How to Recover a Domain After Your Emails Start Going to Spam

You notice it on a Tuesday. Replies dropped from eleven a day to one. Opens look fine — they always do — but nobody is answering, and a seed test puts every message in the spam folder at Gmail and in Quarantine at Outlook. The domain you spent six weeks warming has turned into a liability.

Domain reputation damage is recoverable more often than people assume, but the recovery is slow, boring, and mostly about what you stop doing rather than what you start doing. Below is the sequence that actually works, with realistic timelines, plus the point at which retiring the domain is simply the cheaper decision.

Two things to accept before you start. Reputation is scored on a rolling window — roughly the last 30 days of behavior, weighted toward the most recent days — so time is a required ingredient and no trick shortens it. And reputation lives on your domain, mailboxes, and sending IP, not on your software. Switching tools without changing behavior changes nothing.

Step 1: Stop sending today, not at the end of the week

The most common mistake is letting the in-flight campaign finish. "We'll just let the last 400 go out." Those 400 emails land in spam folders, generate more negative signal, and push the domain further down. Every additional bad send resets your rolling window and adds a week to the recovery.

Pause every sequence on the affected domain. If you run several domains and only one looks damaged, pause that one and keep the others live — but verify first that they are genuinely clean, because reputation problems usually share a root cause across the whole setup. The same unverified list is often feeding all of them.

Do not pause warm-up. Warm-up traffic is the only thing still producing positive signal while cold sends are stopped, and it is the mechanism you will use to rebuild. If anything, raise it. This is the one scenario where running warm-up at full volume with zero campaign sends is exactly right.

Step 2: Find out which layer is actually broken

"Going to spam" is a symptom with at least four possible causes, and they need different fixes. Isolate the layer before you change anything, or you will change nine variables at once and never know which one mattered.

Pull the data before you theorise. Google Postmaster Tools gives you domain reputation, spam rate and authentication pass rates — if you have not verified your domain there, do it now, it takes five minutes and backfills some history. Microsoft SNDS only helps if you control the sending IP through an SMTP relay; it tells you nothing about Microsoft 365 shared outbound. Run a blocklist check against Spamhaus DBL for the domain and SBL/CSS for the IP, plus SURBL and Barracuda. Read your DMARC aggregate reports and look for sending sources you do not recognise, because a spoofing campaign riding your domain is a real and frequently missed cause of a sudden collapse.

Then look at your own numbers. Gmail's published complaint threshold is 0.3%; healthy cold outbound sits below 0.1%. Bounce rates above 5% damage you on their own regardless of complaints. If you want the exhaustive cause list, we covered it in Why Are My Cold Emails Going to Spam? — the goal here is to name one or two specific causes, not to fix everything simultaneously.

Step 3: Fix the root cause before you fix the reputation

Reputation repair without a root-cause fix just buys you a slower version of the same crash. Four causes account for most collapses.

Unverified or stale lists. A 7% bounce rate is enough to move a domain from High to Low inside a week. Re-verify everything still queued, move catch-all domains into a lower-priority segment you send to at reduced volume, and delete any contact older than 90 days that has not been re-checked. Lists decay at roughly 2% a month; a list you bought in March is meaningfully worse in September.

Volume that outran what the domain earned. Jumping from 20 a day to 150 a day per mailbox because "warm-up finished" is the classic. A mature mailbox on a mature domain tops out around 40–50 cold sends a day. More volume needs more mailboxes, not more per mailbox.

Authentication drift. Someone edited DNS, SPF picked up an eleventh lookup and started returning permerror, or DKIM was rotated at the provider and never republished. Re-verify SPF, DKIM and DMARC alignment from scratch instead of assuming last year's setup survived a year of IT changes. Authentication failures are cheap to fix and brutally expensive to ignore.

Copy that earns complaints. Not spam-word copy — complaint-earning copy. Irrelevant targeting, no visible unsubscribe, five follow-ups crammed into eight days. If your complaint rate is above 0.1%, this is your problem, and no amount of warm-up will paper over it.

Step 4: Handle blocklists, but don't panic about them

Most cold-email domain problems are not blocklist problems. Gmail and Microsoft largely use internal reputation systems that no one can delist you from, which is why the fix is time and behavior rather than a form.

