Advanced Email List Segmentation: Moving Beyond Basic Demographics

Most e-commerce brands segment their email list the same way they did a decade ago. They split by gender. They filter by geography. They create a “VIP” segment for customers who have spent above a certain threshold. Then they pat themselves on the back for doing personalisation.

The problem is that demographic segmentation tells you almost nothing about what someone is likely to buy next, how likely they are to churn, or how they prefer to engage with your brand. Demographics describe people. Behaviour predicts people. And in email marketing, prediction is where the revenue lives.

This guide covers the full progression from basic demographic segmentation to behavioural segmentation to RFM modelling to predictive segments — and explains how to implement each layer without creating a segmentation system so complex that it becomes unmanageable.

The Limitations of Demographic-Only Segmentation

Demographic segmentation — age, gender, location, income bracket — is a starting point, not a strategy. The core problem is that two customers with identical demographics can have entirely different buying patterns, engagement preferences, and lifetime values.

Consider two women in their 30s who both purchased a skincare product from your brand six months ago. One has purchased four times since, opens every email you send, and spends an average of $140 per order. The other has not purchased again, has not opened an email in three months, and may have only bought once because of a promotion.

Demographic segmentation treats these two customers identically. Behavioural segmentation treats them as what they actually are: one is your most valuable customer profile, and one is a candidate for a win-back campaign. The revenue implications of treating them differently are significant.

Behavioural Segmentation — The Foundation

Behavioural segmentation uses actions (or the absence of actions) to define audience groups. The three most valuable data sources for behavioural email segmentation are purchase history, browse behaviour, and email engagement.

Purchase History Segments

Purchase history is the richest signal you have. At minimum, you should be building segments around:

First-time buyers — customers who have purchased exactly once and need nurturing toward a second purchase
Repeat buyers — customers who have purchased two or more times (the threshold for demonstrating genuine loyalty intent)
Category buyers — customers who have only ever bought within a specific product category
High-value buyers — customers above your average order value threshold
Lapsed buyers — customers who purchased but have not returned within a defined window (typically 90-180 days depending on your category)

Each of these segments warrants different messaging, different offer strategy, and different email frequency.

Browse Behaviour Segments

If your ESP is connected to your e-commerce platform (Klaviyo + Shopify being the most common combination), you have access to on-site browse events that let you segment by what someone looked at even if they did not purchase.

A subscriber who has viewed your highest-margin product category three times in two weeks without buying is not the same as a subscriber who has never visited your site. One has demonstrated clear intent. The other requires awareness-building first. Sending both contacts the same campaign content is a missed opportunity.

Email Engagement Segments

The simplest engagement-based segmentation approach uses three tiers: highly engaged (opened or clicked in the last 30 days), moderately engaged (opened or clicked in the last 90 days), and unengaged (no activity in 90+ days).

Highly engaged contacts can tolerate higher send frequency and will perform well on promotional campaigns. Moderately engaged contacts benefit from content-first emails that remind them why they subscribed. Unengaged contacts should receive reduced send frequency or be routed into a re-engagement flow before returning to normal campaigns.

RFM Modelling — Turning Data Into Customer Value Scores

RFM stands for Recency, Frequency, and Monetary value. It is the most widely used quantitative framework for customer segmentation in e-commerce, and when applied to email marketing, it creates a remarkably clear picture of customer value and next-best-action.

How RFM Works

Each customer in your list is scored on three dimensions:

Recency — How recently did they make their last purchase? More recent purchases score higher.
Frequency — How many total purchases have they made? More purchases score higher.
Monetary — How much have they spent in total? Higher spend scores higher.

Each dimension is scored on a scale (typically 1-5 or 1-3), and the combination of scores creates a customer archetype.

The Most Valuable RFM Segments for Email

A customer scoring high across all three dimensions (recent purchase, many purchases, high total spend) is your Champion. These are your brand advocates, most likely to respond positively to VIP offers, early access to new products, and loyalty programme content.

A customer who scored high historically but has not purchased recently (high F and M, low R) is a Slipping Champion — someone who used to be deeply loyal but is at risk of churning. These customers should receive personalised win-back content that acknowledges their history with your brand.

A customer with a recent first purchase but low frequency and moderate spend is a Promising New Customer — someone worth investing in to drive that second purchase that dramatically increases predicted lifetime value.

In Klaviyo, you can build RFM segments dynamically using the predicted CLV, days since last order, number of orders, and total spend properties. The segments update automatically as customer behaviour changes, meaning your email programme is always working with current data.

Predictive Segments in Klaviyo

Klaviyo’s predictive analytics features take segmentation a step further by using machine learning to forecast future behaviour — not just describe past behaviour.

Predicted Customer Lifetime Value

Klaviyo calculates a predicted CLV for every profile based on purchase history patterns. This lets you build a high-CLV segment — customers predicted to be your most valuable over the next 12 months — and treat them with appropriate investment. Higher-value welcome series, exclusive offers, priority access to new launches, and proactive retention content.

Churn Risk

Klaviyo’s churn risk prediction identifies customers who are statistically likely to stop purchasing based on patterns in their engagement and purchase history. These customers should be in an active retention flow — not waiting for you to notice they have gone quiet.

The combination of “high CLV, high churn risk” is the most urgent segment in any e-commerce email programme. These are your most valuable customers who are about to leave. Identifying and acting on this segment proactively is worth significant revenue.

Next Purchase Probability

This prediction identifies customers who are statistically likely to purchase soon. This segment is valuable for timing your highest-performing campaigns — sending your best offer or your most exciting product launch to people who are already primed to buy maximises the conversion rate of your best content.

Managing Complexity Without Losing Control

One of the most common mistakes in advanced segmentation is creating so many segments that the system becomes unmanageable. When every campaign requires manually checking a matrix of 30 different segments, most teams either make mistakes or abandon the framework entirely and revert to sending to everyone.

The solution is a layered architecture that separates your always-on automated segments (which run in flows and update dynamically) from your campaign-specific targeting logic.

Your dynamic segments — RFM tiers, engagement tiers, CLV bands, churn risk — should be set up once and maintained by the ESP. Your campaigns use these segments as targeting conditions without requiring you to rebuild the logic each time.

A practical implementation looks like this:

Set up 5-7 master segments that define your core audience groups
Apply engagement tier targeting as a universal filter on all campaigns (so unengaged contacts automatically receive less frequent sends)
Use purchase history and RFM data to vary offer and content within campaigns rather than creating entirely separate campaign versions for each segment

This approach captures 80% of the revenue upside from advanced segmentation with a fraction of the operational complexity.

The Revenue Impact of Behavioural Segmentation

The shift from demographic to behavioural segmentation produces measurable revenue changes, typically visible within 60-90 days of implementation.

The most consistent improvements come from three areas. First, win-back campaigns for lapsed customers become significantly more effective when they are triggered based on actual purchase gap rather than a calendar date. Second, post-purchase flows convert more second purchases when the content is based on what the customer actually bought rather than your most popular products across the board. Third, promotional campaigns generate higher revenue per send when the offer is matched to the customer’s spend tier rather than being one-size-fits-all.

Brands that move from demographic-only segmentation to a full behavioural and RFM framework typically see email revenue increase by 20-40% from the same list size — not because they are sending more emails, but because each email is doing more work.

Segmentation architecture is one of the first things Excelohunt builds when working with a new client. We audit existing segments, build out the behavioural and RFM framework in Klaviyo, and connect segmentation logic to both automated flows and ongoing campaigns so the entire programme gets smarter over time.

Looking to implement these strategies with expert support?

List Management — learn how we implement this for clients Book a free strategy call with Excelohunt →