Last Updated on August 14, 2026 by Click Raven
Crawl errors, decay curves, cannibalization, position drift week over week: content teams track all of it. The same teams push every new article to a mailing list that has not been checked since the day it was imported. That list is a distribution channel and it degrades on its own schedule, in many cases faster than the pages it points to. The difference is that nobody produces a report about it. Search Console flags a coverage problem within days. An aging mailing list produces no alert at all, only a quiet decline in the number of people who receive the thing you published.
Distribution is the part of the audit nobody runs
Across 565 US and UK news sites tracked by Chartbeat in July 2025, traffic broke down as 39.26% internal, 19.03% search including Discover, 17.26% external links, 12.99% social referrals and 11.46% direct. Email does not appear as a line in that breakdown. Most of it hides inside direct traffic, since mail clients strip or fail to pass a referrer, and analytics tools only catch the visit when the send carries UTM parameters.
That measurement gap is the whole problem. A channel filed under “direct” by default gets audited the way direct traffic gets audited, which is to say never. The article gets a full SEO postmortem and no distribution postmortem. Impressions, average position, click through rate against the expected curve, all reviewed. Delivered count, bounce rate and the true size of the addressable segment, none of it.
The connection to organic performance runs through distribution and return visits, and it stops there. A newsletter that reaches 8,000 inboxes instead of 12,000 sends fewer readers to the page, produces fewer returning sessions and gives the piece a smaller pool of people who might cite it, link to it or search for it by name three months later. That is the extent of the relationship. Nothing in the ranking algorithm reads your open rate.
Why a list ages faster than the pages it promotes
Contact records decay on a clock that has nothing to do with a publishing calendar. MarketingSherpa research, cited in HubSpot’s Database Decay Simulation, puts B2B contact data decay at 2.1% per month, an annualized rate of 22.5%. Apply that monthly rate across 24 months and a file imported two years ago and never touched has lost roughly 40% of its usable addresses, without a single unsubscribe click and without any signal in the reporting.
Employment churn is the main driver. US Bureau of Labor Statistics data put median tenure with a current employer at 3.9 years in January 2024, the lowest reading since 2002, with 22% of wage and salary workers holding a year or less of tenure with their employer. Each of those moves kills a work address. People update a newsletter subscription on the way out the door roughly never, and IT deletes or hard bounces the mailbox within a few weeks of the exit interview.
Domain migrations do the same damage in a single afternoon. A rebrand, an acquisition or the consolidation of two mail domains invalidates an entire cohort at once. That failure mode looks different in the reporting: a bounce spike concentrated in a handful of domains rather than a slow drift across the file. Abandoned free mailboxes are the slowest of the three and the hardest to see, because Gmail and Outlook addresses often keep accepting mail long after the owner stopped opening any of it.
The three numbers to pull after every send
Pull the same three numbers after every send and the list starts reporting on its own condition.
Delivered count is the denominator for everything else. A campaign showing a 45% open rate on 6,000 delivered out of 11,000 sent is a worse outcome than 38% on 10,800 delivered, and the open rate hides that completely. Reading engagement percentages without the delivered count underneath them is how a shrinking list stays invisible for a year.
Hard bounce rate is the number that describes the list rather than the content. GetResponse, analyzing more than 4.4 billion messages sent by its customers in 2023, reported an average bounce rate of 2.33% across industries. Opt in lists maintained with any regularity sit well under 1%. A hard bounce rate climbing from 0.4% to 2% across three consecutive sends points at the recipient file, since the subject line, the copy and the template all changed between those sends while the file stayed the same.
Providers publish the ceilings, which makes this measurable against something external. Amazon SES documentation recommends keeping the bounce rate under 5% and states that a bounce rate above 10% may cause SES to pause an account’s ability to send. The same page sets the complaint rate target under 0.1%, with a pause possible above 0.5%. Bounces and spam complaints are separate metrics, and they tend to move together: a file full of dead addresses is also a file full of people who forgot they ever subscribed.
The third number is the split between hard and soft. Soft bounces come from full mailboxes and temporary refusals, and they clear on their own. Hard bounces are permanent: no such mailbox, no such domain. Rising soft bounces with flat hard bounces usually indicate a reputation issue at one specific provider, worth a look at Postmaster Tools. Rising hard bounces indicate the file is aging, and no amount of subject line testing will move that number.
Cleaning cadence: quarterly, then again before the big send
Because decay runs continuously while sends happen on a calendar, the cleaning schedule needs two layers.
Quarterly, run the full file through validation. Three months at 2.1% monthly decay works out to about 6% of the list gone, enough to push a bounce rate from healthy to visible, small enough to fix without a heavy batch job. A quarterly rhythm also matches how most content teams already operate, since it lines up with the same review cycle that drives content pruning and refresh decisions.
Before a large send, the work is narrower and faster. Validate anything added since the last full pass, plus any segment dormant for more than 6 months. Addresses collected at a conference or typed by hand into a spreadsheet deserve their own pass, since they carry typo rates that organic signups never approach.
Manual checks cover the long tail. If you only need to check a handful of addresses before a send, a free email tester will tell you in seconds whether the mailbox still accepts mail. That handles the practical cases: an embargo list of 20 journalists, or the address a colleague forwarded with a typo somewhere in the domain.
One rule holds regardless of cadence. Leave a gap between the first cleaning pass on a badly aged file and the send that follows it. A pass on a neglected list can remove a large share of the recipients at once, and an abrupt drop in volume against a mailbox provider attracts the same scrutiny as an abrupt increase.
When suppression beats re-engagement
There is a difference between a subscriber who stopped opening and a mailbox that stopped existing. The first is an engagement problem, the second is a data problem, and they call for opposite responses.
Re-engagement fits a valid address with a real person behind it who has ignored the last 10 sends. One message asking whether they want to stay on the list, with a working one click unsubscribe, is cheap and settles the question either way. Google‘s guidelines back this directly, telling senders to periodically confirm that recipients want to stay subscribed and to consider unsubscribing those who never open.
Suppression wins once the address itself is in doubt. A hard bounce goes onto the suppression list immediately and permanently, with no retry, because mailbox providers read repeat sends to a known dead address as evidence that the sender ignores bounce feedback. Role accounts and catch all domains belong there as well, since they inflate the delivered count while producing no sessions.
Deletion becomes the right call in one specific case: a segment where the hard bounce rate exceeds the engagement rate. At that point the segment costs more in sender reputation than the traffic it returns, and reputation applies to every send, including the ones going to the healthy part of the file. A newsletter reaching 4,000 engaged subscribers drives more people to a published article than one nominally holding 12,000 records and landing half of them in a spam folder.
The audit takes an afternoon. Export the file, check the age of each record against the date it was added, run validation, then compare the hard bounce rate across the last 6 sends. Content teams already do exactly this for pages, tracking which URLs lost impressions and which ones stopped converting. The mailing list works the same way, and the numbers to read are fewer.