If you are listed, Spamhaus DBL has a free removal request and will relist you immediately if the behavior continues — so fix first, request second. Barracuda, SORBS and UCEPROTECT are worth requesting removal from, but their impact on B2B inbox placement is usually modest. Never pay a removal fee; legitimate blocklists do not charge, and the ones that do are not worth the money. For the internal provider filters that cause most of your pain, there is no form at all. Budget three to six weeks of clean sending.

Step 5: Re-warm slowly, with traffic that produces real replies

Once the root cause is fixed, you rebuild positive signal. This is warm-up again, but from a worse starting point than a brand-new domain, because the provider already holds negative history it weighs against you.

A ramp that works on a damaged domain looks like this. Week one is warm-up traffic only — 10 to 20 messages per mailbox per day, with replies and folder moves, and zero cold sends. Week two keeps warm-up running and adds five cold sends per mailbox per day aimed at your warmest segment: prior repliers, opted-in contacts, existing customers, anything where a genuine human reply is likely. Week three moves to 10–15 cold sends per mailbox per day, but only if seed tests show primary-inbox placement; if they don't, hold at week two volume rather than pushing through. Weeks four to six increase by about five a day every three days, capped at 60–70% of your previous peak for the first full month back.

Two rules decide whether this works. Run a seed-mailbox inbox placement test before each volume increase, not after — placement data that arrives after you have already sent 2,000 emails is a post-mortem, not a control. And do not resume the campaign that caused the problem. Same copy, same list, same outcome in about ten days.

How long recovery actually takes

Assuming the root cause is genuinely fixed, here is what to expect. Mailbox-level damage where one account is throttled and the others are fine clears in 7 to 14 days. A domain sitting at Low in Postmaster Tools takes three to five weeks to climb back to Medium or High. A Bad rating takes six to ten weeks and is roughly a coin flip on whether it fully returns. A Spamhaus DBL listing combined with a Bad rating should be treated as unrecoverable for outbound purposes.

You will know it is working when three indicators move together: seed tests show the primary inbox again, Postmaster domain reputation climbs a tier, and reply rate returns to within 70% of its old level. Open rate moving on its own means nothing — security scanners inflate opens regardless of where the message landed, which is why reply rate is the metric that tells the truth here.

When to stop trying and retire the domain

Be honest about the arithmetic. A sending domain costs roughly $10–20 to register, and a fresh domain plus mailboxes needs about four weeks of warm-up before it carries meaningful volume. If you are staring down eight weeks of uncertain recovery, starting clean is usually faster and cheaper than nursing the old domain back.

Retire it if Postmaster reputation reads Bad and has not moved after three weeks of clean behavior, if you have been listed on Spamhaus DBL more than once, or if the domain was registered less than 90 days ago and therefore has no positive history to fall back on. Retire it permanently for outbound if it is your primary company domain — in that case move cold sending to a separate domain and never bring it back.

One thing not to do: register five replacement domains and blast from all of them on day one. Same behavior, same outcome, five times the cleanup.

How Sendvanta helps you avoid the second crash

Recovery is expensive enough that the real goal is never needing it twice. Sendvanta scores deliverability health in layers — mailbox, domain, DKIM and IP — and automatically slows or pauses sending when risk thresholds are crossed, rather than letting a campaign run to completion while the domain burns. Drafts get SpamAssassin-scored before they send, seed-mailbox placement tests show where mail is actually landing, and engagement is classified bot-aware, so a scanner-inflated open rate can't hide a collapsing reply rate for two weeks.

Warm-up runs inside your own Gmail, Microsoft 365 or SMTP mailbox using real labeled messages, so it keeps generating positive signal during a recovery pause. Connecting your own mailboxes is free on every plan, including the free tier: $0 forever, no credit card, 1,000 active leads, 3,000 emails a month, up to 10 of your own SMTP mailboxes. Paid plans start at $29/mo, and if you need a replacement sending domain you can register one for a one-time $20 and add managed mailboxes at $4/mo each. The health scoring and placement tooling are detailed on the features page.