• View
• Edit
• History
• Print

Mail Set-Up

In addition to using best opt-in and bounce handling practices, how you send your mail can impact your delivery.

Many ISPs block emails because they cannot verify the sender’s identity, or because of some other error on the part of the sender. There are some basic things you can do to avoid this:

• Have proper forward and reverse DNS set up for your domain
• Always use a static IP address
• Have an SPF record (this is somewhat controversial, but it will rarely hurt, and can help)

Reverse DNS

The Domain Name System, DNS, is how computers identify and find each on the Internet, somewhat like a telephone directory. To contact another computer, a DNS lookup would take the domain name and check which IP addresses are associated with it.

Spammers often use false domain names or IP addresses to hide where their spam is coming from – so in order to try to foil this, ISPs will perform a “reverse DNS lookup” on incoming email. This takes the connecting IP address, and checks to see if a registered domain name is associated with it, and if so, if it is the domain present in the email headers. If not, the message may be a forgery or just a sender without a correct DNS entry. Either way, the message will probably be blocked as spam, which is why reverse DNS enabling is important if you’re going to be sending mail.

A good resource for checking whether your rDNS is working properly, and for help in setting it up and troubleshooting it, is DNS Stuff.

Your IP Address

Your IP address can be a source of problems for several different reasons:

- If you inherited it, and it turns out to be already blocked in the anti-spam blacklists because someone who had it before you used it for spamming. This is a legitimate block by the receiver, ISP or spam filter, but a problem for you to get this changed.

- Your IP address is already blocked because you spammed from it! Or someone thinks that you did. Whether you did or not is irrelevant, because either way you need to fix the situation.

- Your IP address is in the same IP block as a spammer's IP address, and the entire IP block is blacklisted because the ISP or host who controls the IP addresses won't remove the spammer. Yes, this can and does happen! Sometimes the only thing an email sender can do is either convince their own ISP to get rid of the spammers, or take their email sending business elsewhere and use another ISP – which affectively will get you a new IP address.

Static IP Addresses

Static IP addresses are important because these are used by web servers and email servers and affectively give you a complete DNS record.

Dynamic IP addresses, by contrast, change every time you connect to the network or Internet. They’re typically used by personal computer users with dial-in or DSL accounts. You can’t normally make a complete DNS record for a dynamic IP address, so they’re rarely used for email servers. You even find that some mailing list management software (e.g. ListManager) won’t function with dynamic IP addresses configured.

So why is it such a big deal? Most ISPs won’t accept email directly from email servers with dynamic IPs, as the messages are highly likely to be either spam, or from a virus-infected computer. This is why ISPs tell their dynamic IP customers to use the ISP’s own outbound (fixed IP) mail server. If you attempt to end run this, and set up your own mail server through your dynamically-assigned IP address connection, expect to have your email blocked.

Your Sending Name

Research shows that email recipients care more about who an email is from than the subject or content. When deciding which emails to delete, open or complain about, the sender is the single most influential deciding factor. This means that more people open an email when the

- From name

- From address, and

- Subject line

are all appropriately branded. As well as being opened more often, these emails are also reported less as spam.

Where the Email is Coming From

Non-existent email addresses

It isn’t just spammers who send email from non-existent email addresses – completely legitimate companies mistakenly do this too. You may see a registration confirmation or message which says "Do not reply to this email, instead contact us at foo@bar.com, and in these cases the "from" address of that email is often not a real email address which can actually be verified by the receiving ISP (for example in this case it might be registration@bar.com.

This is an unwise practice, guaranteed to cause you trouble. It’s a sure sign of spam to a receiver, and highly likely to cause you deliverability issues.

Blocked Email Service Providers

Using an Email Service Provider (ESP) which is being blocked for sending spam can happen, and unless you do careful checks and due diligence when choosing your email sending service, you could get nipped by this problem. These services operate by you giving them your content, and they email it out for you – and there are some disreputable ones out there! There are also some highly reputable ones. Take care in your choosing. Generally speaking, you can with confidence use any of the Email Service Providers that are ISIPP SuretyMail accredited. (See http://www.isipp.com/suretymail.php)

Viruses

Computer virus writers have written viruses that use an infected computer as a spamming machine. Once infecting the computer, the virus takes the user’s address book and finds email addresses, and then sends out its virus and/or spam message to everyone in the address book. That could get your email or IP address blocked if reported as spam.

