Gmail Manage Subscriptions: where your send count lands

Gmail Manage Subscriptions puts every sender a reader gets mail from on one screen, sorted by how often they send, with a count next to each name and an unsubscribe button beside it. Google shipped it on July 8, 2025. More than a year later, nobody outside Google can say for sure how far back that count looks, and that one missing number decides which bucket your brand sits in.

The buckets are "20+ emails recently", "10-20 emails recently" and "10 or fewer emails recently". We took the niche send-frequency medians we publish and ran them through both windows testers have proposed. The result is lopsided. Under one reading, almost nobody looks noisy. Under the other, almost everybody does.

What Gmail Manage Subscriptions shows each reader

Each row is one sender, with a count of how many emails it sent the reader "in the past few weeks", to use the phrase from Google's launch post. One tap unsubscribes, and Gmail sends the request on the reader's behalf through the sender's List-Unsubscribe header. It rolled out on web, Android and iOS, in select countries first.

Three details matter more than they look. MarTech reports the list is sorted by volume first and alphabetically second, so your heaviest-sending competitors sit at the top of the screen. Zeta Global notes that each sender address gets its own row, so a brand mailing from news@ and deals@ shows up twice. And the count is personal. Gmail shows how many emails you sent that reader, not how many you sent your list.

That last one is why MarTech's headline called the feature a win for segmented senders. A subscriber you held back from four promotional sends sees a smaller number than one who got everything. Suppression finally shows up somewhere the reader can see it.

One oddity: the labels overlap. A sender at exactly 10 fits both "10-20" and "10 or fewer", and one at 20 fits "20+" and "10-20". Google hasn't said which way the edges round, so treat any count sitting on 10 or 20 as a coin flip.

Three weeks or three to four months?

Brian McKenna, VP of CRM at DMi Partners, wrote in MarTech on July 29, 2025: "Based on our testing, 'recent' refers to emails sent within the past three weeks." On March 19, 2026, Chad S. White of Zeta Global wrote that "recently" appears to be in the range of 3-4 months. Neither published a method, and the gap between the two is a factor of four to six.

So either one of them is wrong, or Google changed the window somewhere in those eight months. Google hasn't said. Its own copy, "the past few weeks", reads a lot closer to three weeks than to four months, and if we had to bet, that's where we'd put the money. But we wouldn't plan around a bet, so the table below runs both.

Where each niche median lands in Gmail Manage Subscriptions

The table takes the median sends per week for each niche from our newsletter statistics hub (Newsletrix analysis corpus, Q1 2026) and multiplies it out. Three weeks is x3. For three to four months we use 13 and 17 weeks as the bounds. To be plain about the method: this is arithmetic on figures we already publish, not a new measurement. We haven't counted anyone's Manage Subscriptions rows. It also assumes a reader who gets every send.

Niche	Median sends/week	3-week count	Bucket (3 weeks)	13-17 week count	Bucket (3-4 months)
B2B SaaS	1.0	3	10 or fewer	13-17	10-20
Ecommerce / DTC	2.5	7.5	10 or fewer	32.5-42.5	20+
Media / News	5.0	15	10-20	65-85	20+
Creator / Personal	1.0	3	10 or fewer	13-17	10-20
Fintech	1.2	3.6	10 or fewer	15.6-20.4	10-20, touching 20+
Developer tools	1.0	3	10 or fewer	13-17	10-20
Health & wellness	1.5	4.5	10 or fewer	19.5-25.5	Straddles 10-20 and 20+
Education	2.0	6	10 or fewer	26-34	20+

Under three weeks, seven of the eight medians sit at 10 or fewer. Only media and news, at five sends a week, reaches 10-20. A B2B SaaS newsletter at its median cadence shows a 3, which on a busy reader's screen is practically invisible.

Under three to four months, nobody stays at 10 or fewer. A single weekly send lands at 13-17. Ecommerce, media and education go to 20+, health and wellness straddles the line, and fintech touches it at the 17-week end.

So the dispute decides what drives your row. Under the short window, a steady cadence barely registers and a spike is what moves you. Under the long window, your baseline has already picked your bucket and a campaign week hardly shifts it. Good engagement doesn't exempt you either. In the same dataset, the top quartile of ecommerce senders by engagement runs two to four sends a week, and at four a week the three-week count is 12. That's 10-20 before a single promotion goes out.

Black Friday is where retailers get caught

Our Black Friday send volume benchmarks, built from mail received between November 1 and December 5, 2025, show the median retail brand roughly doubling its cadence in peak week. Against the 2.5-a-week ecommerce baseline, that's about five sends in seven days. The top decile runs four to six times baseline, which is 10 to 15 emails in one week from one brand.

Put that into a three-week window. The median brand gets five in peak week plus 2.5 in each of the two weeks around it, for a count of 10. That's exactly the edge Gmail's labels leave unresolved. The top decile gets 10 to 15 plus five, so 15 to 20.

