Child pages
  • .Maximizing deliverability for your emails v9.0.0

Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

  • Authentication
    Email authentication is a collection of techniques aimed at equipping messages of the email transport system with verifiable information. Its purpose is to validate the identities of the parties who participated in transferring a message, as they can modify the message. The results of such validation can then be used in delivery decisions that do not imply any "content filtering" mechanism. There are several authentication methods (FCrDNS,  ADSP, SPF, DKIM, DMARC) and each of them go one step further in assessing sender validation and reputation.

  • Commercial email
    An email message that has as one of its purposes that of participation in a commercial activity. Commercial email typically promotes a product or service.

  • Transactional email
    A one-to-one email message, usually sent as a result of a specific user's action (e.g. online order). Transactional emails are topically receipts, confirmations, reminders, or any non-commercial email personalized specifically to the recipient. Legally speaking (e.g. CASL), a transactional message that includes commercial content becomes a commercial email (see above).

  • Content Filtering / Fingerprinting
    Techniques aimed at evaluating the content of an email message (specific words, URLs or "chunk") to detect spam messages or spammy patterns. Content filtering is more utilized in the corporate world where system administrators may set content restrictions on what employees can receive. In the consumer world, instead, major ISPs (e.g. Google, Yahoo!, etc.) see authentication, reputation and user interaction (see "Engagement") as more reliable than content filtering in detecting spam, though they may use both.

  • Engagement
    A "measure" of how subscribers respond to and act on the email messages you are sending to them. Positive engagement - such as views, time spent viewing, clicks, and replies - will likely improve the sender reputation and deliverability. Negative engagement, meant as the absence of any action or, worse, deleting or marking a message as spam, is very likely to cause deliverability issue in the short term.

  • Envelope Sender / Return Path
    An e-mail address to which asynchronous bounce messages are delivered. As this email address is included in the Header section of the email, its domain takes part in the reputation assessment process.

  • Dedicated IP
    An IP address used exclusively for one sender or for a portion of its email traffic (e.g.. transactional emails). When a dedicated IP is used, email traffic being sent from that IP address is isolated to that specific sender. Consistent sending frequency - and of course high quality of the messages being sent - are crucial factors in building and maintaining a good reputation for dedicated IP addresses. Lack of sending volume and/or frequency can cause lack of reputation for the dedicated IP, which can lead to deliverability issues. For this reason, a dedicated IP address may or may not be a recommended solution.
     

  • Shared IP
    A group of IP addresses used for multiple customers that share common reputation metrics and allow them - as a whole - to maintain a consistent sending frequency.


  • Rate limiting / Throttling
    Rate limiting is the process that ISPs use to delay the delivery of unwanted (or unknown) email, filter spam, and ensure that wanted (e.g. transactional) emails reach the inbox in a timely manner. Each ISP has its own sending limits on a per-hour and/or per-day basis, and they can throttle the sending volume when it’s too high or too low.

     


  • Domain / Apex domain
    The right portion of the domain used by a sender when sending emails (e.g. mycompany.com). It is the root of all reputation and authentication mechanisms, and should be directly linked to the sender's corporate web site or brand identity.
     

  • Subdomain
    A lower-level domain. If mycompany.com is the top-level (apex) domain, news.mycompany.com is a subdomain of it. Since usually the apex domain is already configured to properly serve a sender's corporate web site, and any modifications to it could have unwanted side effects, it is usually recommended that a sender create subdomains to be used for email messaging purposes (3rd level domains such as news.mycompanyname.com and 4th level domain such as bounce.news.mycompanyname.com). The choice of domain, sub-domain, and naming conventions is important because it can have a significant effect on how ISPs and anti-spam authorities will consider the email stream. Please see Configuration steps below for more information.
     

  • Web interface domain:
    A subdomain that will be used:

    • in all tracked links in your email messages;

    • in the URL of the Web version of the message;

    • in the URL of all Web pages used by the system (e.g. subscription confirmation landing page);

    • in your MailUp admin console URL.

...

  1. Pick the FROM domain
    Which domain will you be using to send emails with MailUp? Your top-level domain (i.e. the apex domain as discussed above) or a subdomain (e.g. news.mydomain.com)? In the first scenario, the FROM EMAIL would be something like updates@mydomain.com, whereas in the second it would be something like updates@news.mydomain.com. The decision should be based on whether you have access and can modify the DNS records of that domain. Check with the person in your organization that has access to your domain management system to find the answer. In the examples below we are assuming that the sending domain corresponds to the apex domain (mydomain.com). If you cannot modify the DNS records of your apex domain, then you will need to set up a subdomain (eg news.mydomain.com) and refer to that one (in place of mydomain.com) in the steps outlined below.
     
  2. Verify your FROM EMAIL
    Now that you have picked the FROM domain, create a FROM EMAIL under that domain, and verify it in your MailUp account. To prevent abuse, MailUp requires that the FROM EMAIL is verified before it can be used. Verification is very simple: MailUp will send a verification message to the provided FROM EMAIL address, and you will need to click the link contained in the message. You can verify the FROM EMAIL when you configure a List in your MailUp account, when you set up a new mailing or when you add a new From email in Deliverability checkup page.
     
