March 8, 2010
Email deliverability is a term that encompasses all the issues involved with getting your permission based emails to land in your subscriber’s inbox. Improving deliverability involves actions that can be taken on behalf of email marketers to ensure that emails are delivered to the recipients that signed up to receive them. The success of your email marketing campaign hinges on your ability to ensure that these mailings land in your subscribers inbox. The ability to deliver a mailing in a subscribers inbox is almost entirely dependent on how ISPs calculate your Sender Reputation. How they calculate your sender reputation score determines whether your mailing is received in the inbox, the junk mail folder, or even delivered at all.
Sender reputation is associated with the IP address of the mail server you are using to send messages from. ISPs score a sender’s reputation by giving different weight to various factors related to email marketing activity and then use some type of equation to determine your reputation. In essence, your sender reputation indicates to ISP the trustworthiness of the source of the email that is being delivered. What constitutes a trustworthy source will vary from ISP to ISP so in order to build a strong sender reputation you will need to familiarize yourself with all of the factors that ISPs look at when determining a score.
In the past ISPs would block emails based solely on one of these filters but as weaknesses in this approach became inherent they can no longer rely on any one method. Each of the following filtering methods play a role in your ISPs’ delivery decision:
- Content Filtering – Uses the email subject line and the body content to come up with a spam count score. One of the most popular spam filtering packages is SpamAssassin.
- List Quality Filtering – A lot of spam lists particularly paid subscriber lists contain a lot of bogus email which results in high bounce rates. ISPs use this to detect which lists contain bounces that exceed a certain quantity and then disallow all emails from this IP address or sending domain.
- Volume Filtering – Because spammers send bulk emails without regard to accuracy or volume this checks the number of simultaneous connections that are opened (known as threads) at any one time with your ISP. The server may then reject all messages based on the number of open connections.
- IP Address Filtering – When an IP address is added to this filtering list the process is known as “blacklisting.” Once blacklisted all emails from a particular IP address are disallowed.
- Domain and URL Filtering – Checks the email domain used to send a mailing as well as any URLs included in the message.
Whitelisting and Feedback Loops
Being reported as spam by your subscribers will affect your ability to deliver to all your subscribers. A feedback loop is a service provided by ISPs where they will forward complaints made by email recipients to the original sender. ISPs collect these complaints through report spam pages on webmail pages, email clients, or through help desks. Joining feedback loops provides a way for you to clean your list and identify a problem that exists in your marketing strategy, e.g. subscribers do not know they subscribed to your list due to deceptive marketing material. If a subscribers email address appears in other subscriber lists then you can also add them to exclusion lists to prevent them from reporting you as spam multiple times. Every time a subscriber reports you as spam it harms your sender reputation. If you get enough complaints you could be blacklisted or have your site shut down by ISP. Some ISPs provide whitelisting and compaint feedback loops as methods for improving deliverability. Those that do typically require senders to have explicit opt in permissions for subscribers.
- Whitelists -The opposite of a blacklist where the ISP receives requests from legitimate companies to be added to a list of pre-authorized e-mail addresses from which mailings can be delivered regardless of spam filters
- Feedback Loops – A reporting mechanism whereby an ISP provides the sender with data including unsubscribes and spam complaints.
Most people have seen a “Report Spam” button on their email client. When this is clicked on a spam complaint is logged at the ISP level or it is also relayed back to the sender if they are a part of a feedback loop. If you receive too many spam complaints then you will damage your sender reputation. A good way to reduce your spam complaints is to ensure that all of your subscribers are opt-in.
Spam complaints do not track the reason why recipients make a spam complaint. It is possible that even if subscribers opt-in to your list they may forget they are subscribed or they may no longer find that your mailing is relevant. If you receive a high number of spam complaints on a list where your subscribers have opt-in to then this is usually a good indicator that your mailings are either irrelevant or sent to frequently.
In order to establish a Sender Reputation you must first set up some type of Sender Authentication. When people became aware of spam they learned to avoid any emails that came from unknown email addresses. In response spammers learned to hide the true origins of their emails. To combat this sender authentication protocols were established to verify that emails were coming from the addresses that they say they are coming from. This is used to prevent domain forgery and spoofing and provides a framework for helping ISPs to distinguish between legitimate email senders and spammers. ISPs Identifying and verifying a claimed domain name has been authenticated or authorized for sending from a MTA makes it possible to treat suspected forgeries with suspician, reject known forgeries, and block email addresses from known spamming domains.
- Sender Policy Framework (SPF) – a record that allows you to determine which computers can send emails on behalf of your domain. Adding an SPF record to your domain name’s TXT entry, while not required, can help improve email delivery rates by reducing the chance that the emails you send will be seen as spam. It can also help prevent others from sending spam and using your domain name. This is used by Bellsouth, AOL, Gmail, and MSN/Hotmail.
- Sender ID – is very similar to SPF record except this extends the verification process to include the purported responsible address included in the header. Used by MSN/Hotmail
- Domain Keys – an authentication standard that is designed to verify the DNS domain of email sender and the message integrity. All outgoing emails are digitally signed with a private encryption key to match a public key that is published in the sender’s DNS record. Used by Gmail, Yahoo, SBCGlobal, British Telecom, Rogers Cable, Rocket Mail, etc.
- DKIM – an enhanced authentication standard that allows a person to verify that a message comes from the domain that it claims that it came from.
Your system admin should be able to assist you with ensuring that the following technical configurations are in line as they can improve or harm your sender reputation.
- IP Address – because email originates from this address you need to establish a low history of spam complaints, spamtrap hits, and low bounce rates in order to have a positive reputation that will affect your long term deliverability. If you wish to qualify for whitelists, feedback loops, and reputation services, your IP address must have low spam complaints, unsubscription management, and proper setup for the domain associated with it.
- Sending Domain or Subdomain – Domain registration and domain age are two factors for establishing a positive reputation. Newly registered domains are regarded with suspicion as spammers often hop from domain to domain. If a sending domain has a bad online reputation it will result in low deliverability rates.
- RFC Compliance – these are information documents used as governing standards for internet traffic. RFC 2821: Simple Mail Transfer Protocol and RFC 282: Internet Message Formats relate to email reputation.
- Reverse DNS – used to identify the domain name associated with an IP address. The IP address is the only data that can not be forged and not having this enabled is in violation of RFC standards and a requirement for many ISPs. If this is not enabled or is configured improperly you must immediately contact your server admin.
- Bounce Management – An email address is considered dead and should be removed from your list if it bounces 3 consecutive times or if the time between the most recent consecutive delivery rejection is in excess of 15 days.
Tariq Ahmad is a email delivery & marketing specialist at ActiveCampaign – a web software company that develops email marketing software, survey software, and help desk software. ActiveCampaign devotes a lot of resources to email marketing testing, deliverability research, and general market research to help businesses succeed with their online marketing.