Why Google Workspace Cold Email Deliverability Drops After Scaling Volume
Campaigns were working. You doubled volume or added more inboxes. Everything fell off a cliff. Here's the mechanism and the fix.
Your cold email campaigns were working. Replies were coming in. Open rates looked healthy. Then you increased volume — maybe doubled daily sends or added more inboxes — and everything fell off a cliff. The campaigns are identical to what worked before. The only thing that changed was volume.
Why This Happens
Gmail tracks sending volume per domain and per IP. When volume increases suddenly, Gmail's systems treat it as a risk signal by design. Spammers and compromised accounts both exhibit sudden volume spikes, so Gmail's default response is to throttle or redirect to spam until the sender proves the new volume is legitimate.
Google's guidelines are explicit: they recommend increasing volume slowly and state that suddenly doubling previously sent volumes can result in rate limiting or reputation drops.
Here is what happens mechanically when you scale too fast:
- Your domain hits Gmail's internal volume threshold and messages start getting deferred with 4.7.28 SMTP errors, or they get silently routed to spam without any bounce signal.
- Your engagement rate drops because more emails are going to spam, which means fewer opens and replies per message sent. Gmail interprets this declining engagement as evidence that recipients don't want your mail. It's a feedback loop.
- If you added new sending IPs by bringing on a new outreach tool or new inboxes on different infrastructure, those new IPs have no reputation. Gmail treats them as unknown.
Step-by-Step Diagnosis
Check Google Postmaster Tools for your domain reputation trend. Look at the date you increased volume and see if domain reputation dropped from High or Medium to Low around the same time.
Check your SMTP logs or outreach tool dashboard for deferral errors. A 4.7.28 error means you have exceeded Gmail's sending quota. Google's guidance says to stop sending for at least 10 minutes when you see this, then resume with a single connection.
Compare engagement metrics before and after the volume increase. If open rates dropped proportionally more than volume increased, more emails are going to spam.
Run a placement test from each inbox separately. Use the sending limit planner to confirm your per-inbox volumes are within safe thresholds.
The Fix Path
Reduce volume back to where things were working. Go back to the daily send volume where engagement metrics were healthy and hold there while reputation recovers.
If you saw 4.7.28 errors, follow Google's specific guidance: stop for 10 minutes, identify the cause, then resume slowly.
Ramp volume back up over 2–4 weeks instead of jumping. The benchmark is starting new domains at 5–10 emails per day and increasing over 4–6 weeks. Even established domains need a gradual ramp when increasing beyond their established baseline.
If you added new inboxes, warm them independently before putting them into production campaigns. Each new inbox needs its own warmup period. Use the warmup readiness checker to confirm they're ready before activating.
When to Replace Instead of Repair
If you caught the drop quickly and reduced volume within a few days, domain reputation will likely recover in 1–2 weeks of disciplined sending at lower volume with strong engagement.
If you pushed through the drop and kept sending at high volume for weeks, the damage is deeper — recovery could take 4–6 weeks and you may need to rest some inboxes entirely.
For agencies with client commitments, the answer is often both: repair the damaged inboxes by resting them while rotating client campaigns onto clean, prewarmed inboxes from WarmInboxes. This maintains campaign continuity without forcing more volume through damaged infrastructure that will only get worse.
Mistakes That Make This Worse
- Pushing more volume to "make up for" lost replies
- Adding multiple new inboxes and putting them into full production on day one
- Ignoring deferral errors and SMTP 4.7.28 codes
- Not separating new infrastructure from established infrastructure during testing
- Assuming that warmup on one inbox transfers reputation to other inboxes on the same domain
Run the checks first
Before replacing anything, run a free inbox placement test. You might find the issue is DNS, not the domain — and save yourself a week of unnecessary work.