In addition, on receipt of one of the infected messages, opening the message may infect the receiver’s computer. Thus to disguise the source of the infection, and trick people into opening the message and unleashing the virus, often these viruses send out the spam or virus emails from an email address that it gets from the address book, so the emails are sent out in someone else’s name as if it came from their email address. Again, if reported as spam you could be blocked.

SPF

SPF authenticates that the domain from which the email is purportedly being sent is in fact a domain authorized to send through the IP address and attendant mail server. An SPF record is the specific DNS entry that lists which mail servers are associated with which domains and IP space.

At present SPF is the most widely deployed of all of the existing email authentication mechanisms. Although SPF got a bad rap when spammers also adopted it, it did not aid the spammers, and SPF was never considered to be an anti-spam mechanism.

Some, but not all, ISPs require SPF implementation in order to be considered for their whitelisting programs.

It’s fairly easy and straightforward to set up an SPF record, and it adds authentication to your outgoing email, increasing deliverability, and also helps to protect your brand against phishing. To set up an SPF record, visit

http://spf.pobox.com/

or

http://www.openspf.org/wizard.html?mydomain=&x=19&y=9

You can check your SPF entry at:

http://www.dnsstuff.com/pages/spf.htm

Sender ID

Sender ID also authenticates an email sender at the domain-name level, but it is different from SPF in that it looks at slightly different information. As with SPF, Sender ID confirms that the server used is authorized to send out email on behalf of the domain name listed. But it doesn’t verify the “envelope address” as SPF does, it examines the PRA – the purported responsible address, or the “message from” address inside the message.

Sender ID is Microsoft’s own email authentication protocol, although it incorporates and extends aspects of SPF.

Sender ID was developed to help prevent email spoofing and phishing, and while it’s controversial, Microsoft uses it to verify email incoming to MSN and Hotmail accounts, so it’s not a bad idea to be Sender ID compliant if you’re sending emails to recipients at MSN or Hotmail addresses.

Note: A properly formatted SPF record will serve for both SPF and Sender ID.

DomainKeys

DomainKeys is the result of a partnership between Yahoo and Cisco. It’s an encryption-based authentication method which requires new software, and sometimes hardware, for both senders and receivers.

So how does it work? DomainKeys includes an encrypted digital “signature” at the end of each outgoing email message. This digital signature enables a recipient to:

Verify the sender’s domain name

Confirm if the message has been altered in any way

Match the “from” address to the domain name of the sender (to prevent forgeries) and

Trace the message back to the sender’s domain name. This point is critical as it may deter spam as spammers simply don’t want to be traced.

How is this different from SPF and Sender ID?

DomainKeys can have expiration dates set, and they can be replaced or revoked, so an email service provider could give every user of their system their own unique DomainKeys profiles, and modify them whenever necessary, e.g. if they’re caught spamming.

DomainKeys is intended for ISPs and email servers, but could be tailored for use by individuals – for example your ISP may not use DomainKeys technology, but you could add it as a filter in Outlook, Thunderbird, etc. as an end user.

There’s even discussion about building “sender reputation” into DomainKeys, so if you are a persistent spammer, that information might be built into your DomainKeys profile, so ISPs could block your mail even if it passes their regular checks!

For more information on DomainKeys, visit:

http://antispam.yahoo.com/domainkeys

Failure to Comply with CAN-SPAM

Another factor in non-delivery of emails is simple failure to comply with the Controlling the Assault of Non-Solicited Pornography and Marketing Act of 2003, or CAN-SPAM. But it isn’t really that hard to comply, so make sure that all your messages do!

Email must contain a valid return address

Your physical address must be clearly displayed

Emails must truthfully identify the sender and subject matter in the headers

All commercial mailing list email must provide an unsubscribe option which must be available for 30 days

Removal of unsubscribes must take place within 10 business days

Other ways to stay on the right side of CAN-SPAM include:

Gathering email addresses for mailing lists only from reputable sources

Never harvesting email addresses from web sites or other sources

Not selling your email lists to third parties