This document will introduce you to the concepts and best practices related to deliverability as well as the tools offered by Adobe Campaign Classic to optimize sending your deliveries.
Last update: 2019-05-30
Deliverability or how to measure the success of your campaigns reaching your recipients' inbox without bouncing, or being marked as spam.
Adobe Campaign Deliverability is a paid service that comes available in different offers. Contact the deliverability or commercial service.
The deliverability rate depends on numerous factors, particularly:
Maintaining your domain and IP address reputation is essential. Here is some advice for setting up a new platform:
To improve your email deliverability rate and make sure that your emails reach your recipients, the email must respect a certain number of rules:
To avoid exhausting your recipients, delete duplicate email addresses from your target. This step protects your sending reputation and ensures good quarantine management. Adobe Campaign offers the necessary tools to implement these recommendations and avoid the risk of being blacklisted by the ISP.
To avoid duplicates as much as possible, the following actions must be carried out:
Quarantine management is presented in the detailed documentation.Read More
The main internet service providers (ISPs) and Web mail providers manage whitelists from recognized email message senders. Adobe Campaign helps you with the process of getting certified on the best whitelists.
There are two types of whitelists:
In addition, there are several techniques, configurations, and tools that you can use to improve your deliverability rate:
For more information, refer to the detailed documentation.Read More
Use the tools offered by Adobe Campaign to monitor your platform's deliverability.
The Deliverability package gives you access to:
You can also use the Delivery throughput report which gives you an overview of the entire platform's throughput for a given period.Read More
Are you experiencing a deliverability problem? You may find the solution here...
To optimize the deliverability of your Adobe Campaign emails, we recommend using the best practices listed below. Deliverability problems are generally linked to measures of protection against spam implemented by internet service providers and mail server administrators.
Email deliverability refers to the set of characteristics that determine a message's ability to reach its destination, via a personal e-mail address, within a short time, and with the expected quality in terms of content and format. These characteristics fall into four main categories: data quality, message and content, sending infrastructure, and reputation. Together, they form the foundation of a successful email deliverability program.
The deliverability rate is the number of sent emails that were successfully delivered to its recipients.
Here is a list of the key points to check to ensure good deliverability.
Spammers try to conceal their real identity and as a consequence make their servers difficult to identify. A legitimate network configuration that does not try to hide the identity of the server is essential to sending emails in large volumes.
Spammers often use address generators based on lists of frequent names and first names; in addition, they rarely process technical notifications sent back by mail servers. A high rate of invalid addresses is often interpreted as a sign of spam. Double opt-in mechanisms and effective handling of technical bounce messages make it possible to avoid this.
ISPs usually have a prominent means of reporting a received message as spam. This makes it possible to identify unreliable sources. By rapidly honoring opt-out requests, making regular use of a given list, verifying consent through a double opt-in system, and implementing feedback loops, you can reduce complaint rates.
ISPs and other organizations (refer to http://www.projecthoneypot.org/) make use of mailboxes that do not correspond to physical persons but are created simply to trick spammers. These so-called "honey pot" addresses are published on the Web in order to be collected by spambots and thus catch illegitimate senders. The use of a double opt-in mechanism precludes this sort of address being added to a list. When using a third-party list, you must be sure of the methods employed by its maintainer.
To a lesser degree, the content of certain messages can lead certain filters to detect it as spam. The use of certain words, the use of exclamation points in the subject line and within the messages are read as tell-tale signs of spam. Spammers are also known to replace text with images to stop offending text from being analyzed automatically by anti-spam filters. In response to this, a message (in HTML format) with a high proportion of images, or images as attachments, may end up being blocked.
Spammers make programmed deliveries to maintain their reputation over time. They sometimes need to adapt their marketing plan to meet the best practices imposed by the ISPs and so, after a peak in reputation (ramp-up), they configure regular deliveries.
Starting to send emails on a new platform is a sensitive step because the platform does not have any history of use and no reputation (when the sending IPs have never been used for this purpose). ISPs are naturally suspicious of IP addresses that have never been used to send email and that suddenly start to send large volumes of email traffic. In effect, spammers generally use "unknown" IP addresses (that is to say addresses that have never been blacklisted) to send the largest possible number of messages before detection.
