A Quiet Update That Is Costing Senders 11 to 19 Points of Placement
If your Gmail metrics have looked shakier than usual since late July, you are not imagining it.
Gmail has been rolling out an inbox-placement algorithm update incrementally since mid-July 2026. By late August, multiple deliverability vendors — including Validity and GlockApps — confirmed the change: Gmail now down-weights broadcast campaigns sent to low-engagement segments far more aggressively than it did before.
Here is what you will learn in this article: what Gmail actually changed, why it is tanking delivery for senders with large dormant segments, how the Promotions tab relevance sorting works, and the exact segmentation and suppression playbook that protects inbox placement.
For senders running large, lightly segmented lists, the reported impact is real.
Average inbox placement on Gmail has dropped between 11 and 19 percentage points for operators who have not adjusted their send cadence or list hygiene.
This is the biggest shift to hit email inboxes since Apple's Mail Privacy Protection disrupted open-rate reporting in 2021. Understanding what changed is the difference between protecting your Q4 revenue and watching your reach quietly collapse.
What Actually Changed in Gmail's Ranking Model
For years, Gmail's delivery decisions were dominated by two things: whether your mail passed authentication, and how clean your domain's long-term reputation was.
The 2026 update shifts the balance of power. Recent engagement velocity now outweighs historical sender reputation.
In plain terms: Gmail no longer cares as much that your domain has a clean 18-month sending history if the last 30 days of recipient interaction look flat. The algorithm looks at how people are behaving with your mail right now, and it ranks you accordingly.
What Is Engagement Velocity?
Engagement velocity is the speed and consistency with which recipients interact with your mail right now. Gmail measures it over a rolling window, typically the last 30 days, using signals like opens, clicks, replies, forwards, and folder moves.
A concrete example makes the concept clear. Imagine two domains with identical 18-month histories. Domain A gets 40% of recipients opening its last three sends. Domain B gets 12% opening its last three sends. Under the 2026 model, Gmail ranks Domain A higher, regardless of the fact that both domains look identical on historical reputation.
The key takeaway: engagement velocity is a current snapshot, not an accumulated score. It changes with every campaign you send.
The 60-Day Negative Signal
The most consequential part of the change is how Gmail now treats inactive subscribers.
Previously, subscribers who had not engaged were treated as neutral — they simply did not count toward your engagement signals. In the 2026 model, subscribers who have not opened or clicked in 60 or more days are now treated as active negative signals.
This is a critical distinction. It is no longer just that unengaged addresses contribute nothing positive. They actively pull your overall sender score down, and the damage applies to the entire send, not just to those individual addresses.
A subscriber who would have engaged loses inbox visibility because a large dormant segment on your list is dragging down the campaign. This is the mechanism behind the 11–19 point placement drops senders are reporting. It also means the natural decay of engagement over time now carries a direct, measurable cost, and the question of how long a subscriber stays valuable is no longer academic — it determines whether your next send reaches the inbox.
Why a Clean History No Longer Protects You
The most disorienting part of this update is that it punishes senders who did everything right for years.
A domain that built a strong reputation through careful list building and consistent sending can still see placement fall if its recent engagement has gone flat. Gmail's model now treats recent activity as the primary signal, and a long, clean history alone is no longer enough to keep you at the top of the inbox.
This is the shift that most confuses teams. Deliverability professionals are used to thinking in terms of accumulated reputation. The 2026 model rewards momentum, not history. You are only as strong as your last 30 days of engagement.
The Vicious Cycle
The update also creates a feedback loop that makes the problem worse over time.
- Low engagement sends your placement down
- Lower placement means fewer subscribers see your mail
- Fewer opens and clicks mean even lower engagement signals
- Your placement sinks further
This is why the update disproportionately affects senders who keep broadcasting to dead segments. Every send to a dormant address trains Gmail to treat your mail as ignorable, which depresses visibility for your engaged subscribers too.
The Gmail Promotions Tab Is Now Relevance-Sorted, Not Chronological
A second change compounds the first. The Gmail Promotions tab — the inbox tab that holds promotional and bulk email — no longer sorts strictly by recency.
Gmail has been shifting toward relevance-based sorting since September 2025, when "Most Relevant" became the default sort option. By 2026, this ranking is driven by the same engagement-velocity signals as the broader inbox update.
| Gmail Change |
When |
Impact |
| "Most Relevant" default in Promotions tab |
September 2025 |
Email ordered by predicted relevance, not arrival time |
| Engagement-velocity weighting rollout |
Mid-July 2026 |
Recent interaction outweighs historical reputation |
| 60-day inactivity as negative signal |
Confirmed late August 2026 |
Unengaged segments drag down the whole send |
| Users can still sort by recency |
Ongoing |
Most users keep the default relevance sort |
The practical consequence: delivery no longer guarantees visibility. An email can land in the Gmail inbox and sit functionally unseen because the algorithm ranked it below a dozen senders the subscriber engages with more.
