Over the last few months, several ISPs have announced or started testing feedback loops in the new Abuse Reporting Format (ARF). For the uninitiated, a feedback loop is a system by which the ISP provides the sender a copy of a message that a subscriber has reported as spam – usually by hitting a “report spam” button. The ISPs that have announced or are testing feedback loops are as follows:
AOL: Currently have feedback loop. Testing ARF format
Earthlink: Currently beta testing feedback loop in ARF format
Excite: Provides periodic list of email addresses of complainters
Hotmail: Beta testing ARF feedback loop
Outblaze: Currently beta testing feedback loop in ARF format
Yahoo: Announced intentions to provide feedback loop this year in ARF
The ARF format is a kind of machine parseable email message that has standard data elements in standard fields. The standardization is great because you can use the same code to pull the information out of these complaint messages.
Why are we so excited about this? Simply put, this information is deliverability gold. The #1 reason that commercial emailers run into delivery problems is that their complaint rates are too high. This data, when analyzed appropriately, can allow you to figure out where there are pockets of complaints inside of your email program. For example, working with one client we recently found:
* The best co-registration data partner had complaint rates that were 20% of the worst co-registration data partner. The implication here was to drop the worst data partner(s).
* Subscribers that were in their first two weeks on the list had 10x higher complaint rates than clients that had been on the list longer. However, this held only for third-party marketing messages and not for the company newsletter. Mailing new clients from separate IP addresses made sense here. In addition a more detailed review of the client’s lifecycle mailing approach was in order
* Particular subject lines / creatives drove much higher complaint rates (in some cases up to 20x the average). This indicated a need for more pre-mail testing.
This is the sort of information that can be gleaned from feedback loops. Best of all, it’s free.