You cannot expect to reach operational speed in terms of output at the very start of the production phase. Furthermore, you should not attempt to send messages at this rate as it might lead the ISPs to block the sending addresses and to severely compromise the rest of the start-up phase.
Starting a platform often happens when using a list of addresses for the first time and which may not be fully qualified. If you send to invalid addresses or to honeypot addresses this will contribute to diminishing the reputation of the platform. If you have a list of invalid addresses, it is in your best interests to import it into the quarantine table (Administration/Campaign Management/Non deliverables Management/Non deliverables and addresses) before sending for the first times. If, all the same, you wish to requalify the invalid addresses, it is by far preferable to do this once the reputation of the platform is established and bit by bit in order to "dilute" the use of bad addresses over time.
To summarize the principles to be followed when starting up:
Here is some additional information on editing email content.
Certain ISPs check the validity of the sender address (From) before accepting messages. A badly formed address may result in it being rejected by the receiving server. You must make sure a correct address is given at the instance level (menu Tools > Advanced > Deployment wizard...) or in the most frequently-used scenarios.
By default, when the message is analyzed, a typology rule checks whether an opt-out link has been included and generates a warning if it is missing. You can change this rule so that an error is raised rather than a simple warning and stop a delivery from going out without this link.
You must check that the opt-out link works correctly before each time you send. For example, when sending the proof, make sure the link is valid, that the form is on-line and that validating this changes the value of the No longer contact this recipient field to Yes. You should make this check systematically because human error is always possible when entering the link or when changing the form.
If a problem is detected concerning unsubscription after the delivery is started, it is still possible to perform an unsubscription manually (using the mass-update function, for example) for those recipients who click the opt-out link even if they were not able to confirm their choice.
As a general rule, do not try to get in the way of recipients who want to opt-out by requiring them to fill out fields such as their email address or name, for example. The form should have one validation button only, and reconciliation should be performed on the encrypted identifier only. Requesting additional confirmation is not reliable: a user may have two email addresses redirected to the same box (for example: firstname.lastname@example.org and email@example.com). If the recipient is able to remember the first address only and wishes to unsubscribe via a message sent to the other one, the form will refuse this because the encrypted identifier and the email address entered will not match.
Adobe Campaign can be configured to work with SpamAssassin. This makes it possible to score emails to determine whether a message runs the risk of being considered as spam by the anti-spam tools used upon receipt.
Before starting a delivery, the Preview tab enables you to evaluate the risks. A warning message gives the result of the test as in the following example:
The More... link enables you to view the analysis type, from the the Anti-spam checking.
If a high level of risk is detected, the following message is displayed:
The reasons for this risk are given in the Points > Rule > Description section of the Anti-spam checking tab.
SpamAssassin must be installed and configured on the Adobe Campaign application server.
For more information on configuring SpamAssassin in Adobe Campaign, consult the Installation guide.
Below you will find details on duplicate and quarantine management.
You have the possibility to monitor the sent email volume by IP address. A schema extension is needed for this. You need to extend the broadlogs table to add the "public identifier" and create a workflow to extract and display the data. Contact Adobe if you need this.
Having duplicate email addresses can have multiple consequences:
Besides this side-stepping of opt-in procedures, this situation will likely lead users to consider the messages as spam and to trigger a blacklisting procedure at the ISP.
You must be especially prudent when performing operations on the database:
There are, all the same, cases in which duplicates between the different partitions are normal. For example, when sending for third-parties or different company entities, it is logical for the same person to be a recipient for different reasons. It is, however, rarely normal to find duplicates within the same partition.
Adobe Campaign manages a list of quarantined addresses. The recipients whose addresses are quarantined are excluded by default during the delivery analysis: they are not targeted. An email address can be quarantined for example when the inbox is full or if the address does not exist. In all cases, quarantining corresponds to the specific rules detailed below.
Quarantine management is presented in the detailed documentation.
