                            1
CompuServe (SPRY)                       11                            3
IntelliQuest survey                     42                           10

(Source: Seidman's Online Insider)

The calculations for AOL and CompuServe (CSi) are somewhat misleading,
though SPRYNET's numbers  ought to be pretty clean, because the total
monthly network hours for CompuServe (CSi) and AOL each include roughly
1 million International customers.  The impact is bigger for AOL than
CompuServe because CompuServe didn't begin offering flat-fee pricing until
Oct. 1.  In other words, two-thirds of CompuServe's recent quarterly
report
wasn't affected by the shift to flat-fee.  Neither AOL nor CompuServe
offer
flat-fee pricing  to international subscribers.

** Rounded

***Rounded and based on a 30-day month (month total/4.29 weeks)

Of the above numbers, the SPRYNET figures really surprised me.   I would
have guessed it would be higher.  I would think SPRYNET, with 311,000
subscribers, is fairly representative of large ISPs (EarthLink, Erols,
Prodigy Internet, MindSpring, Netcom, AT&T WorldNet, etc.).  But let's
just say for a moment that we can buy that the average hourly use per
week of an AOL user is nearly twice that of an ISP user (some may not be
surprised at that thought, but I'd point you to the SPRYNET numbers!).
But let's  just say we can live with that.  I still can't live with what
this means in the aggregate.

AOL has 9 million domestic accounts responsible for about 45 million
hours per week or about 5 hours per week per account (users vs. accounts
doesn't matter for this piece because if we believe that there would be
more users than accounts, thereby lowering the overall average number of
hours per user, we still know the total hours). IntelliQuest, on the other
hand, said there are 56 million users averaging 9.8 hours a week each, for
an astonishing 548 million hours per week.

So even with a 1-to-1 relationship between accounts and users, you're
looking at 16 percent of the users being responsible for only 8 percent
of the traffic.  We could easily assume that AOL was responsible for 1.5
users per account.  I think excluding the workplace you'd almost have to
make that assumption across the board.  If domestically, AOL, the
Microsoft Network, CompuServe and WorldNet account for about 14 million
accounts and you say there are as many total ISP accounts as there are
online service accounts (and based on all the other numbers I've seen,
that should definitely get you in the ballpark), you're talking about 28
million accounts in the United States.  If that's right and the 56
million is right, there are about two adult users per account.

But let's say that with AOL it's 1.5 users per account, or 13.5 million
total.  In that scenario, AOL represents 25 percent of the users.  And I
don't have the data, but I bet IntelliQuest's own data will show at
least that much.  (My guess is actually it would show much more than
this; when things are inflated, they are usually inflated across the
board.  I didn't get the number of AOL users the study suggests, but my
guess is that it's more than AOL itself claims!)  Are you willing to
believe that 25 percent of the users are responsible for only 8 percent
of the traffic?  Throw in CompuServe's numbers and it's even worse.  And
if all the traffic is on the ISPs, then how come AOL's usage is almost
twice that of a fairly big ISP, SPRYNET?

IntelliQuest's data suggests 2.4 BILLION HOURS A MONTH of usage.  So,
2.4 billion hours of usage and so many companies losing money?

*Online Commerce*

More good news from IntelliQuest: There are 8.7 million people who will
purchase $7 BILLION worth of goods a year electronically, based on the
current run rate.  Compared with the far-fetched hours stuff, this seems
pretty spot-on to me. If Dell Computer is doing $2 million in sales a
day, that's pushing a billion dollars.  Several travel services are
claiming over $2 million a week in online reservations. Amazon.Com says
it has served more than a million unique customers and, based on sales
from July through September, it's on the road to making $152 million on
an annual basis (and that's definitely still trending up).  Then there
are software and CDs. Throw in electronic stock trades, and $7 billion
doesn't seem such a shocking number.

On a category (not a dollar) basis, the top responses were:

1. Books
2. Software
3. Computer hardware
4. Reports for investing

Not really surprising, either -- it takes a LOT of books to equal a
computer.  The investment-reports ranking is surprising to me only
because more and more brokerages are giving a lot of this information
away to those who have accounts with them.

*Other Nuggets*

- 87 percent of users use e-mail
- 83 percent use the World Wide Web

In terms of actual usage (time spent doing something), the study ranked
the top five categories as:

1. Web
2. E-mail
3. Obtaining information about a hobby or personal interest
4. Obtaining information about a product or service
5. Obtaining general news

In addition to the 56 million users, the study calculated 10 million
former users, who no longer use any service.  The major reasons cited
for this were changes in circumstances or availability (left school or
no longer have access to a computer), and lack of interest or usefulness
to the ex-users.

As mentioned, 42 percent access from multiple locations, but 30 percent
access exclusively from home and 16 percent access exclusively from
work.

My thanks to IntelliQuest's managing director of Internet services, Tom
Fornoff, who was very helpful with providing additional information.


