Why Your Bounce Rate Spiked After a Sending Pause

Your bounce rate was fine — 1%, 2%, maybe 3% — and then you took a break from sending. Weeks, maybe months. You came back, hit send, and the bounce rate jumped to 8%, 10%, 12%. Amazon SES started warning you, or paused your account outright.

Nothing about your content changed. Nothing about your list changed. So what happened?

The break is what happened. And it's one of the least-understood bounce-rate mechanics in email marketing.

The pause-bounce mechanism

An email list is a perishable asset. Addresses go bad every month — people change jobs, abandon inboxes, corporate domains get restructured. The industry rule of thumb is 20–25% list decay per year, which means even a three-month pause kills roughly 5–6% of your list. Not soft bounces — hard bounces. Gone.

When you send regularly, dead addresses bounce a little at a time, and you suppress them a little at a time. Your bounce rate stays low because the decay is spread across dozens of sends.

When you pause, all that decay compounds into a single send. Three months of dead addresses hit at once. Your effective bounce rate for that one campaign can easily be 3–4x your usual. A list that ran 2.5% weekly can spike past 10% on the first send back — and 10% is where SES's automated enforcement kicks in.

The documented case: ~5,000 subscribers, a newsletter running on SES for years, a sending pause of a few weeks. The first send back spiked the bounce rate past the tripwire, the account was suspended, and the appeal was rejected. The list didn't decay — it decayed all at once, on one campaign, in front of an automated judge.

"Just reduce your bounce rate" misses the point

Most bounce-rate advice assumes the problem is who is on your list. Bad signup forms, purchased lists, no double opt-in. That's the usual checklist — and it's fine advice for a list with chronic bounce problems.

But a pause-induced spike isn't a list-quality problem. It's a timing problem. Your signup sources were fine. Your content was fine. The addresses simply expired while you were away. The fixes are different:

1. Don't resume with a full-list blast. Ever. A warm-up send to your most-engaged segment (last 90 days of opens/clicks) gives you a clean reputation signal first.

2. Verify before the first send back. Run the list through a hygiene check after a pause — not at signup. The addresses that were valid in January can be dead by June. This is the tool that would have saved that list: check, suppress, then send.

3. Suppress aggressively. Every hard bounce from the resume send needs to be suppressed immediately — not after "a few strikes." SES evaluates your bounce rate over a representative volume of recent sends — a rolling window, not a strict per-campaign rule. But it doesn't forgive a second spike while that rolling rate is already over.

4. Watch the engaged-segment split. If your engaged segment bounces clean and the full list doesn't, the dead weight is in the cold addresses. Cut them. A smaller list that delivers beats a bigger list that gets you suspended.

5. Know SES's actual thresholds. SES starts warning around 5% bounce rate and suspends around 10% (AWS's published numbers — cite docs.aws.amazon.com/ses at publish). Those are evaluated over a representative volume of recent sends, a rolling window — and one bad send is all it takes to drag that window over the line.

The mistake I made (so you don't)

The resume send was treated like a normal send. Same full list, same workflow. No list check, no warmup, no segmentation. One campaign, ~5,000 addresses, and the stale ones all bounced at once.

The appeal process doesn't care that you were on a break. SES enforcement is automated and the appeal review is opaque — "rejected" came with no actionable feedback. That's when the real lesson lands: the rail you're on matters as much as the list you're sending. When the enforcement is a black box, one bad campaign can end years of work.

That's why I now run newsletters on infrastructure I control, with failover to a second sending rail — if one rail trips, the other keeps delivering. But that's the advanced class. For today, the move is simpler: verify after every pause.

Before your next send after a break

Quick checklist:

• [ ] Run list hygiene on the full list (hard-bounce prediction, not just syntax checks)
• [ ] Suppress everything the hygiene check flags as dead
• [ ] Segment to engaged addresses (opened/clicked in the last 90 days)
• [ ] Send the engaged segment first, watch the bounce rate
• [ ] Expand to the rest only if the engaged send stays under 3%

Skip any step and you're gambling your account on one campaign. I lost that gamble.

Free download: the ListProof self-audit checklist. A one-page, step-by-step version of the resume-send hygiene process above — what to check, what to suppress, and the exact send order after a pause. Free, no signup theater. [CTA placeholder: direct PDF download, no email capture — signup-vs-download decision: no-signup, matches his copy]

Want someone else to do it? We run a $899 done-for-you List Rescue audit: we analyze your list, hand you a hard-bounce prediction figure in writing, and if the safe segment bounces above the prediction on your first send, we re-run the analysis free. [Details on the ListProof audit page.]