Your 9 a.m. send can appear below a three-day-old email from a brand the recipient opens more often. Sending time matters far less than it used to. What matters is how much your specific sender has earned the subscriber's attention.
How This Interacts With the Bulk Sender Rules
It is important to be precise about the relationship between this update and the bulk sender requirements that Gmail and Yahoo introduced in 2024.
Authentication is still mandatory. So is RFC 8058 one-click unsubscribe for marketing mail from bulk senders. None of that has changed.
But the 2026 update clarifies the role these rules play. Authentication is no longer the test. It is the ticket that lets you sit for it.
A sender who fails SPF, DKIM, or DMARC, or who lacks compliant unsubscribe headers, gets rejected at the door. But a sender who passes every compliance check still faces the engagement test. The algorithm now decides where compliant mail actually lands based on recent engagement velocity.
This is why you can have a technically perfect setup and still see declining placement. Compliance keeps you in the game. Engagement determines how well you play it.
The Playbook: What to Do Now
The good news is that the response to this update is well understood. It is not mysterious, and it does not require abandoning email. It requires operating the way high-performing ecommerce programs have always operated.
1. Segment by Engagement Recency
Stop treating your list as a single audience. Build engagement tiers and treat them differently.
| Tier |
Last Engagement |
Strategy |
| Active |
Within 30 days |
Full send frequency, primary offers |
| Warm |
31–60 days |
Reduced frequency, lighter touch |
| Cold |
61–90 days |
Re-engagement sequence, not broadcasts |
| Suppressed |
90+ days |
Exclude from general sends |
Each tier should receive different send frequency, different subject line approach, and different offer treatment. This is the single highest-leverage change you can make under the new algorithm.
2. Suppress, Don't Just Clean
Suppression is now a growth lever, not a cleanup chore.
Cutting sends to subscribers with no engagement in 90-plus days reads as a loss in reach on paper. It performs as a gain in visibility in practice. When you concentrate your volume on demonstrated-intent segments, your engagement density rises, and the algorithm rewards you with better placement for everything you send.
For a deeper breakdown of the tiers and the win-back mechanics, see our guide to writing a re-engagement email and our glossary entry on email engagement recapture.
3. Run One Re-Engagement Sequence, With an Exit
The disengaged get exactly one deliberate win-back sequence, with a clear endpoint.
After that sequence, non-responders leave the active file. Endless monthly attempts at the same dead segment are precisely the behavior the new algorithm penalizes. A single, well-timed sequence with a strong incentive either brings them back or confirms they are gone.
4. Move Your DMARC Policy Forward
Data from inbox testing in late August 2026 shows that senders with DMARC set to p=none are recovering more slowly than those already at p=quarantine or p=reject.
If your DMARC is still at p=none, moving it to p=quarantine is a meaningful step. It is not directly a ranking signal, but it strengthens your authentication posture and is associated with faster recovery curves under the new model. Our guide to email authentication methods covers the full migration path.
5. Test Placement Before You Send
Before any large broadcast, verify where your mail is actually landing.
Run a placement test across Gmail, Yahoo, and Outlook before the send. If your Gmail placement is under 85%, do not send until you have diagnosed the drop. Pre-send testing turns deliverability from a reactive afterthought into a proactive check. Our email deliverability checklist walks through the full pre-send audit.
6. Watch the Complaint Rate Closely
The update makes list quality more consequential, because a rising complaint rate compounds the engagement problem.
Keep your spam complaint rate well under 0.1%. Gmail's hard ceiling is 0.3%, but you never want to approach it. A complaint from someone who could not cleanly unsubscribe costs you far more reputation than a missed open ever will. Our article on why emails go to spam covers the full diagnostic order.
How to Tell if Your Domain Is Already Affected
You may be wondering whether this update has already hit your account. A few checks will tell you quickly.
- Compare engagement to placement. If opens and clicks have held steady but inbox placement has fallen, that is a signal of re-ranking, not a content problem.
- Look at your unengaged share. Calculate the percentage of your last campaign that went to subscribers with zero engagement in 90 days. That share is what the algorithm is grading you on.
- Check your Gmail metrics separately. Gmail-specific placement and engagement tell a different story than your overall ESP averages.
If any of these reveal a large dormant segment, the fix is the playbook above: segment, suppress, and re-engage before you add more volume.
Key Takeaways
- Gmail's 2026 algorithm update, rolled out since mid-July and confirmed in late August, now ranks senders on recent engagement velocity rather than historical reputation
- Subscribers with no engagement in 60+ days are treated as active negative signals, dragging down placement for the entire send
- The Promotions tab sorts by relevance, not recency, so delivery no longer guarantees visibility
- Authentication and RFC 8058 compliance remain mandatory, but they are the entry ticket, not the ranking factor
- The response is to segment by engagement recency, suppress dormant segments, run one re-engagement sequence with an exit, and test placement before large sends
Related Articles
Related tools: See how your list's engagement compares across segments with the Email Engagement Score Calculator, or measure the deliverability impact of your sending setup with the Email Deliverability Calculator.
