Feedback loop (FBL)

How spam complaints find their way back to the sender, and why attribution is the hard part.

A feedback loop (FBL) is a channel through which a mailbox provider sends spam complaints back to the sender who caused them. Most providers send one report per complaint in the Abuse Reporting Format (ARF); Gmail instead publishes aggregate spam rates tied to a header the sender sets.

Why it matters

Without feedback loops a sender is blind to complaints: reputation degrades, mail starts landing in spam, and the sender never learns which campaign or which list segment caused it. With them, every complaint becomes an immediate suppression and a data point. Enrollment is free at Yahoo and Microsoft. The catch is attribution: ARF reports frequently redact the recipient's address for privacy, so the report alone may not tell you who complained, only that someone did.

In practice

Yahoo and Microsoft operate per-message loops: each complaint produces an ARF report (RFC 5965) containing the original message. Gmail offers no per-message loop at all. Instead, you stamp each send with a Feedback-ID header, and Postmaster Tools reports aggregate spam percentages per identifier, useful for spotting a bad campaign, useless for suppressing an individual. In both worlds, attribution depends entirely on identifiers you embedded in the message before it left, because the redacted report is only as good as what survives inside it.

How Email Fast handles it

Email Fast stamps every message with a Feedback-ID and a VERP-token Message-ID. When an ARF report comes back, even with the recipient redacted, those identifiers attribute it to the exact message and recipient, and the complainer is suppressed instantly, as described under spam complaint. FBL enrollment is tied to live delivery and arms with sending at launch.