An email whitelist is a list of email addresses or domain names from which an email blocking program will allow messages to be received.
If you want to be able to authorize new IP addresses yourselves, a package is available upon request. Please contact your Adobe administrator. For more details on this tool, refer to this technote.
There are two types of whitelists:
To be accepted by these whitelists, the sender must pass a series of tests based on a technical verification (its email server must not be an open relay but should have a static IP) of the infrastructure or its activity (delivery frequency, volume, number of complaints).
If the sender does not follow one of these rules, it may be deleted from the whitelist. In its Adobe Campaign Email Deliverability package, Adobe Campaign offers an accompanying expert consulting service for the certification process for non-commercial whitelists.
Commercial white lists are based on a system that allows the sender to bypass antispam filters altogether or be assigned incremental points as they enter the system. These paying whitelists (CPT or on an annual basis) are offered by systems such as Return Path Sender Score.
ISPs are free to use these services and the number of ISPs can vary depending on the whitelist. A sender can therefore be more confident when sending his messages by having a delivery guarantee. Certain whitelists also offer to open images and activate links.
Appearing in a whitelist is an undeniable asset for any email campaign. In its Adobe Campaign Email Deliverability package, Adobe Campaign offers a commercial whitelist certification service such as CSA and Return Path Sender Score.
Here are a few definitions of the main technical terms.
Adobe Campaign checks whether a reverse DNS is given for an IP address and that this correctly points back to the IP.
MX rules are used to control the speed at which the Campaign MTA (Message Transfer Agent) sends emails to each individual email domain or ISP (e.g. hotmail.com, comcast.net). These rules are typically based on limits published by the ISPs (e.g. do not include more than 20 messages per each SMTP connection).
TLS (Transport Layer Security) is an encryption protocol that can be used to secure the connection between two email servers and protect the content of an email from being read by anyone other than the intended recipients.
SPF (Sender Policy Framework) is an email authentication standard that allows the owner of a domain to specify which email servers are allowed to send email on behalf of that domain. This standard uses the domain in the email's "Return-Path" header (also referred to as the "Envelope From" address).
DKIM (Domain Keys Identified Mail) authentication is a successor to SPF and uses public-key cryptography that allows the receiving email server to verify that a message was in fact sent by the person or entity it claims it was sent by, and whether or not the message content was altered in between the time it was originally sent (and DKIM "signed") and the time it was received. This standard typically uses the domain in the "From" or "Sender" header. To insure the security level of the DKIM, 1024b is the Best Practices recommended encryption size. Lower DKIM keys will not be considered as valid by the majority of access providers.
DMARC (Domain-based Message Authentication, Reporting and Conformance) is the most recent form of email authentication, and it relies on both SPF and DKIM authentication to determine whether an email passes or fails. DMARC is unique and powerful in two very important ways:
DMARC can leverage the reports generated by 250ok.
SMTP (Simple mail transfer protocol) is an internet standard for e-mail transmission.
Adobe provides a dedicated IP strategy for each customer with a ramp-up IP in order to build a reputation and optimize delivery performance.
Below you will find details on the Delivery throughput report as well as the different monitoring tools offered by Adobe Campaign via the deliverability package. Here are some additional guidelines on deliverability monitoring:
This report contains information on the delivery throughput of the entire platform for a given period. To measure the speed at which the messages are delivered, the criteria are the number of messages sent per hour and the size of the messages (in bits per second). In the example below, the first graph shows the successful deliveries in blue, and the number of erroneous deliveries in orange.
You can configure the values displayed by changing the timescale: 1-hour view, 3-hour view, 24-hour view, etc. Click Refresh to confirm your selection.
The technical monitoring report is available via the Supervision universe, in the Deliverability section.
It includes a number of deliverability quality indicators for your platform (the indicators are updated daily at 9 AM).
In addition, you are able to receive a daily report by email at a specified address. Please let us know the requested email address by email.
