Skip to content
Blog  /  Deliverability

Why bounces compound — and how to stop the spiral

A close look at how a single bad campaign can cost you six months of sender reputation.

If you send to a list with a 6% bounce rate, you’re not losing 6% of your revenue. You’re losing closer to 40%. Here’s the math.

TL;DR

Bounce rate is a leading indicator, not a lagging one. Mailbox providers use the first 3–5% of your bounces as a signal to throttle everything else — even the addresses that would have delivered fine. The damage compounds silently across sends.

2.1%
Bounce rate where Gmail starts throttling sender reputation
38%
Drop in observed open rate at a sustained 5% bounce vs 0.5%
14×
Revenue loss-to-verification-cost ratio at the cliff

The dataset

We sampled 400M sends from 1,100 Cleanify customers between October 2024 and February 2026. Only senders with consistent monthly volume and at least one verification pass in the period were included. We bucketed sends by pre-send bounce rate (from the verification result) and then joined to post-send performance reported via Mailgun, SendGrid, and Postmark webhooks.

Where a sender used our white-label platform, we anonymised at the reseller level first.

Open rate vs. bounce rate

The first chart is the headline. Open rate falls faster than bounce rate rises — the curve is concave almost everywhere it matters.

Open rate by pre-send bounce rate
400M sends · all senders with ≥10k monthly volume · Oct 2024 – Feb 2026
36%28%20%12%4%0%1%2%3%5%8%+Pre-send bounce rateThrottling begins (2.1%)

Between 0% and 2%, open rate is effectively flat. Beyond ~2.1% — the point most Gmail-heavy senders hit throttling — the drop-off accelerates. A sender at 5% bounce rate loses about a third of opens they would have gotten at 0.5%, and they still paid to send to every one of those addresses.

Inbox placement

Open rate is downstream of placement. Seed-inbox testing across Gmail, Outlook, and Yahoo accounts tells a similar story with less noise.

Inbox placement by bounce rate tier
Seed-inbox sample · Gmail, Outlook, Yahoo · Jan–Feb 2026
<0.5%0.5–2%2–4%4–6%6%+97%94%81%62%41%97%94%81%62%41%

The inflection is sharp between the 2–4% and 4–6% tiers. That’s where reputation scoring stops being forgiving.

The revenue curve

Revenue follows opens, but with extra leverage. If your average open-to-revenue conversion is stable, the revenue curve compounds the open-rate drop with the send-volume cost of each unwanted bounce.

“We were blowing past our verification budget because we thought it was expensive. It turned out to cost about $180 a month. We were losing $9,000 a month in bounced revenue.” — CRM lead at a Series B commerce brand (200k active subscribers)

Put simply: verification is usually 1–3% of the revenue recovered by verifying. The trade isn’t close.

Where the cliff is

For any sender asking “how clean is clean enough,” the answer from this data is:

  • Under 1% — safe for any provider, any list
  • 1–2% — Gmail will forgive you; Outlook will start flagging
  • Above 2% — reputation erodes on every send until you clean
  • Above 5% — you’re being throttled across providers right now

What to actually do

Three things, in order:

  1. Verify before import. Run every new list — purchased, scraped, event-sourced — through verification before it touches your ESP.
  2. Re-verify every 90 days. Addresses decay at 2–3% per month. Yesterday’s clean list is tomorrow’s medium-dirty list.
  3. Segment “risky” separately. Don’t send your best content to role-based or accept-all addresses. Isolate, test, and expect worse engagement.

None of this is new. What’s new is being able to see exactly what the sloppy version costs you. If you want to run the math on your own list, we built a free bounce calculator that takes 30 seconds.