Adapting to ever-changing anti-spam policies
Laurent Garnier, deliverability director, NP6
Three ways to maximize inbox placement rates and optimize email campaigns
Sending the right message to the right person at the right time is fundamental to email marketing today. The current anti-spam screening methods used by the main U.S. Internet service providers (ISPs)—Outlook, Gmail, Yahoo, etc.—are not the only drivers of this. Indeed, it is the recipient's reactions to your message that will ultimately determine the placement in their inbox. Analyzing the current email deliverability landscape, we can clearly identify three optimization tools:
This is the foundation of good deliverability. Combined with the transparency of the first-point-of-contact-message—which defines the nature and the frequency of the messages to come, as well as requesting a confirmed, double opt-in—the data gathering will be optimal. In fact, a double opt-in registration confirms the accuracy of the address to ensure that it belongs to the person who registered. The confirmation message must be short and effective, with only one click. Consider an automatic recovery email may if there is no response.
Although 20 to 25% of the addresses are "lost" in this type of registration, it must be borne in mind that this is a first filtration step ensuring the high quality of the addresses collected. If using a simple opt-in, the same number of addresses gathered is likely to be lost over time and the time used to purge the database will be a real handicap for deliverability.
The email database
It is imperative that the permission granted at the point of collection is respected at all times throughout the life of the customers in the email database. It is highly recommended that the recipient is reminded of the context of their subscription and that a clearly visible, and simple to execute, unsubscribe link is highlighted. In fact, a simple and convenient unsubscribe process will greatly limit the spam alerts and therefore be beneficial for deliverability. High marketing pressure (i.e., too many emails) is often the cause of subscribers' disengagement. Consequently, it's necessary to adapt emailing frequency in light of responses. Email marketing must comply with recommendations defined by the law.
The infrastructure used to send emails must respect and properly implement authentication protocols in force (SPF, DKIM, dMarc, etc.). It must also be able to effectively manage and interpret feedback from emails that weren't opened/received and the complaints, whether issued directly by the user or from feedback loops made available by some operators. A good sending architecture should be able to communicate with the different email domains. Each domain has its own requirements that one must be able to integrate to the sending servers.
Following the daily developments related to the distribution of messages and applying them is a full-time job that should not be overlooked. It's essential to understand the reasons for the distribution problems, instead of simply removing recipients from sending lists. ISPs are evolving and equipping themselves with more effective filtering solutions that are clearly focused on recipient behavior.
Since email operators filter increasingly depending on behavior, email marketers must know how to elicit a positive reaction from customers by proposing a relevant and appropriate offer that answers their needs. Easier said than done, but segmentation tools based on conversion rates or predictive analytics can facilitate this process. Based on the generalization of fine segmentation, the study “Five in Five” from IBM, released in December 2011, predicted that spam will disappear within five years. But what if all this was simply a matter of evolution? I cannot refrain from quoting Charles Darwin: “The species that survive are not the strongest, nor the most intelligent ones, but those that adapt best to changes.”
Laurent Garnier is deliverability director for NP6.