Micro Segments Feel Precise and Behave Like Spam: The Fatigue Maths of Slicing Too Thin
Segmentation has a seductive direction of travel. If a broad campaign works, a version for new parents works better, and a version for new parents who bought in the last sixty days and live in two cities should work better still. Each step is defensible in the meeting where it is proposed. The failure only appears in aggregate, in the customer's inbox, where the same person now qualifies for nine micro segments, receives nine lovingly tailored campaigns a fortnight, and unsubscribes from a brand that believes it has been communicating with unusual precision.
Count sends per person, not campaigns per segment
The unit of fatigue is the person, and no segment plan that counts campaigns instead of per person totals can see the problem. Build the view that matters: distribution of marketing contacts per customer per month across every segment and trigger, with the overlap shown. Most teams find a familiar shape, a large group sensibly contacted, and a valuable minority, often the most engaged customers, sitting in the tail receiving three times the intended volume because every segment wants them. They were the best audience for each campaign individually and the worst treated audience collectively.
Thin segments cannot be measured, so they cannot be managed
A segment of four hundred people split across variants produces differences nobody can distinguish from noise, which means creative and offer decisions for micro segments are made on noise and defended with confidence. Below a workable size, the honest choices are to merge the segment upward until its results can be read, or to serve it with a rule based personalisation inside a larger campaign, product and content selection varying while the campaign, and its measurement, stay whole. Personalisation inside a campaign scales down gracefully. Campaign proliferation does not.
Eligibility rules need a traffic controller
Where many segments persist, someone or something has to arbitrate: a contact policy at the profile level that caps total volume, ranks competing campaigns by expected value per send, and enforces cooldowns after high intent triggers so a browse abandon email and a segment campaign do not land an hour apart. That arbitration is unglamorous infrastructure and it is the difference between a segmentation strategy and a pile of audiences. Assign the policy an owner with authority to say a campaign waits. A cap that any team can override for an important launch is a suggestion, and inboxes are not governed by suggestions.
Precision should be spent where difference is real
Some differences justify separate treatment: lifecycle stage, because a new customer and a five year customer genuinely need different things; language and region, because the message must change; a stated preference, because the customer asked. Many celebrated micro segments differ only in a behavioural detail the creative does not actually use, producing identical emails with different audience names. Audit the calendar for campaigns whose content would be unchanged if their segments merged. Merge those, bank the production time, and spend the precision budget on the segments where the message, offer, or timing truly diverges.
Triggers count, and they are the worst offenders
Teams auditing fatigue usually count campaigns and forget triggers, because triggers belong to the lifecycle team and live in a different tool. The customer counts everything. Browse and cart abandon flows, back in stock alerts, price drop notices, review requests, and loyalty updates can add several contacts a week to an active customer on top of campaign volume, and they cluster exactly when the customer is most engaged and therefore most likely to be in every campaign segment too. Put triggers under the same contact policy as campaigns, with priority given to the message closest to the customer's own action, and let the campaign wait.
Keep reading
Related guides on this site: RFM Segments: Setting Thresholds From Your Own Data, Not a Template Churn Risk Scores Are Useless Without an Action Window Practical answers about personalization.
Frequently asked questions
What is campaign fatigue?
The decline in engagement and rise in unsubscribes when a customer receives too many marketing contacts, usually caused by aggregate volume across segments rather than any single campaign.
How small can a segment be before results are unreadable?
It depends on conversion rate and send volume, but segments in the low hundreds per variant rarely produce differences distinguishable from chance. Merge upward or personalise inside a larger campaign.
What is a contact policy?
A profile level set of rules capping total marketing contacts, ranking competing campaigns, and enforcing cooldowns, applied across all segments and triggers.