Research8 min read

Synthetic engagement signals Google stopped counting

A short list of what Gmail used to count that it now mostly doesn't — opens, pool-replies, pool mark-as-important, tracking-pixel fires. And what it's counting instead.

Every warmup vendor's sales pitch assumes Gmail is still counting the engagement signals it was counting in 2020. Most of them aren't worth what they were.

TL;DR

Opens are noise post-Apple MPP. Synthetic replies from coordinated pools are devalued. “Mark as important” inside known pools contributes little. Tracking-pixel fires are useless for reputation. What Gmail is counting: complaints, filtered-at-delivery, authentication alignment, diversity of real engagement, and dwell-then-reply.

Opens

Apple Mail Privacy Protection prefetches every image in every email regardless of whether the user opened the message. Most corporate networks pre-click links for security scans. The result: “opened” is a synthetic signal for a majority of mail. Gmail never relied heavily on opens for reputation, but any vendor that tells you their warmup works because “opens went up” is measuring something irrelevant.

Replies inside coordinated pools

Reply rate is a real signal. Reply rate inside a detected coordination cluster is discounted at the filter layer. The graph-closure signal (see pool fingerprinting) identifies clusters where the replies are mutual and the contents are fraternal. Those replies don't count like a real prospect reply does.

“Mark as important” from pool members

This was the warmup pool's ace in the hole. Filters now match the action to the actor. A mark-as-important from a pool member with 0 out-of-pool activity is weighted near zero; the same action from a long-time active Gmail mailbox with diverse legitimate activity still counts. Pools are running their plays against a scoring system that's figured out who's doing the scoring.

Tracking-pixel fires

Pixel fires were never a Gmail reputation input — but a lot of operators talk as if they were. They measure something for you (opens, kinda, post-MPP poorly). They measure nothing for Gmail's sender classifier.

What Gmail counts now

  • Complaint rate — the “Report Spam” button. Heaviest input. Kept below 0.1% is great, 0.3% is the public threshold.
  • Filtered-at-delivery rate — the proportion of your mail the content filter marks as spam before the user sees it. Visible in Postmaster Tools.
  • Spam-trap hits — one hit can crater an IP/domain for weeks. Zero tolerance.
  • Authentication alignment — DMARC pass rate, DKIM signature validity, SPF record correctness.
  • Engagement diversity — replies, forwards, adds-to-contacts coming from many unique domains, not a repeat pool.
  • Dwell-then-action — messages that the user opened for > 5 seconds and then replied/forwarded/archived-to-inbox. The combination is hard to fake.
Check your real engagement-driven placement

Run a free placement test to see what Gmail actually files your mail as. Our seed mailboxes aren't in any warmup pool — the placement they see is the placement your prospects see.

What to do with this

  1. Stop celebrating open rates. They aren't noise-free and weren't the signal anyway.
  2. Watch complaint rate daily. Any tick above 0.2% pauses the sequence.
  3. Watch Postmaster domain reputation weekly. If it slips, something is wrong now — fix it before it slips further.
  4. Drive real replies. Nothing built from synthetic engagement outperforms a modest real reply rate from real prospects.

FAQ

Is this officially confirmed by Google?

Nothing here is lifted verbatim from a Google press release. It is inferred from published Gmail sender documentation, Postmaster Tools data models, and independent placement-test observations across tens of thousands of senders. Google doesn't publish its classifier weights.

Do warmup tools outright admit this?

No. Their pricing model requires that the signal is still worth what it was. Expect marketing to stay bullish until the category is undeniably cooked.

What signals still make my warmup worth it?

The audit-and-alert parts of a bundled warmup tool are real. The pool traffic is what's decayed. If you're paying for a tool, you're paying for the non-pool parts.
Related reading

Check your deliverability across 20+ providers

Gmail, Outlook, Yahoo, Mail.ru, Yandex, GMX, ProtonMail and more. Real inbox screenshots, SPF/DKIM/DMARC, spam engine verdicts. Free, no signup.

Run Free Test →

Unlimited tests · 20+ seed mailboxes · Live results · No account required