Frequency Fatigue Is a Preference Signal. Read It Before It Becomes an Unsubscribe.

Opens decay, then clicks, then someone reports spam out of irritation rather than malice. The fatigue curve is readable weeks before the unsubscribe spike, if anyone graphs it per customer instead of per campaign.

Email fatigue announces itself quietly and per person, which is why campaign level reporting misses it. The campaign dashboard shows a healthy list average while, inside it, a specific customer's opens slide from most sends to some sends to none, their clicks die a month before their opens do, and the final event, an unsubscribe or a spam report, looks sudden only because nobody was watching the slope. Read per customer, fatigue is one of the most predictable events in marketing. The customers who fatigue share a cause more often than a profile: the store increased cadence around a sale season and never came back down, or a second programme, a new product line, a loyalty flow, started mailing the same people on a second clock, and the two clocks together doubled the load while each team reported its own frequency as reasonable. Fatigue is a portfolio problem wearing a per campaign disguise. The fix starts with counting what a customer actually receives from every programme combined, which is the only frequency the customer's inbox has ever experienced.

The combined cap is the only honest number

Set a house cap on total sends per customer per week across campaigns, automations, and product lines, and let programmes bid against it by expected value instead of each owning an unlimited claim. This is an argument inside most companies, because every programme can prove its own sends perform. The inbox does not grade programmes separately. A customer who gets nine good emails in a week is a customer having a nine email week, and the fatigue curve prices the total, not the intentions.

Automations deserve special suspicion because they are invisible in campaign planning. A browse abandonment flow, a replenishment reminder, and a loyalty nudge can land on the same customer on the same day, each correctly triggered, each individually sensible, summing to a pile up no human approved. A daily collision report, customers receiving more than a set number in 24 hours, is a one query piece of hygiene that routinely embarrasses the first team that runs it.

Related reading on this site: PrefEngine and Stated Preferences Lie Politely. Revealed Preferences Bill Accurately. cover the neighbouring problems in more depth.

Recover slope, not just the endpoint

Intervene on the slope, while the customer still opens something. A customer whose engagement halves over three weeks gets cadence halved in return, automatically, and the best performing programme keeps its slot while the rest pause. That trade, fewer sends to cooling customers, is usually revenue positive within a quarter, because the sends cut were heading for zero anyway and the preserved subscribers resume buying at their natural pace. Waiting for the unsubscribe and then suppressing is locking up after the horse has not only left but filed a spam report on the way out, which costs deliverability for the whole list.

Win back flows aimed at the fully lapsed should be honest about why they left. One email, we have been sending too much, here is a monthly option, outperforms the discount ladder for the fatigue segment, because the problem was never price. Misdiagnosing fatigue as price sensitivity and answering volume with discounts trains your best customers to wait for the apology coupon.

Let customers set the ceiling and keep it

A stated frequency preference outranks every model. The customer who chose monthly has opted out of your cadence experiments, your sale season exceptions, and your reactivation urgency, and honoring it through the loud months is what makes the preference system real. The stores that keep these subscribers longest treat the chosen ceiling as a deliverability asset: a cohort that never fatigues, never reports spam, and opens at rates the main list has not seen since it was small. That cohort is worth more as a kept promise than as a few extra sends in December, and it is not close.

Frequently asked questions

What are the early signs of email fatigue in a customer?

A per customer decline in opens over several weeks, with clicks usually falling first and faster. The pattern is visible in individual engagement history well before the unsubscribe or spam report, which campaign level averages tend to hide.

Why do customers receive more email than any single programme sends?

Because campaigns, automations, and separate product lines each mail on their own clock. The customer experiences the combined total, so fatigue management needs a house cap across all programmes and a collision report for customers receiving many sends in a short window.

How should a store respond when a customer's engagement starts falling?

Reduce that customer's cadence while they still engage, keeping the best performing programme and pausing others. Acting on the slope preserves subscribers; waiting for an unsubscribe and suppressing afterward also loses the deliverability cost of any spam reports along the way.

Should sale seasons override a customer's chosen email frequency?

No. A stated frequency preference is the customer's ceiling and honoring it during peak season is what keeps that cohort engaged and spam free over the long term. Overriding it converts a deliberate preference into evidence the setting is decorative.