CompuServe's Numbers
====================

Excluding special charges, CompuServe lost about the same amount of
money ($4 million) for the quarter ended Oct. 31 as it did for the
quarter ended July 31.  With special charges, CompuServe reported a loss
of $13.5 million on revenue of $205 million.  The special charges
related to bonuses that will be paid to employees for staying on through
the WorldCom/AOL transition.  The bonuses won't be paid until the deal
closes sometime early next year, but CompuServe has to pay the bonuses
whether the deal goes through or not.

Network Services revenue continued to grow, rising to more than $84
million for the quarter and approaching Interactive Services revenue,
which dropped to $116.1 million in the October quarter vs. $124.2
million in the July quarter.  Including the ever-meaningless NIFTY
licensee accounts (meaningless to CompuServe, not NIFTY), CompuServe was
relatively flat, falling from 5,341,000 subscribers to 5,321,000, for a
loss of 20,000 customers.  But the NIFTY licensee accounts grew by more
than 100,000 (unfortunately, the license royalties are bupkis so it
really is meaningless).  Where it counts, CompuServe lost customers from
July to October -- 101,000 domestically, 29,000 in Europe and 17,000 in
other geographies.  SPRYNET, however, was up 24,000 subscribers.

Most interesting to me was the news that at the end of October only 8
percent of CompuServe's eligible subscribers had elected to go with the
flat-fee service that was offered on Oct. 1, with half of the
subscribers to the $24.95 plan being new users.  Now, I understand why
AOL chief Steve Case wants to keep CompuServe separate.  It's an
apathetic bunch.  Eight percent elect flat-fee, most of the rest have
five "free" hours, and there's no traffic on the network.  There are
lots of customers not doing anything -- including logging on.  Oh, sure,
this happens on AOL too, but even before flat-fee AOL was averaging
around eight hours (or twice what CompuServe is currently averaging).  I
wonder how many companies are paying for accounts for people who no
longer use the service.  Why would AOL want to draw any attention to
that?  CompuServe is spending $43 million a quarter in marketing and
still losing subscribers.  What will AOL have to spend in marketing to
generate growth for CompuServe?

I really do hope I'm wrong, but I still think the CompuServe subscriber
base will churn off -- only it will churn off very slowly so AOL will
definitely leave it alone for a while and get whatever it can out of
that base.  By the time AOL kills CompuServe, AOL probably will have 15
million to 20 million subscribers and it won't have made a difference.
Meanwhile, the current run rate for revenue from Interactive Services
(CSi) is $460 million a year and falling...

David Simoms, Managing Director, Digital Video Investments takes a
look at what losses AOL will inherit at:
< http://www.wellengaged.com/engaged/seidman.cgi?c=online&t=16&q=19 >

Happy Thanksgiving!  See you in 2 weeks.

Stock Watch for the Week Ended Nov. 21, 1997
============================================

Courtesy of InfoBeat's CLOSING BELL < http://www.infobeat.com >.

                                      52 Wk     52 Wk    P/E     Week
SECURITY                    CLOSE     HIGH       LOW    Ratio    CHNG
---------------------------------------------------------------------
AT&T Corp................   55 11/16 54 15/16 30 3/4      18   +16.0%
Amazon Com Inc...........   53 7/8   66       15 3/4            +7.7%
America Online Inc.......   74 7/8   91 1/8   27 3/8            +2.1%
Apple Computer Inc.......   18 3/16  29 3/4   12 3/4            -1.3%
At Home Corporation Ser A   19 1/16  30 5/8   16 5/8            -1.6%
C/Net....................   24 1/4   46 1/2   15 3/4            +1.5%
CMG Info Svcs. Inc.......   21 3/4   28 1/2   10 7/16            0.0%
CUC Intl. Inc............   29 1/2   32 1/8   19 1/4      53    +3.2%
Cmp Media Inc Cl A.......   18 5/8   29 3/8   13 3/4      27    -1.3%
CompuServe Corp..........   13 1/16  14 9/16   8 7/8            +8.8%
Concentric Network Corp..   10       16        9 3/4            +2.5%
Cybercash Inc............   15 3/4   30       10 1/2           +12.5%
Earthlink Network Inc....   19       22 1/2    8 5/8           +16.9%
Excite Inc...............   26 1/64  35        6                -0.8%
FTP Software Inc.........    2 1/2    8 5/8    2 1/2            -6.9%
GTE Corporation..........   48 1/2   52 1/4   40 1/2      16    +5.1%
H & R Block Inc..........   40 3/16  42 1/4   27 3/4      44    +4.0%
Hewlett Packard Company..   63 5/8   72 15/16 48 1/8      22    +3.2%
IBM......................  105 9/16  109 7/16 63 9/16     18    +4.0%
Individual Incorporated..    3 27/32 11 7/8    2 5/8            +4.2%
Infoseek Corporation.....   11 1/2   14 1/2    4 3/8            +1.0%
Lycos Inc................   30 5/8   42        9 1/2            -2.3%
MCI Communications Corpor   43 9/16  43 7/16  27 5/16     44    +4.8%
Mecklermedia Corp........   22 1/2   30       16 1/2      118   +16.8%
Microsoft Corporation....  137 7/8   150 3/4  74 1/8      52    +3.4%
Mindspring Enterprises In   28 5/8   31 1/2    5 1/4            +5.0%
Netcom On Line Communicat   19 7/16  21 7/8    7 7/8            +2.3%
Netmanage Inc............    3 1/16   8 1/2    2 1/2             0.0%
Netscape Communications C   29 3/4   65       23 1/2            -2.4%
Network Solutions Inc. Cl   16 5/16  26 3/4   11 3/4      82    +2.7%
Onsale Inc...............   18       35 1/4    4 5/8            -5.2%
Open Market Inc..........   10 3/8   17 3/4    6 1/2            -6.2%
Oracle Corporation.......   35       42 1/8   22 25/64    49    +1.6%
Psinet Inc...............    7 15/32 14 1/8    5 1/2           +16.0%
Quarterdeck Corp.........    2        7 1/8    2               -11.1%
Security First Network Ba    7 5/8   14 1/4    5 1/4            -3.1%
Silicon Graphics Inc.....   13 3/16  30 5/16  12 5/8      53    -5.3%
Sprint Corporation.......   56 3/8   59 3/8   37 1/2      25    -1.9%
Spyglass Inc.............    7 15/16 18 7/8    6                 0.0%
Sun Microsystems Inc.....   35 11/16 53 5/16  25 1/2      19    +3.0%
Vocaltec Communications L   21 5/8   33 1/4    4 1/2           +10.1%
Worldcom Inc.............   32 5/8   39 7/8   21 1/4            +9.2%
Yahoo Corporation........   52 7/8   58 5/8   11 9/64          +10.4%
Dow Jones 30 Industrials. 7,881.07                              +4.0%
---------------------------------------------------------------------

