Recovering a Damaged Sender Reputation: A Realistic 30-Day Plan

You can recover a damaged sender reputation, but not on a fixed timetable. Spend the first week finding and removing the cause, then send only to recipients who clearly want your mail, at a steady low volume, widening slowly while watching Postmaster Tools. Thirty days is a sensible window to see whether the trend turns; it is not a guarantee.

That caveat matters, because most recovery advice online promises a schedule that no mailbox provider has ever published. Google's own sender guidelines say only that "it can take time for improvements in spam rate to reflect positively on spam classification" (Google). What follows is a plan built around the few concrete things providers do say, plus the order of operations that actually changes the numbers.

First, work out what is damaged

Reputation is tracked separately for your sending domain and your sending IP, and separately by each mailbox provider. Before planning anything, find out which one is hurt and where.

Google's Postmaster Tools has separate IP Reputation and Domain Reputation dashboards. Yahoo states that "each IP and DKIM domain has a reputation" (Yahoo Sender Hub). Microsoft's SNDS reports data about individual IPs, and you can only see the IPs you are responsible for (SNDS).

What you see	Likely damaged	Where to confirm
Gmail sends to spam, Outlook and Yahoo fine	Gmail domain or IP reputation	Postmaster Tools: Spam Rate, Domain Reputation
Every provider filters you, on every ESP you try	Domain reputation	Postmaster Tools Domain Reputation; your bounce logs
Problem started when you moved ESP or IP	IP reputation	Postmaster Tools IP Reputation; SNDS if the IP is yours
Hard rejections quoting a blocklist	A listing on a public blocklist	A blocklist lookup on your sending IP and domain
Rejections mentioning authentication	Not reputation at all: a DNS fault	Deliverability audit

The last row is worth checking first, because it is the cheapest to fix and is frequently mistaken for a reputation problem. A broken SPF include or an expired DKIM key looks identical to "Gmail hates us" from the outside. If the symptom is Gmail-only, work through why emails go to spam in Gmail first.

Google defines its reputation levels in words, not numbers. "Low" is a "history of sending a significant volume of spam regularly", where mail "is likely to be marked as spam". "Bad" is a "history of sending a high volume of spam regularly", where mail "is almost always marked as spam or rejected" (Google). The deeper you are, the longer the climb.

The 30-day plan at a glance

The plan has four phases: stop and diagnose, fix the cause, restart small with engaged recipients, then widen in steps. Each phase has a gate you should pass before moving on, and the gates matter more than the dates.

Week	Goal	What you send	Gate to move on
1 (days 1-7)	Stop the damage, find the cause	Transactional and one-to-one mail only. Pause bulk and cold sends	Cause identified and fixed; authentication passes
2 (days 8-14)	Prove good behaviour	Small daily sends to your most engaged recipients only	Spam rate under 0.1% on the days data shows; bounces near zero
3 (days 15-21)	Widen carefully	Add the next engagement tier, in steady daily amounts	Spam rate holding under 0.1%; no new blocklist listings
4 (days 22-30)	Return towards normal volume	Normal programme, with the cleaned list and new suppression rules	Reputation trending up for 2 weeks; you know why it fell

If a gate fails, stay in that week. Moving forward because the calendar says so is how a two-week dip becomes a three-month one.

Week 1: stop the damage and find the cause

For the first week, pause every bulk and cold campaign and spend the time on diagnosis. Each send made with the original problem still in place adds more of the complaints or bounces that caused it.

Keep sending genuinely wanted mail: receipts, password resets, replies to people who wrote to you. That traffic is the best evidence you have that your domain sends mail people want.

Then go through the usual causes in order of how often they turn out to be responsible.

1. A dirty list

A hard bounce tells the receiving server you do not know who you are mailing. A run of them, especially from an old or bought list, is a common trigger for a sudden drop. Pull the bounce logs for the campaigns just before the drop and look at the share of 550 5.1.1 style "user unknown" responses.

Fix: remove every hard bounce, then verify the rest of the list before it is mailed again. Anything a verifier cannot confirm should be held back from the restart weeks, not mailed at full volume. Old and bought lists are also where spam traps hide, so they deserve the hardest look.

2. Complaints

Google asks senders to keep the user-reported spam rate in Postmaster Tools below 0.3%, and recommends below 0.1% (Google). Yahoo also asks for below 0.3% (Yahoo). A spike usually traces back to one campaign: a segment that never opted in, a forgotten list, a sudden change in frequency, or an unsubscribe link that was hard to find.

Fix: identify the campaign and the segment, and remove that segment from the restart entirely. Make sure one-click unsubscribe works and is processed quickly.

3. Missing or broken authentication

Run SPF, DKIM and DMARC checks on the exact domain in your From address and on the domain your ESP signs with. If DKIM is signed with the ESP's domain rather than yours, your own domain is not building reputation, and you may be inheriting someone else's.

4. A public blocklist listing

If rejections name a list, follow that list's own removal process once the cause is fixed. Requesting delisting before fixing the cause tends to end in relisting. The steps for the major lists are in how to get removed from an email blacklist.