  3. Configure the SPF record for the sending domain
    Adding SPF authentication is easy. Here is what you need to do:

    • Contact your Web hosting company, domain registrar, or network administrator that manages this domain
    • Tell them that you need to make a change to the DNS (Domain Name System) records
    • If you are not already publishing an SPF record, ask them to add the following TXT record:  

    v=spf1 include:musvc.com ~all 

    • If you already have an SPF record in place (e.g.: you have a TXT record starting with v=spf1) then you should only add the "include:musvc.com" before the final "all" keyword

    Example: v=spf1 include:mydomain1.com include:mydomain2.com include:musvc.com ~all 

    • Wait 24-48 hours: it takes a bit of time to changes to the DNS to propagate around the Internet
    • Run the SPF test in Deliverability checkup to confirm that the SPF record has been successfully updated.

  4. Enable DKIM authentication
     Adding DKIM authentication is easy. Here is what you need to do:
    • Contact your Web hosting company, domain registrar, or network administrator that manages this domain
    • Tell them that you need to make a change to the DNS (Domain Name System) records
    • Ask them to create the following two CNAMEs (replace "mydomain.com" with your domain)
      (1) ml01._domainkey.mydomain.com 
     
    • ... and point it to
      ml01.dkim.musvc.com
      (2) ml02._domainkey.mydomain.com
     
    • ... and point it to
      ml02.dkim.musvc.com
     
    • If a CNAME cannot be created, you may also establish DKIM authentication by adding the following TXT records to the DNS settings. Please contact us for additional details
     
    • Wait 24-48 hours: it takes a bit of time to changes to propagate around the Internet
    • Run the DKIM test in Deliverability checkup to confirm that the CNAMEs have been successfully updated.

  5. Configure a Web interface domain (optional)
    If you wish to use a custom Web interface domain (see the Glossary above for a definition), create a C-NAME in your domain management system (e.g. news.mydomain.com) and point c.mailup.com
    For more information, please see MailUp account settings. Please note that this configuration is available only for PRO and ENTERPRISE clients. 

  6. Configure a custom Envelope Sender (optional)
    Using a custom Envelope Sender (see the Glossary above for details) you can to "align" it with the FROM EMAIL address, which allows for more advanced sender configurations, as mentioned above. This address can address can be any email account of your choice under the same domain a subdomain as the one used for the FROM EMAIL (e.g. if the FROM EMAIL is news@mydomain.com  the Envelope Sender could be bounce@mydomainbe bounce@bounce.mydomain.com). In order for the MailUp system to be able to process bounces, it will need to access sent to that address. There are two options:

    1. You may provide us with POP3 access to that mailbox (port 110, PLAIN no TLS or SSL)

    2. You may forward messages received by that mailbox to a specific email address that we will provide

  7. PTR of SMTP servers (For dedicated IPs):
    If your email streams will be delivered through dedicated SMTP servers, each one of them should have a PTR aligned with the base host domain. Example: 

    mx67202.mydomain.com A 93.174.67.202
    mx67203.mydomain.com A 93.174.67.203

    Each PTR should have the same SPF / Sender ID records as the sending domain:
    mx67202.newsletter.mydomain.com TXT v=spf1 include:musvc.com ~all 
    mx67202.newsletter.mydomain.com TXT spf2.0/pra include:musvc.com ~all
     


  8. Enable DMARC
    Since DMARC is built upon SPF and DKIM all the previous steps are required before enabling DMARC.

    The proper TXT record (_dmarc.mydomain.com) should be added to the DNS settings for your sending domain.
    It can change depending on what you want your DMARC policy to be.

    A simple DMARC record is the following: v=DMARC1; p=quarantine; pct=100; rua=mailto:dmarc.rua@mycompany.com; ruf=mailto:auth-reports@mycompany.com.
    where:
    * v is the version, DMARC1 is the only version available at the moment.
    * p is the policy. Allowed values are *none* (take no action, just collect data and send reports) *quarantine* (treat with suspicion unqualified mail) *reject* (block any unqualified mail for the domain)
    * pct is the percentage of non-aligned messages that should be rejected (from 1 to 100 where 100 means all the messages)
    * rua: Send aggregate reports to this address (should be closely monitored)
    * ruf: Send forensic (detailed) reports to this address.

    Note that the email addresses that receive the aggregate and detailed reports (“rua” and “ruf”) can be on any domain, not necessarily the domain used for the authentication, for reporting purposes only.

    We strongly suggest ramping up DMARC use slowly by using the p=none policy at first. Monitor your traffic and look for anomalies in the reports (eg.: messages that are not yet being signed)
    Then, once you have verified that all legitimate messages are correctly being authenticated, move to "quarantine."
    Review the results again (look also in your spam folder) and when you're absolutely sure all of your messages are signed, change the policy setting to "reject" to make full use of DMARC.

    You can also leverage the pct tag to sample your DMARC deployment. If you want to be extremely conservative, after moving to the quarantine policy, you may start with pct=1 and then move to 10, 25, 50, 100...