Subscription Information
========================

To subscribe to this newsletter by e-mail:

Send e-mail to LISTSERV@PEACH.EASE.LSOFT.COM and in the BODY of the
message type:

SUBSCRIBE ONLINE-L FIRSTNAME LASTNAME

Example: Subscribe Online-L Robert Seidman

If you'd like to remove yourself from this mailing list, send a message
to LISTSERV@PEACH.EASE.LSOFT.COM and in the body of the message type:

SIGNOFF ONLINE-L

A Web-based version of the newsletter is available at:
< http://www.onlineinsider.com >
<FONT  COLOR="#0f0f0f" BACK="#fffffe" SIZE=3>

----------------------- Headers --------------------------------
Return-Path: <owner-online-l@PEACH.EASE.LSOFT.COM>
Received: from  relay09.mail.aol.com (relay09.mail.aol.com [172.31.109.9]) by air09.mail.aol.com (v36.0) with SMTP; Sun, 23 Nov 1997 23:51:37 -0500
Received: from VMS.DC.LSOFT.COM (vms.dc.lsoft.com [206.241.12.2])
  by relay09.mail.aol.com (8.8.5/8.8.5/AOL-4.0.0)
  with ESMTP id VAA05235;
  Sun, 23 Nov 1997 21:44:45 -0500 (EST)
Received: from PEAR.EASE.LSOFT.COM (206.241.12.19) by VMS.DC.LSOFT.COM (LSMTP for OpenVMS v1.1a) with SMTP id <8.B18E882B@VMS.DC.LSOFT.COM>; Sun, 23 Nov 1997 21:39:08 -0400
Received: from PEACH.EASE.LSOFT.COM by PEACH.EASE.LSOFT.COM (LISTSERV-TCP/IP
          release 1.8c) with spool id 11613577 for
          ONLINE-L@PEACH.EASE.LSOFT.COM; Sun, 23 Nov 1997 21:40:01 -0500
Approved-By: robert@ONLINEINSIDER.COM
Received: from robertse (dialin-3.white-plains.bestweb.net [208.218.53.38]) by
          miro.bestweb.net (8.8.5/8.8.5) with SMTP id VAA16412 for
          <online-l@peach.ease.lsoft.com>; Sun, 23 Nov 1997 21:39:08 -0500 (EST)
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 4.71.2038.0
X-MimeOLE: Produced By Microsoft MimeOLE V4.71.2039.0
Message-ID:  <000701bcf881$f42709c0$2635dad0@robertse>
Date:         Sun, 23 Nov 1997 21:37:47 -0500
Reply-To: Robert Seidman <robert@onlineinsider.com>
Sender: "Seidman's Online Insider" <ONLINE-L@PEACH.EASE.LSOFT.COM>
From: Robert Seidman <robert@onlineinsider.com>
Subject:      Seidman's Online Insider - November 23, 1997:
To: ONLINE-L@PEACH.EASE.LSOFT.COM

</PRE></HTML>