Opt-in fatigue and thread health
Consent toggles are late indicators. By the time a LINE OA block or a WhatsApp stop arrives, the conversation has usually been tired for weeks. Messaging Engagement Analytics that only watches the legal switch will congratulate itself on a green dashboard and miss the quieter exit: people still “in” who no longer start threads.
We call that gap opt-in fatigue. It is not a medical metaphor. It is a ledger line: consented contacts whose last useful thread is older than your brand’s own promised cadence.
A Bangkok beauty calendar
A retailer on our cycle twelve list kept OA consent high through a season of drops. Thread starts per consented contact slid from 0.31 a week to 0.11. Nothing in the vendor UI looked like a crisis. Agents, however, were answering a higher share of irritated reopenings—“why are you pinging me”—which is thread health turning into queue noise.
They retired two mid-week broadcasts, which felt commercially brave and statistically modest. First-hour coverage rose because the remaining sends landed on people who still had a reason to talk. Consent barely moved. Fatigue did.
How to see it without a new tool
You need three timestamps: last explicit consent event, last thread start initiated by the customer, last promotional send. If the send keeps happening and the customer start does not, you are spending goodwill that the toggle cannot see.
Do not average this across the whole OA. Split by entry reason: care, loyalty, drop alerts. Drop-alert cohorts fatigue faster; pretending they have the same stomach as care cohorts is how calendars get loud.
The Engagement Atlas keeps this on layer two and layer three together. Reading Seat recordings cover the annotation. Live critique in Inbox Signal Lab is stricter, because someone will ask whether your “still opted in” segment is actually still a conversation. That question is the work.