By day 7 you should be able to write one sentence that explains why reputation fell. If you cannot, do not move on. A restart without a known cause is a guess.

Week 2: restart with your most engaged recipients

In week 2, resume sending only to the people most likely to open, reply or click, at a small and steady daily volume. The aim is to give mailbox providers a run of days where almost nobody complains and nothing bounces.

Define "engaged" with whatever signal you trust most. Recent replies, clicks and purchases are better than opens, because opens are inflated by privacy features on some mail clients. For a cold outbound programme, the engaged segment may be very small: people who replied in the last few months.

Two rules for this week:

• Steady beats large. Send roughly the same amount each day rather than one batch. Google advises senders to "avoid introducing sudden volume spikes" (Google).
• Nothing unverified. Every address in the restart should be a known-good one. A single hard bounce in this phase costs you more than it would in a normal week, because the denominator is small.

This is not the same as warming up a new domain, where you are building a reputation from nothing. You are rebuilding one that has negative history, which is why the audience matters more than the daily number. If you are starting a fresh domain or inbox instead, follow the warm-up schedule.

One practical wrinkle: Google notes that Postmaster Tools "dashboards might not include all data on days when your outgoing email volume is low" (Google). Small restart volumes can leave gaps in the charts. Treat missing days as no data, not as good data.

Week 3: widen carefully

In week 3, add the next tier of recipients in steady daily amounts, and keep watching spam rate and bounces each day. The restart only counts as working if those numbers hold as the audience grows.

A sensible way to tier a list:

Tier	Who	When to add
A	Replied, clicked or bought recently	Week 2
B	Engaged in the last few months	Week 3
C	Subscribed but quiet for longer	Week 4, if B held
D	No engagement for a long time	Re-permission campaign, or remove

Tier D is where many recoveries fail. Mailing dormant contacts is how you find recycled spam traps and the people who no longer remember signing up. Google recommends that senders "periodically send messages to confirm that recipients want to stay subscribed" (Google). A one-off "do you still want these?" message to tier D, with removal for non-responders, is usually worth more than the volume it gives up.

Week 4: return towards normal and lock in the fix

By week 4 you can move back towards normal volume, provided spam rate has stayed low and reputation has trended upwards for two weeks. This is also the week to make sure whatever broke cannot happen silently again.

For Gmail specifically, there is one concrete milestone. Google states that bulk senders with a user-reported spam rate above 0.3% are ineligible for mitigation, and that they "will be eligible for mitigation when their spam rates remain below 0.3% for 7 consecutive days" (Google). Google also says that after you take corrective action, you should "wait up to 7 days for the spam rate to reach a level that falls within compliance" (Google). If you need to contact Google about delivery, get through that 7-day window first.

Permanent changes worth making now:

1. Verify before every import. Lists from events, partners and lead sources go through verification before they ever reach a campaign.
2. Automatic suppression. Hard bounces suppressed after one occurrence; repeated soft bounces suppressed after a short run.
3. A sunset rule. Recipients who have not engaged for a set period stop receiving campaigns.
4. One alert. A daily check on Postmaster Tools spam rate, so a rise is noticed in days rather than at the next campaign review. Our guide to Postmaster Tools explains each dashboard.

What not to do during recovery

The common mistakes are attempts to escape the reputation rather than repair it. Most of them fail because the cause travels with you.

• Do not jump to a new domain or IP as a first move. A new one has no history, which providers also treat cautiously, and if the cause was the list, the new domain inherits the same list.
• Do not stop sending entirely for weeks. Silence gives providers nothing new to measure. Keep low, wanted traffic flowing.
• Do not send the old full list "one last time". That is the send that turns Low into Bad.
• Do not judge success by open rates. Opens are unreliable. Use spam rate, bounce rate and Postmaster reputation.
• Do not buy a dedicated IP to fix a domain problem. If the damage is to the domain, a new IP does not reset it. The trade-offs are in shared or dedicated IP.

When 30 days is not enough

If reputation is still Low or Bad after a month of clean sending, the cause is probably still present or the damage was deeper than it looked. Go back to week 1 rather than pushing volume.

Signs you are not done: spam rate keeps bouncing above 0.1% even to engaged segments, new blocklist listings appear, or a particular provider keeps rejecting while others recover. In the last case, focus on that provider's own tools and published requirements rather than the general plan.

The short version

• Find out what is damaged (domain or IP, and at which provider) before changing anything.
• Week 1: pause bulk mail, find and fix the cause. Do not move on without a one-sentence explanation.
• Week 2: send only to your most engaged recipients, in steady daily amounts, nothing unverified.
• Week 3: add the next engagement tier only if spam rate holds under 0.1%.
• Week 4: return towards normal, and add verification, suppression and a sunset rule so it does not recur.
• For Gmail, the concrete milestone is 7 consecutive days below 0.3% spam rate. Beyond that, no provider publishes a recovery clock, so judge by the trend, not the calendar.

Before the restart in week 2, a quick deliverability audit of your sending domain rules out the DNS faults that no amount of careful sending will fix.