And those are floors, because the weeks around peak week aren't baseline weeks. Early-access sends start around November 10 to 12. Extended-sale and last-chance sends run through the Tuesday and Wednesday after Cyber Monday. This year Black Friday is November 27 and Cyber Monday is November 30, so a reader who opens Manage Subscriptions on Wednesday, December 2 is looking at a three-week window that starts on November 11. That's the whole campaign, ramp to tail, and none of your quiet October.

Here's the part we think most retail teams haven't priced in: under the short window, the sort punishes a spike far harder than a steady cadence. A brand that sends three a week all year and 12 in peak week sits higher on that December 2 screen than a brand that sends four a week and holds it straight through November.

The 300% unsubscribe claim is mostly double counting

Zeta's March piece reported that senders' unsubscribe rates "have jumped by 200% or even 300%". That line travels without the rest of the paragraph. In the same article, Zeta explains that Gmail submits multiple unsubscribe requests every time someone clicks the link in Manage Subscriptions, which inflates total unsubscribe rates, and calls the spike "mostly an illusion". Unique unsubscribe rates rose only modestly.

DMi's July 2025 finding was that its clients had not seen a statistically significant increase in unsubscribe rates for Gmail subscribers. The two fit together once you separate events from people. If your ESP logs every one-click request as its own unsubscribe, your December report will show a spike that is mostly Gmail repeating itself. Deduplicate by address before anyone panics, then compare against the unique rates in our unsubscribe rate benchmarks.

We'd also push back on treating every unsubscribe from this screen as a loss. A reader who finds the button in two taps is a reader who doesn't hit Report spam instead. Gmail's bulk sender requirements put the hard ceiling at 0.3% spam complaints and ask senders to stay under 0.1%, and our spam complaint rate benchmarks put best-in-class senders at 0.01% to 0.05%. An unsubscribe costs you one address. A complaint counts against your reputation with every Gmail reader you have.

What to change before November

Start with your own numbers, per From address and per segment. Count what your most-mailed segment received in the last three weeks and in the last 13 to 17, then find your row in the table. If your ongoing cadence runs above 2.5 times your niche median, our send frequency benchmarks by niche already put you in burnout territory, Gmail or no Gmail.

Then open the Black Friday calendar and cut from the back.

The tail sends after Cyber Monday go first. Our Black Friday analysis made the case on engagement grounds: the least differentiated subject lines of the window, landing on a list that has just absorbed a week of pressure. Gmail adds a second reason. Under a three-week window, a send on November 12 drops out of the count by December 3. A send on December 2 stays in it until December 23. The ramp ages out early, while the tail sits in the count for most of December, which is exactly when our Black Friday data shows unsubscribe rates are still elevated.

If you can't bring yourself to cut the tail, segment it. Send the extended-sale email only to people who clicked during peak week. Everyone else sees a smaller number on December 2, and that is the whole point of a per-reader count.

Split mailstreams if your newsletter and your promotions share an address. Zeta recommends it, and we agree. Our reason: a reader tired of promotions can drop that row and keep the newsletter. It has a cost. Two rows means two unsubscribe buttons, and a reader in cleanup mode may tap both.

For the sends that survive, make each subject line say why it exists. "Early access ends tonight" with a real deadline beats a third "Don't miss out", so run the survivors through our subject line tester before they get scheduled. If you'd rather know where competitors sat last November than guess, count their sends from received mail. Our Milled comparison covers what a retail-only archive can and can't tell you.

Do your own count now, before the November ramp muddies the baseline.

Frequently asked questions

How does Gmail decide who appears in Manage Subscriptions?

Gmail lists your subscription senders sorted by how many emails each one sent you recently, then alphabetically. Each sender address gets its own row, so a brand mailing from two addresses appears twice. The count is per reader: it shows how many emails that sender sent you, not how many it sent its whole list.

What does 'recently' mean in Gmail Manage Subscriptions?

Google has not published the window. Its launch post says 'the past few weeks'. DMi Partners tested it in July 2025 and put the window at three weeks, while Zeta Global estimated 3-4 months in March 2026 without publishing a method. Plan for both: the short window for campaign spikes, the long one as the worst case for your baseline.

Will a weekly newsletter show up as 20+ emails?

No, not under either reading. One email a week counts 3 over three weeks, which is the 10-or-fewer bucket, and 13 to 17 over three to four months, which is 10-20. Getting past 20 takes about seven sends a week under the three-week window, or roughly 1.2 to 1.6 a week under the longer one.

Does Gmail Manage Subscriptions increase unsubscribe rates?

It depends on what you count. Zeta Global reported total unsubscribe rates up 200 to 300 percent, but also found Gmail sends multiple unsubscribe requests per click, so most of that spike is double counting and unique unsubscribe rates rose only modestly. DMi Partners saw no statistically significant increase for Gmail subscribers in July 2025. Deduplicate by address before you report.

Does Black Friday week move my bucket?

Under a three-week window, it can. The median retail brand roughly doubles its cadence in Black Friday week, which puts a three-week count near 10, right on the boundary between buckets. A top-decile brand sending 10 to 15 emails that week reaches 15 to 20 even if the weeks either side stay at baseline, and they rarely do.

Related reading