At 6:12 AM -0800 1/30/97, Brad Knowles wrote:
>It's just the nature of the beast of large
>communities -- we don't have any more than our share of clueless
>customers (not much, anyway ;-), it's just that because we've got
>eight million customers, we've got a lot more total clueless people
>than other places.
Hate to say it, Brad, but as someone who's defended AOL in the past, I
have to now say -- I don't think so.
Right now, I'm doing an address verification probe over on
solutions.apple.com, sending out mail to each subscribed address to
check for bounces. So far, I've sent out about 22,000 of the 52,000
e-mails (I've got the process throttled to avoid overloading any
server, including my own, so mail is going out once every 7-10 seconds.
It'll take about 5 days to finish the entire probe).
As part of the message, I say, quite explicitly and in capital letters,
DO NOT REPLY TO THIS MESSAGE. I also give an address that people can
mail to if they have questions/problems/need help, whatever.
So far, just FWIW, my bounce rate is about 2%. Given that I try to keep
my mailing lists pretty clean from bounces, I find that number
fascinating, because there seems to be an awful lot of bouncemail
that's not getting back to my server. Some servers just seem to not
send errors back to bulk priority at all, or do so intermittently, so
this standard-class probe is getting a bunch of things sent to it.
Of course, a number of people are using this as a hook into getting
help, either to update addresses or to get off the lists. That was part
of it (and, frankly, the "DO NOT REPLY..." was *just* to see who reads
and follows instructions and who doesn't. I was curious).
I have picked up about 250 messages in response to this probe of an
administrative nature -- that means wanting addresses changed or
wanting to be unsubscribed. That ignores vacation bots and the like,
which I'm seeing a lot less of than I'd expected.
Of that 250 (roughly) messages so far, about 60% are coming to the
requested address. The other 40% are from people replying despite the
instructions and returning messages to the bounce-test mailbot.
Now, for the interesting part. Of that 40% ignoring the instructions, a
solid 80% of those are AOL people. Only about 20% of the AOL accounts
sending me mail on this followed that instruction.
So, I've sent about 3,500 e-mails to AOL accounts, and 18,500 e-mails
to non-AOL accounts. My response is about 120 to the wrong address, of
which about 95 were AOL addresses. And about 140 to the requested
address, of which about 25 were AOL addresses.
Right off the bat, there's a discrepancy: AOL responses are about 3.5%
of the subscriber base, while for the rest of the net, it's only 0.7%.
80% of those responses were misdirected, either because the users
didn't bother to read the instructions or chose not to follow them,
while for the rest of the net, that number was quite small. I don't
have to run this through a stat package to see it's statistically
significant -- AOL users under identical situations are a *LOT* worse
at reading and following instructions. They don't bother, or they
can't, or whatever. A few, who's responses ranged from haughty to
obnoxious to one rather abusive twit sent responses that boiled down to
"I don't care what you say, I'm responding here, and dammit, you do
what I tell you to do."
And the responses I got back from AOL made it clear these folks don't
know how to operate the machinery. Period. Despite my system sending
instructions on subscription, and re-sending them to the list every two
weeks, these people in many cases had no clue how to unsubscribe, had
no clue how to get help, never bothered to read any of the
instructions, don't know basics like e-mailing postmaster for help --
nothing. And when the information *is* presented to them, there's a
high level of simply choosing to ignore it, doing what they want, and
assuming it'll work anyway.
I'm *not* seeing this from Compuserve (also probed), or earthlink
(whihc is probably a very distant second in cluelessness so far), or
any other service. It's kinda scary -- AOL users don't read the
instructions, don't learn the basics,a dn then just sit back and TAKE
IT in a very passive way, or they lash out and start doing random
things, like screaming at the list, screaming at users posting to the
list -- anything but the right thing. And from my feedback of the last
couple of days, they're not terribly interested in learning. They want
pack slaves to do all this for them. I'm *not* seeing that, except for
a very isolated person here or there, anywhere else, but it's fairly
common in the culture of the users I'm working with on AOL right now.
This is not good. AOL has a huge amount of work to do to clean up their
user base. Statistically, they *are* much different from the rest of
the net, and with a high level of cluelessness and a low interest level
in learning this stuff. And there's enough arrogance mixed into the
mail I'm getting ("I don't know, I don't wnat to know, it's your job to
fix this for me") that I'm not interested in defending AOL much right
now. They're different, in very negative ways, and if I were to roll
these numbers up formally and identify this in detail, I can prove it
statistically.
I'm not going to bother - the general trends are enough for me. But you
ought to know about this, since this is fairly good data that I'm
generating in a concentrated form with a good control against AOL,
since I'm annoying users across the world equally here... The fact that
something like an order of magnitude more AOL users by percentage of
subscribers are responding is a huge warning sign all on its own --
because they're responding because they don't have the *basic*
information needed to even find a post master or *ask* for help, until
some address pops up they can latch onto. That's scary, or it should
be, especially since that information is dumped into their mailbox on a
regular basis, if they'd only read it. But the other scary thing is
they're proving quite conclusively they don't bother to read
instructions, and then wonder why things don't work....
Anyway, just some data to chew on. It turns out, from what I can see,
that AOL users are different than the rest fo the net, not in positive
ways. AOL's *not* just bigger. They're doing a rotten job of educating
their users on how to use these internet tools, and their users are
doing a rotten job of using the tools we give them, starting with
instructions and help files. And I see a general lack of tacking
responsibility for themselves and expecting others to take
responsibility for them in their responses to me, which also aren't
showing up from any other service....
Use, or ignore, as you wish... Data's always fun, because everyone can
interpret it differently. This is mine...
--
Chuq Von Rospach (chuq@apple.com) Apple IS&T Mail List Gnome
<http://www.solutions.apple.com/>
Plaidworks Consulting (chuqui@plaidworks.com) <http://www.plaidworks.com/>
(<http://www.plaidworks.com/hockey/> +-+ The home for Hockey on the net)
Follow-Ups:
References:
|
|
