As a veteran of the so-called permission marketing industry, I have offered
my pithy comments to the ones made below.
So. Let's say you are that company. You have tens or hundreds of
thousands of clients with millions or tens of millions of email
addresses.
[ ... ]
What kind of system will make this all work with maximum success?
Perfection is not a goal, it's not possible. What's going to give
optimal results?
And don't say "confirmed opt-in". That's a policy. Maybe right,
maybe wrong. But anyone can say it. How are you going to enforce
compliance with the above list managers?
You raise some good points (most of which I've snipped). However, I
will say "confirmed opt-in". And I would enforce the compliance in
part by using complaints and DNSBL listings.
1: Actually, I will say "confirmed opt-in with yearly renewals". Every
new list you get, you require all list members to "renew" their
subscription to the mailing list and they must continue to renew their
subscription every year or so. Save these confirmations and renewals
so that if there is any dispute, you have something to show that you
are not sending unsolicited email. Unlike initial subscriptions, I
think it would probably be reasonable to send maybe two emails about
the renewal. Just like initial subscriptions, the renewal request
must not contain any advertising and the description of the mailing
list should be brief and approved by the bulk mailers.
A: This is not going to happen. Marketers spend many years and many
dollars to build legitimate lists through their websites and POS and they
will not voluntarily chop it in half just because someone didn't 're-up'.
Many customers who get email newsletters selling products buy something once
or twice a year, and often don't open the messages for months at a time, but
continue to want to subscribe - they buy seasonally for example. This
approach is not practical and has no chance of being adopted.
2: Bulk mailers should get as many IP addresses as possible, preferably
some in different CIDR blocks. New customers (not subscribers) with
their own lists would be put on "high risk" IP addresses that DNSBLs
may quickly blacklist. As time passes and the number of emails sent
to the list increases without complaints, you can move the lists to
lower risk IP addresses. How many different pools of risk levels you
need will have to be determined by experience.
A: The legitimate email marketing houses already use either individual
netblocks, MTAs or subdomains (or a combination) to separate their clients
out. This is a very effective way to make sure that high risk clients can
be isolated. Moving them around different netblocks as you describe would
only make it look like something fishy was happening with the list. And
enough complaints would get the client fired, most likely.
3. Bulk emailers need to watch DNSBLs and abuse complaints very
carefully. If complaints start coming from a particular mailing list,
then they need to moved into higher risk IP blocks or segregated into
an IP address by itself.
A: They watch spews and other dnsbl's very closely. Getting off of
those lists is very difficult. More important to most are relationships
with top ISPs, those being Yahoo, MSN, AOL, Earthlink, etc.
4. New customers should be required to physically sign a contract, and
place a "large" damage deposit which will be refunded when the mailing
list has reached a low enough risk category. Maybe a $1000 +
$0.01/email deposit would work.
A: No customer thinks its list is dirty. If you think that their list
is dirty enough to consider escrow, it is best not to work with that client
in the first place.
5. The bulk emailer could then almost advertise about how their low
risk
customers are rarely, if ever, placed on DNSBLs. Actually, this kind
of procedure would probably scare away most spammers also, so the bulk
emailer would mostly have to deal with people who don't quite
understand email rather than people who are outright trying to deceive
them.
A: This happens organically by not being a spammer, but still sending
bulk commercial 'permission' mail.
6. Basically, I think that nothing really needs to be changed at the
technical level. It is all a matter of bulk emailers implementing
policies that keep them out of trouble rather than send the most
email.
A: You are wrong. The legitimate mailers that you target with these
proposals typically already do this right. Spammers do not adhere to these
rules and will not. These spammers are the ones that this group wants to
shut down. They do not comply with 'best practices' that we promulgate
here. That is why legislative and other sociological approaches will not
work - only technical solutions will. Economic approaches are a very close
second, since spam is really an economic issue at its heart.
7. Some might complain that this type of system will cost a lot more
money, but I'm not so sure that is true. The low risk lists will cost
the bulk emailer below average costs and they will probably be able to
charge above average fees and that can pay for the expense of making
sure the lists are clean.
A: The email marketing business is highly competitive, making this final
assertion/suggestion also less than practical.
Dave Hendricks
Permission Technology
Suite 100
1261 Post Road
Fairfield CT, 06430
203 254 7222
203 254 9222
dave(_at_)permissiontechnology(_dot_)com
www.permissiontechnology.com
-----Original Message-----
From: asrg-admin(_at_)ietf(_dot_)org [mailto:asrg-admin(_at_)ietf(_dot_)org] On
Behalf Of wayne
Sent: Wednesday, April 30, 2003 10:12 PM
To: IRTF Anti-Spam Research Group
Subject: Re: So what's the *right* way to be a bulkmailer? (Was RE: [Asrg] A
New Plan for No Spam / DNSBLS)
In <p06000f13bad47eac81f5(_at_)[192(_dot_)168(_dot_)1(_dot_)104]> Kee Hinckley
<nazgul(_at_)somewhere(_dot_)com> writes:
So. Let's say you are that company. You have tens or hundreds of
thousands of clients with millions or tens of millions of email
addresses.
[ ... ]
What kind of system will make this all work with maximum success?
Perfection is not a goal, it's not possible. What's going to give
optimal results?
And don't say "confirmed opt-in". That's a policy. Maybe right,
maybe wrong. But anyone can say it. How are you going to enforce
compliance with the above list managers?
You raise some good points (most of which I've snipped). However, I
will say "confirmed opt-in". And I would enforce the compliance in
part by using complaints and DNSBL listings.
Actually, I will say "confirmed opt-in with yearly renewals". Every
new list you get, you require all list members to "renew" their
subscription to the mailing list and they must continue to renew their
subscription every year or so. Save these confirmations and renewals
so that if there is any dispute, you have something to show that you
are not sending unsolicited email. Unlike initial subscriptions, I
think it would probably be reasonable to send maybe two emails about
the renewal. Just like initial subscriptions, the renewal request
must not contain any advertising and the description of the mailing
list should be brief and approved by the bulk mailers.
Bulk mailers should get as many IP addresses as possible, preferably
some in different CIDR blocks. New customers (not subscribers) with
their own lists would be put on "high risk" IP addresses that DNSBLs
may quickly blacklist. As time passes and the number of emails sent
to the list increases without complaints, you can move the lists to
lower risk IP addresses. How many different pools of risk levels you
need will have to be determined by experience.
Bulk emailers need to watch DNSBLs and abuse complaints very
carefully. If complaints start coming from a particular mailing list,
then they need to moved into higher risk IP blocks or segregated into
an IP address by itself.
New customers should be required to physically sign a contract, and
place a "large" damage deposit which will be refunded when the mailing
list has reached a low enough risk category. Maybe a $1000 +
$0.01/email deposit would work.
The bulk emailer could then almost advertise about how their low risk
customers are rarely, if ever, placed on DNSBLs. Actually, this kind
of procedure would probably scare away most spammers also, so the bulk
emailer would mostly have to deal with people who don't quite
understand email rather than people who are outright trying to deceive
them.
Basically, I think that nothing really needs to be changed at the
technical level. It is all a matter of bulk emailers implementing
policies that keep them out of trouble rather than send the most
email.
Some might complain that this type of system will cost a lot more
money, but I'm not so sure that is true. The low risk lists will cost
the bulk emailer below average costs and they will probably be able to
charge above average fees and that can pay for the expense of making
sure the lists are clean.
-wayne
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg