The List Hygiene Protocol That Protects Reputation

Every mailing list decays. Addresses are abandoned, inboxes fill, interest fades and consent records become stale. Left unmanaged, that decay shows up as rising complaint rates and falling deliverability, and it accelerates: poorer placement reduces engagement, which lowers the signals mailbox providers use to judge you, which lowers placement further.

Hygiene is the work of interrupting that loop.

Classifying failure correctly

Not all failures mean the same thing, and treating them identically is the most common hygiene error.

A hard bounce is a permanent failure — the address does not exist or the domain rejects all mail. These must be suppressed immediately and permanently. Continuing to send is the strongest possible signal of a poorly maintained list.

A soft bounce is temporary: a full mailbox, a server temporarily unavailable, a message too large. These should be retried a limited number of times with increasing delay. An address that soft-bounces repeatedly over weeks has effectively become a hard bounce and should be suppressed.

A block is a policy decision by the receiving server, often affecting many messages at once rather than one address. Blocks are not list problems; they are reputation or configuration problems, and suppressing the addresses will not help.

A complaint means the recipient marked you as spam. This is more serious than an unsubscribe and should trigger immediate suppression, plus investigation into which campaign or source produced it.

Engagement decay

Addresses do not fail; they just stop being read. Somewhere between thirty and fifty percent of a typical list has not engaged in the last ninety days. Sending to all of them dilutes your engagement signals, because mailbox providers weight aggregate recipient behaviour when deciding where future mail lands.

The practical approach is tiering by recency. Recipients who engaged in the last thirty days get full frequency. Those in the thirty-to-ninety window get reduced frequency and your strongest content.

Beyond ninety days, run a re-engagement sequence — typically two or three messages — and suppress anyone who does not respond.

The re-engagement step matters because it distinguishes disinterest from abandonment. Someone who has not opened in six months may still be a customer who wants your mail; someone who ignores an explicit "do you still want to hear from us?" has told you the answer.

Consent, recorded properly

Hygiene is not only about deliverability. Under GDPR you must be able to demonstrate consent, which means storing the timestamp, source and form for every opt-in — not merely a boolean subscribed flag. When a data subject request arrives, retrieving a defensible record under time pressure is where teams discover their records are inadequate.

Consent state must also propagate. An unsubscribe in one channel must suppress in all of them, including any automation that may be running. This is where separately maintained suppression lists fail: the marketing tool knows about the unsubscribe, the transactional system does not, and the contact receives mail they explicitly declined.

Imports are where lists get poisoned

Most hygiene damage enters through imports, not through sending. A purchased or scraped list produces complaints at rates that will damage a domain quickly, and it is prohibited by every major provider. Even legitimate imports need care: validate addresses at the point of entry, preserve subscription status from the source system, and never default an unknown status to subscribed.

The cheap intervention is validation before import rather than cleaning afterwards. Cleaning a list already in the database means you have already stored bad addresses and must now identify them, whereas validating on entry means they never arrive.

The routine

Hygiene is a schedule, not a project. Weekly: review complaint and bounce rates per campaign. Monthly: recalculate engagement tiers and suppress repeated soft bouncers.

Quarterly: run re-engagement for the lapsed tier and suppress non-responders. Continuously: validate all imports at entry. Each step is small; skipping all of them for a year produces a deliverability problem that takes just as long to reverse.

Sunset policy, and doing it without losing revenue

Suppressing unengaged contacts sounds simple until the contacts are customers. A pure engagement-based sunset rule will eventually suppress people who read your mail without clicking and buy without reading it. Purely behavioural signals miss the passive reader.

A workable sunset policy combines two things: an explicit re-engagement ask, and a commercial safeguard. The re-engagement message makes the choice visible — stay or go. The safeguard is excluding recent purchasers or active account holders from suppression regardless of engagement, because their relationship is demonstrated by something more reliable than email behaviour.

When contacts do get suppressed, do not delete them. Suppressed contacts retain their history and can be reactivated if they engage later, whereas deleted records lose the context that makes reactivation possible. The goal is to stop sending, not to destroy data.

Measuring whether hygiene is working

Hygiene produces measurable movement, and watching it tells you whether the effort is paying off. Delivery rate should rise as invalid addresses are removed. Complaint rate should fall as suppressed contacts stop receiving mail they do not want.

Click rate often rises without any content change, simply because the denominator is now people who read.

The metric to watch most closely is complaint rate per thousand sends, because it is the earliest and least ambiguous signal. An open rate can be distorted by client behaviour and a click rate by a single well-placed link, but a complaint is a deliberate action by someone who received your mail.

How Cresca supports this

Suppression is maintained at the account level, so an unsubscribe applies across campaigns, automations and future imports rather than only the context in which it happened. Bounce and complaint events arrive through the same event stream that drives analytics and automation triggers, so a hard bounce is reflected everywhere at once.

Pricing

Plans are defined by contact count, which makes hygiene decisions concrete: suppressing an unengaged address frees capacity you can spend on a genuinely engaged one.

The short version

Suppress hard bounces and complaints immediately and permanently. Retry soft bounces a bounded number of times, then suppress. Tier by engagement recency and explicitly re-engage the lapsed before removing them.

Record consent with source and timestamp, and propagate suppression across every channel and automation. Validate imports at entry, because that is where most damage originates.