Some technical terms:
Reputation Authority: This WatchGuard's score is calculated in real time according to the feedback received from their machines worldwide, but also from the different users who use their software. Administrators can use such tools to apply a first level filter on their messaging servers. Most important, this indicator is relevant if customers are sending messages to recipients in the USA. Its impact within Europe is irrelevant and moreover the score is calculated based on the messages volume handled by that firewall: this means that as very low volumes pass through Europe, even one complaint will affect customers' score massively. So, if there are no messages sent to US recipients, just ignore this indicator, as it was specifically implemented for Adobe Campaign customers in the USA.
For your information, clicking on the IP link within the technical report will lead to http://www.reputationauthority.org, where one has the possibility to clean the IP history and get a neutral score again. Nevertheless, this action is limited to a number of times per month. Please also be aware there is no support provided by WatchGuard‘s Reputation Authority (sending delisting requests is then useless). Otherwise this scoring is based on the following:
Message content: for instance: presence of spam words
IP/Domains reputation: for instance: IPs are listed on an RBL
IP configuration: for instance: IPs associated to different domains
Volumes sent by IP: for instance: presence of peaks or significant variations
The Inbox monitoring report is available via the Supervision universe, in the Deliverability section.
It gives an overview of the quality of emails sent over a given period of time. A benchmark comparison is also made with the other platforms that have subscribed to the deliverability service.
This report is available for each delivery, in the Inbox Rendering tab.
The number of rendering reports is limited monthly according to your license agreement. The counter is reset every 25th day of the month.
Adobe Campaign has implemented a dedicated workflow (Update seed network for Inbox Rendering) to enable you to recover the target addresses to use for your proof. This workflow is located in the Administration > Production > Technical workflows node of the Adobe Campaign tree.
This workflow creates and updates the addresses used in the seed network. It runs daily at 5 AM. To force it to run straightway, right-click on the scheduler and select Execute pending task(s) now.
The contacts are inserted into a folder named Inbox Rendering recipients and assigned to a Inbox Rendering recipients group to enable rapid targeting.
Simply add this group in your delivery actions to enable the Adobe Campaign capture process.
The rendering thumbnails can be accessed in the Inbox rendering tab of the delivery a few minutes after sending the emails.
Each sending is counted off against your number of authorized rendering reports. Use the Refresh button to update the list.
Example of a rendering report:
If you have defined specific access rights for your operators, they require read rights for the "Inbox rendering recipients" and the associated group.
This mode implies using the nmsRecipient table. If not, you have to include the target addresses in your contact table.
If you use personalization elements in your emails, the "source" profiles must be specified as a consequence.
Signal Spam is a French service which offers anonymized feedback loop reporting for French ISPs (Orange, SFR).
This service allows you to follow the reputation of the French ISPs and track customers' activity evolution.
Signal Spam also provides direct complaints that end users log through a dedicated interface. Those complaints are then quarantined from the email address database.
250ok is a monitoring solution which provides IP, domain blacklisting and reputation indicators.
The information provided is real-time, which allows for a pro-active assistance. 250ok a complementary solution to the Adobe deliverability internal tools.
This message indicates that you have reached the quota limit for a specific MX and that you have to wait to be able to send another email to this provider.
If you always get the same error message for an ISP, your email or IP may have been detected as faulty by the ISP. Carry out the following recommendations:
If the problem persists, contact the commercial or deliverability services, or Adobe Campaign support.
The status Blacklisted is a result of a feedback loop (when a person reports a message as spam).
The status Quarantined is a result of a soft or hard bounce.
Here are 10 possible reasons: not defined, user unknown, invalid domain, blacklisted address, refused, error ignored, unreachable, account disabled, mailbox full, not connected.
To check whether your IP address is blacklisted, you can use various web sites to verify it:
To check whether your IP address is blacklisted, you can use various web sites to verify it:
Generally the result of the IP address check will return a list that contains details of the blacklist and also the name of the web site that blacklisted the IP address.
By clicking on the link you can access the web site details.
Then, you can request that your web site be delisted from the web site that blacklisted the IP address.
The delisting process may vary depending on the web site. Some sites require you to create an account, while others just need you to provide the IP address.