Incident: Reputation Tanking Overnight

Your sender reputation is how much mailbox providers like Gmail and Outlook trust your email. When it drops, more of your email lands in spam or gets refused. This playbook shows you how to spot a drop, find its cause, and win the trust back over about six weeks.

Signs your reputation has dropped

A drop usually shows up in several places at once.

Strong signs:

• In Google Postmaster Tools (Google’s free dashboard for senders), your spam rate rises or delivery errors appear.
• In Microsoft SNDS (its free dashboard for IP addresses), your IPs turn from green to yellow or red.
• Placement tests (sending to a set of test mailboxes to see where email lands) show more spam-folder placement.
• Your logs show more 4xx (try later) and 5xx (refused) replies, especially 5.7.1 policy refusals.

Weaker signs:

• Opens and clicks fall, though your content and list haven’t changed.
• Email that used to reach Gmail’s Promotions tab now lands in spam.
• Bounces go up.

Don’t trust falling open rates on their own. Apple Mail Privacy Protection and similar features open emails automatically, which makes open counts unreliable.

Google has announced it will retire the domain and IP reputation dashboards (High, Medium, Low, Bad) in Postmaster Tools. Don’t rely on them. Watch the spam rate, delivery errors and compliance status instead.

Find the cause

A reputation drop is a symptom. Find the cause before you try to fix it.

Common causes

Your list:

• Did you mail a new segment: bought, old or never checked?
• Did a recent campaign have more than 2% bounces?
• Has bounce handling stopped, so you keep mailing dead addresses?
• Did you import addresses without checking them?

Complaints (people clicking “report spam”):

• Did a recent campaign get complaints from more than 0.1% of recipients?
• Did you change your From name, subject style or type of content?
• Are you sending more often?
• Is unsubscribing broken or hard to find?

Your setup:

• Did you change email service or sending servers?
• Did your SPF, DKIM or DMARC records change? (These DNS records prove your email comes from your domain.)
• Did you move to a new IP without warming it up (raising volume slowly)?
• On a shared IP, is another sender causing trouble?

Engagement (how readers interact with your email):

• Have opens and clicks fallen over the past 30 days?
• Did one campaign do much worse than usual?
• Do you still mail people who haven’t engaged for six months or more?

Outside events:

• Are you on a blocklist (a public list of spam sources) such as Spamhaus, Barracuda or SpamCop?
• Did a provider change its filters? This is rare, so rule out your own causes first.

How to investigate

1. Match the timeline. Find the day the drop started. Look at what you sent that day and in the 48 hours before.
2. Check each provider. One provider or all? At Gmail, a drop shows in the spam rate. At Microsoft, it shows in SNDS.
3. Check each campaign. Compare bounces, complaints and engagement to find the one that stands out.
4. Check each segment. If you mailed a win-back group, an old segment or a new signup source, look at its numbers on their own.

The recovery plan

Start small, with the readers most likely to welcome your email. Then widen the audience one step at a time.

Phase 1: Stop the damage (days 1–3)

1. Cut volume by 50–75%. Send only to people who opened or clicked in the last 30 days.
2. Pause the segment that caused the problem, if you found one.
3. Fix authentication if SPF, DKIM or DMARC is broken. Check with the DNS lookup.
4. Request blocklist removal, after fixing the cause.
5. Check bounce handling. Suppress (block from future sends) every hard bounce from the bad period.
6. Check complaint handling. Feedback loops tell you when someone marks your email as spam. Make sure you receive them and suppress those addresses.

Phase 2: Rebuild engagement (days 4–14)

• Send only to people who engaged in the last 30 days.
• Send your strongest content: the kind that got the best response before.
• Keep your normal schedule. Don’t send extra to make up for the pause.
• Check opens, clicks and complaints daily.

Good engagement from a small list shows providers that people want your email. Nothing rebuilds trust faster.

Phase 3: Widen slowly (days 15–45)

Start when your numbers improve: the Gmail spam rate falls and SNDS shows green. Then widen one step a week:

• Week 3: people who engaged in the last 60 days. This often roughly doubles your volume.
• Week 4: the last 90 days.
• Week 5: the last 120 days.
• Weeks 6–7: the last 180 days.

Before each step, check that complaints stay under 0.05%, bounces under about 1%, your Gmail spam rate and SNDS status are stable or better, and you’re on no new blocklist. If any number gets worse, stay at the current step until it settles.

Phase 4: Keep it steady (day 45 onwards)

• Keep choosing who to email based on engagement.
• Don’t mail people inactive for six months or more without a win-back campaign first.
• Set alerts for your key numbers.
• Check addresses monthly and review inactive readers every quarter.

Notes for each provider

Gmail

• Watch the spam rate in Postmaster Tools. Gmail and Yahoo require bulk senders (5,000+ emails a day to their users) to stay under 0.3%. Google recommends under 0.1%.
• Check the compliance status. It shows whether you meet Gmail’s sender rules, such as SPF, DKIM, DMARC and one-click unsubscribe.
• Gmail weighs engagement heavily. Send to your most engaged Gmail readers first.
• DMARC alignment matters: the From domain must match the domain that passes SPF or DKIM.
• Recovery usually takes weeks, and longer after a severe drop.

Microsoft

• SNDS shows each IP as green, yellow or red, and counts spam trap hits (emails to addresses that exist to catch bad lists). Yellow is a warning. Red means refusals and junk-folder placement.
• JMRP (Junk Mail Reporting Program) is Microsoft’s feedback loop. Sign up to learn who marks your email as junk.
• Any spam trap hit means part of your list is bad.
• Microsoft leans heavily on IP reputation. On a shared IP, talk to your email service provider.

Yahoo

• Yahoo reacts strongly to complaints and blocks quickly once they pass its limit.
• To recover, keep complaints very low for a long time.
• Yahoo shares less data than Google or Microsoft. Sign up for its Complaint Feedback Loop to learn who marks your email as spam.

Numbers to watch

Check these daily while you recover. They are rough targets, not provider rules.

Number	Phase 1	Phase 2	Phase 3	Phase 4
Daily volume	25–50% of normal	30–40% of normal	Rising step by step	Normal
Complaint rate	Under 0.02%	Under 0.03%	Under 0.05%	Under 0.05%
Bounce rate	Under 0.5%	Under 0.5%	Under 1%	Under 1%
Gmail spam rate	Falling	Falling	Low and stable	Low and stable
SNDS status	Not red	Green	Green	Green
Blocklists	None	None	None	None

Common mistakes

• Catching up. Mailing your whole list after a pause, to make up for lost time, makes the drop worse.
• Fixing only the symptom. Get off a blocklist without fixing the cause, and you’ll be listed again within days.
• Widening too fast. Jumping from engaged readers to your whole list gives providers no time to adjust.
• Not watching. Check daily so you catch a setback early.
• Blaming the provider. Filters react to how you send. Change how you send, and they will follow, slowly.
• Changing everything at once. Then you won’t know what helped. Change one thing at a time and measure.

Reputation drops in hours or days and takes weeks or months to rebuild. There are no shortcuts.

Checklist

• Confirm the drop with Postmaster Tools, SNDS or a placement test, not open rates alone.
• Find the day it started and what you sent around then.
• Cut volume and send only to readers who engaged in the last 30 days.
• Fix authentication, bounce handling and complaint handling.
• Fix the cause before you ask to be removed from a blocklist.
• Widen one step a week, only while the numbers stay healthy.
• Keep sending based on engagement once you’ve recovered.