AI search news · ·
A cold email follow-up that went out 9 to 11 days after the first email instead of 5: 20 sent, 2 replies, both the single word "unsubscribe", 0 bounces and 0 report loads
Of 20 cold follow-up emails sent on 2026-09-18, 9 to 11 days after each first email instead of the designed 5, 2 were answered (10.0%), both with the single word "unsubscribe", 0 bounced and 0 of the linked reports were opened. The method is one read of the sending platform's sent list, received list and lead statuses 22.5 hours after the last dispatch, joined to site analytics by the campaign tag on the report links. The 132 emails sent before the pause had drawn 0 replies of any kind.
The question
A sequence is paused mid-flight and the follow-ups go out late. What do the recipients do with a follow-up that arrives after the thread has gone cold? This is the count for one campaign of ours: a two-step sequence to agencies whose first email links a free report already run on one of the agency's own clients. It was paused on 2026-09-11 with 132 emails sent, so that one sentence could be replaced (the record of that edit), and activated again 7.1 days later. The follow-up is designed to go 5 days after the first email. The rows stranded by the pause got it 9 to 11 days after (median 11).
Day one, in counts
| Reading | Count |
|---|---|
| Emails sent on the first day after the pause (the daily limit) | 20 |
| Of which follow-ups to rows that got a first email before the pause | 20 |
| Of which first emails | 0 |
| Bodies carrying the edited sentence and its second link | 20 |
| Sent per mailbox (two mailboxes) | 10 and 10 |
| First and last dispatch, UTC | 13:54 and 15:23 |
| Minutes from first to last dispatch | 89.1 |
| Seconds from the activation request to the first dispatch | 125 |
| Designed days between first email and follow-up | 5 |
| Actual days between first email and follow-up: min, median, max | 9, 11, 11 |
What the recipients did
| Reading | Count | Share of the 20 |
|---|---|---|
| Replies of any kind | 2 | 10.0% |
| Replies whose first line is the one word "unsubscribe" | 2 | 10.0% |
| Automatic replies | 0 | 0.0% |
| Bounces | 0 | 0.0% |
| Unsubscribes recorded by the platform's own counter | 0 | 0.0% |
| Report loads carrying the campaign tag since the activation | 0 | 0.0% |
| Of which loads of the version without pricing (the edited sentence's link) | 0 | 0.0% |
| Emails sent to a row after its reply | 0 |
The two replies arrived 7.6 and 157.8 minutes after the follow-up each answered. Both quote the follow-up, and both consist of the opt-out word the footer asks for, one of them with a sign-off under it. Neither is a spam complaint; both addresses were added to the account-wide block list and to the suppression file every later batch is linted against, and neither was answered. The platform's unsubscribe counter reads 0 because it counts clicks on a platform link, which this campaign does not use; an opt-out by reply shows up only in the received list.
Against the campaign's own record
| Period | Emails sent | Replies of any kind | Bounces | Tagged report loads (persons) |
|---|---|---|---|---|
| Before the pause (first emails and follow-ups, old wording) | 132 | 0 | 5 | 2 (2) |
| Day one after the pause (late follow-ups) | 20 | 2 | 0 | 0 (0) |
| Lifetime | 152 | 2 | 5 (3.3%) | 2 |
132 emails drew no reply at all, and the first 20 after the pause drew 2, both opt-outs. Two things changed at once for those 20 rows: the follow-up carried a new sentence with a second link, and it arrived 9 to 11 days after a first email that itself had gone unanswered. The read cannot separate them. The 40 follow-ups sent before the pause, with the old wording, drew 0 opt-outs. The 0 bounces are expected: every one of the 20 addresses had already accepted a first email (the bounce record by verification status).
