Cold Email Isn't Dead. Volume Is. I Read All Three Bulk Sender Policies and the Ceiling Is 30 Complaints Per 10,000.

TL;DR: Cold email is not dead in 2026, but the volume strategy is. On 2026-08-15 I read the current bulk-sender policies published by Google, Yahoo and Microsoft and compared them across four requirement categories. Two of the three publish a numeric spam-complaint ceiling and both set it at 0.30%, which is 30 spam reports out of 10,000 inbox-delivered messages, and Google's own guidance tightens that to 0.10%, or 10 reports per 10,000. Two of three publish a volume trigger and both set it at 5,000 messages a day. All three now require SPF, DKIM and DMARC. The channel still works. Sending as though each message costs nothing does not.

"Cold email is dead" gets declared every January and has never been true. Something did change, though, and it is more specific than the obituary. Between October 2023 and November 2025 the three companies that run the consumer inbox turned enforcement on, and they published the numbers. So I read them.

What did the policy audit show?

I fetched eight vendor pages on 2026-08-15: Google's Email sender guidelines, sender guidelines FAQ, Postmaster Tools docs and October 2023 announcement; Yahoo's Sender Requirements, Sender FAQs and Postmaster blog; and Microsoft's April 2025 high-volume sender post.

Requirement	Google (personal Gmail)	Yahoo	Microsoft (Outlook.com)
Numeric volume trigger	5,000 a day, permanent	none published	5,000 a day
SPF + DKIM + DMARC at p=none minimum	required	required	required
One-click unsubscribe	required	required	recommendation only
Numeric complaint-rate ceiling	0.30%, with 0.10% guidance	0.30%	none published

Three of three require the authentication stack. That is the only thing all three agree on in writing, and it is not a bar every vendor selling you compliance clears: I ran SPF and DMARC lookups against 30 of them. Everything else is two out of three.

Yahoo is the one that will not publish a volume number, and its FAQ says so outright: "A 'bulk' sender is classified as an email sender sending a significant volume of mail. We will not specify a volume threshold." Microsoft is the one that will not publish a complaint number, and its unsubscribe rule sits under a heading called "Additional Email Hygiene Recommendations", worded as "provide an easy, clearly visible way for recipients to opt out." That is guidance, not a threshold.

Google's version has a detail that gets skipped: crossing 5,000 once is permanent. The FAQ says "senders who meet the above criteria at least once are permanently considered bulk senders" and "bulk sender status doesn't have an expiration date."

What does a 0.3% ceiling actually mean in messages?

Multiply it out. At the 0.30% ceiling, 10,000 messages delivered to inboxes buys you 30 spam reports before you are over. At Google's stricter monitoring line of 0.10%, it is 10 reports per 10,000. Google's FAQ is explicit that the softer number already bites: "even today, user-reported spam rates greater than 0.1% have a negative impact on email inbox delivery for bulk senders."

The denominator is the part that turns this from a rule into a trap. Google defines the metric as "the percent of your messages that are delivered to engaged recipient's Inbox and then marked as spam by the recipient." Yahoo said the same thing in its October 2025 Insights announcement: "the Spam Complaint Rate shown in Sender Hub is based only on messages delivered to the inbox." Not messages sent. Messages that landed.

So the ratio moves against you exactly when you are already in trouble. Send 10,000 and get 3,000 into inboxes because filtering took the rest, and your 0.30% budget is now nine spam reports, not thirty. The metric accelerates in the direction you are already falling.

Two things surprised me. The ceiling has no volume floor: Google and Yahoo both list the 0.3% spam rate under requirements for all senders, not just bulk senders. Spraying 200 messages a day across forty domains does not get you under it, because there is no "under it" to get to. And Postmaster Tools will not show you your own number at that scale, since the dashboard "might not include all data on days when outgoing email volume is low." The rule applies to you and the instrument does not.

Why does this kill volume rather than the channel?

Because the count above is a cost sheet, and the volume play was priced at zero. Five dated milestones in twenty-five months, every one in the same direction: Google and Yahoo announce on 2023-10-03; the requirements take effect in February 2024; one-click unsubscribe becomes mandatory and a spam rate above 0.3% makes a bulk sender ineligible for delivery mitigation in June 2024; Microsoft begins enforcing on Outlook.com after 2025-05-05; and Gmail, per its own FAQ, starts "ramping up its enforcement on non-compliant traffic" in November 2025, "including temporary and permanent rejections."

Each new sending domain now needs SPF, DKIM, DMARC, valid forward and reverse DNS, TLS, and a reputation built from nothing. The forty-throwaway-domains architecture did not become illegal. It became a setup and warm-up bill, repeated per domain, defended against a ratio you cannot see. Its entire premise was that sending was free.

The uncomfortable part for anyone hoping to just switch channels: the same ceiling appears where nobody is enforcing anything. Reachium's volume study of 180,155 matured LinkedIn connection requests, an outside dataset measuring LinkedIn and not email, found acceptance peaked at 32.01% for accounts averaging 10 to 19 invites a day and fell to 26.65% at 20 to 29 a day, a gap of 5.36 points, or 3.98 with the single largest contributor removed. The under-10 band was lowest of all at 22.77%, so this is a peak in the middle, not a clean "less is more." Recipients apply a volume tax on a platform with no published complaint threshold. The mailbox providers just wrote theirs down.

I cite no reply-rate or open-rate benchmark here on purpose. Every widely-quoted cold-email benchmark I found measures one vendor's own customers, and open rates have been unreliable since mail-privacy prefetching started firing pixels nobody looked at. I traced fifteen of those stats in the mail merge versus AI personalization piece and checked eight outreach vendors against their own marketing claims in this audit.

What would change my mind?

The strongest counterargument is one the documents make themselves. Google's FAQ says plainly that "the Email sender guidelines don't apply to messages sent to Google Workspace accounts", and Microsoft's post says its rules cover "Outlook.com, our consumer service." Most B2B cold email goes to Workspace and Microsoft 365 tenants, so the letter of these rules does not bind it. What binds it is the same filtering infrastructure and the same complaint signal. Show me a non-vendor sample where B2B reply rates held steady at high per-domain volume through the November 2025 Gmail ramp, and the enforcement is consumer-only in practice as well as on paper.

Two other things would move me. If Google published a way to shed bulk-sender classification, the permanence argument weakens. And if anyone documented a sender sustaining 5,000-plus cold messages a day under 0.1% complaints for a year, the ceiling is survivable at volume and I picked the wrong constraint. Until then the operator move is the boring one: fewer and better-targeted sends, a channel mix that does not stake everything on the inbox, and a real unsubscribe path. The cheapest non-inbox lead is somebody who asked, which is what the comment-to-DM loop is engineered to produce. The LinkedIn side of that mix has its own limits, and they rhyme. Thirty complaints per ten thousand is not much room. It was never meant to be.