Send table
Fieldnote Coffee, Klaviyo, weekly Tuesday newsletter. Rules: open rate = unique opens / delivered; open rate excluding MPP = (unique opens minus MPP opens) / delivered; click rate = unique clicks / delivered; bounce rate = bounces / (delivered + bounces); complaint rate = spam complaints / delivered; unsubscribe rate = unsubscribes / delivered. Segment labels are as given.
| Send | Date | Segment | Delivered | Open rate | Open excl. MPP | Click rate | Bounce rate | Complaint rate | Unsub rate |
|---|---|---|---|---|---|---|---|---|---|
| E1 | 09-15 | Engaged 90 days | 20,000 | 45.00% | 18.00% | 3.00% | 0.30% | 0.03% | 0.15% |
| E2 | 09-22 | Engaged 90 days | 20,200 | 45.00% | 18.02% | 3.02% | 0.31% | 0.03% | 0.14% |
| E3 | 09-29 | Engaged 90 days | 20,100 | 45.02% | 18.01% | 2.94% | 0.29% | 0.02% | 0.15% |
| E4 | 10-06 | All subscribers | 41,000 | 28.00% | 8.98% | 1.71% | 0.99% | 0.08% | 0.22% |
| E5 | 10-13 | All subscribers | 41,200 | 27.50% | 8.81% | 1.67% | 1.03% | 0.09% | 0.23% |
| E6 | 10-20 | All subscribers | 41,100 | 27.01% | 8.39% | 1.65% | 1.01% | 0.09% | 0.21% |
Before and after
Period rates use summed counts over summed delivered, not averages of the send rates.
| Measure | Before (E1 to E3) | After (E4 to E6) |
|---|---|---|
| Delivered | 20,000 + 20,200 + 20,100 = 60,300 | 41,000 + 41,200 + 41,100 = 123,300 |
| Unique opens | 27,140 | 33,910 |
| MPP opens | 16,280 | 23,150 |
| Unique clicks | 1,800 | 2,070 |
| Bounces | 180 | 1,260 |
| Spam complaints | 17 | 106 |
| Unsubscribes | 89 | 273 |
| Open rate | 27,140 / 60,300 = 45.01% | 33,910 / 123,300 = 27.50% |
| Open rate excl. MPP | 10,860 / 60,300 = 18.01% | 10,760 / 123,300 = 8.73% |
| Click rate | 1,800 / 60,300 = 2.99% | 2,070 / 123,300 = 1.68% |
| Bounce rate | 180 / 60,480 = 0.30% | 1,260 / 124,560 = 1.01% |
| Complaint rate | 17 / 60,300 = 0.028% | 106 / 123,300 = 0.086% |
Hypotheses ranked
- Audience change (strongest support). Your change note says that on 10-05 the audience switched from 'Engaged 90 days' to 'All subscribers'. Every send from E4 on uses the new segment, and delivered roughly doubled (60,300 to 123,300 per three sends). Total unique clicks rose (1,800 to 2,070) while the click rate fell (2.99% to 1.68%). That pattern fits the same engaged readers plus many less engaged ones, which divides the same activity by a larger base. Supported by E1 to E6 and the 10-05 note.
- Deliverability strain from the wider list (moderate support). The bounce rate rose from 0.30% to 1.01% and the complaint rate from 0.028% to 0.086% after the switch (E4 to E6). Gmail's sender guidelines say to keep the spam rate below 0.10% and never reach 0.30%. Your ESP complaint rate is a different measure from the Postmaster Tools spam rate, so this needs checking there.
- Machine opens shifting (weak support). MPP opens are a large share of opens in both periods, and Apple Mail Privacy Protection can inflate opens. But the open rate excluding MPP also fell (18.01% to 8.73%), so MPP alone doesn't explain the drop.
Ruled out by your change notes: template change, from address change and tracking change ("Template and from address unchanged. No tracking changes.").
Checks to run
- For hypothesis 1, in Klaviyo, report E4 to E6 only for people who are also in 'Engaged 90 days'. If their open and click rates match E1 to E3, the drop is the audience mix.
- For hypothesis 2, open Google Postmaster Tools for your sending domain and compare the spam rate before and after 2026-10-05. Also check which addresses bounced in E4 to E6.
- For hypothesis 3, compare MPP opens as a share of unique opens in each period: 16,280 / 27,140 before vs 23,150 / 33,910 after.
Questions for you
- Was 'All subscribers' cleaned recently? How old are its least recent sign ups?
- Did anything else happen around 10-05 that isn't in the change notes?
Needs your decision
- Whether to go back to 'Engaged 90 days', or keep 'All subscribers' and test a re-engagement step. I haven't changed any segment or setting.