The three legs pre-registered on 2026-09-17, scored
| Leg | Statement | Reading | Result |
|---|---|---|---|
| A | The object after the load carries the new sentence in both steps; delays, subjects, daily limit and tracking flags unchanged | 2 of 2 bodies carry the link; 6 of 6 compared fields unchanged; 2 of 2 subjects equal the staged ones | held |
| B | Rows sent between the pause stamp and the activation stamp | 0 | held |
| C | The first email dispatched after the activation is a stranded row's follow-up and carries the forward link | follow up, link present: yes | held |
What is still queued
| Reading | Count |
|---|---|
| Leads in the campaign | 88 |
| Complete (both steps sent, or stopped by a reply) | 60 |
| Pending | 28 |
| Pending follow-ups, all late by the same pause | 23 |
| Pending first emails, the first rows to get the edited first step | 5 |
| Emails sent lifetime: first emails, follow-ups | 92, 60 |
The schedule runs Monday to Friday, so nothing was sent on the read day. The 23 pending follow-ups finish on the next two sending days, and the 5 first emails follow them.
What this read does not show
It does not show opens: the campaign tracks neither opens nor links, by design, so a recipient who read the follow-up and did nothing is invisible here. Report loads are read from site analytics by the tag on the link, which misses a recipient who strips the query string and counts nothing from mail scanners that do not run scripts. 20 rows is a small sample and 2 replies is a count, not a rate to plan on. The read is 22.5 hours after the last dispatch and the day after was a Saturday; a Monday reply would not be in it.
The next read, pre-registered
- 1(A) On 2026-09-24: one-word opt-out replies across every follow-up sent with the edited sentence (the 20 of day one plus the pending ones, which finish on 2026-09-22). Expected 3 to 6; falsified if 2 or fewer (day one was the whole effect), or 7 or more.
- 2(B) On 2026-09-24: report loads carrying the campaign tag from the day-one activation to the read, any version. Expected 0 to 2; falsified if 3 or more.
- 3(C) On 2026-09-24: bounces on follow-ups sent with the edited sentence (each address already accepted a first touch). Expected 0; falsified if 1 or more.
- 4(D) On 2026-09-24: replies that are not an opt-out, on any row of the campaign. Expected 0; falsified if 1 or more.
Any leg outside its statement is published as a falsification with the same script and a dated artifact.
What happens when a cold email follow-up goes out late after a campaign pause?
On this campaign, 20 follow-ups went out 9 to 11 days after the first email instead of 5. 2 recipients (10.0%) replied with the one word "unsubscribe" within 157.8 minutes, 0 bounced, and 0 opened the linked report in the 22.5 hours before the read. The 132 emails sent before the pause had drawn 0 replies of any kind.
Does an unsubscribe reply show up in a sending platform's unsubscribe count?
Not here. The platform's counter read 0 with 2 reply opt-outs in the received list, because the counter records clicks on the platform's own link and this campaign's footer asks for a reply instead. The received list has to be read, and the address suppressed by hand, within the day.
Do follow-ups to addresses that already accepted a first email bounce?
0 of 20 did on this day. The campaign's 5 lifetime bounces (3.3% of 152 emails) were all recorded on first emails before the pause.
Artifact
`content/news/resume-day-one-cut-2026-09-19.json` holds the activation stamps, the object readings after the load, the three scored legs, the day-one dispatch counts and gaps, the recipients' readings, the lifetime counts, the four pre-registered statements with their thresholds, and 20 self-checks (0 failed), one of which asserts that no address, link, name or id is in the file. It is produced by `content/news/resume_day_one_cut.mjs` from one read of the sending platform and one site-analytics count; this page is generated from the artifact by `content/news/resume_day_one_post.mjs` and no number on it is typed.
See your number
See which businesses AI names when your client's buyers ask.
Running this for clients? The $249 agency 5-pack audits five businesses, white-labeled.
Who runs this
- Built and operated by Sensara LLC, Atlanta, Georgia — about us and how the audit works.
- See what the report looks like before you run anything — score per engine, the competitors AI names instead of you, and a fix plan.
- We run the same audit on ourselves every week and publish the result: in the latest run AI named AskedAbout in 1 of 144 answers. We report our own numbers the way we report yours.