The difference between an open rate of 18% and one of 35% in the same sector with the same contact base is usually just one thing: segmentation. The mass newsletter that reaches all contacts in the same way is the most inefficient format in email marketing. Advanced segmentation is what transforms email from a noise channel into one of the highest-ROI channels.
The four levels of segmentation
Level 1 — Demographic and firmographic segmentation: sector, company size, job title, location. It’s the starting point. A business management company has clients in hospitality, industry and services. Each segment has different problems, and the same email for everyone has low relevance for everyone.
Level 2 — Behavioural segmentation: based on what the contact has done. Has opened the last three emails vs has not opened any in ninety days. Has visited the pricing page vs only the home page. Has downloaded resource A vs resource B. This segmentation allows far more relevant messages because it is based on real intent signals.
Level 3 — Life-cycle segmentation: new lead, lead in nurturing, active customer, customer at risk of churn, lost customer. Each stage requires a different approach. Sending a new-customer acquisition offer to existing customers is a mistake that damages the relationship. Sending nurturing content to someone who is already a customer ignores the opportunity to upsell.
Level 4 — Predictive segmentation: based on machine learning models that predict future behaviour. Who is most likely to make a purchase in the next thirty days, who is at greatest risk of unsubscribing, which product is most likely to interest whom. This is the most powerful form of segmentation and the one that requires the most data and the greatest technical sophistication.
The RFM model applied to email
RFM (Recency, Frequency, Monetary) is a classic customer segmentation model that applies perfectly to email marketing.
Recency: when the last interaction with emails (open, click) or with the business (last purchase, last contact) took place. Contacts with high recency are more likely to respond positively to emails.
Frequency: how often the contact interacts with emails or makes a purchase. High-frequency contacts are the most engaged and the most valuable.
Monetary: how much the contact has spent or what their estimated spending potential is. High monetary-value contacts deserve more personalised communications and greater contact frequency.
The segments that result from the RFM model have very different characteristics and needs. Champions (high R, high F, high M) need recognition and early access to new products. At-risk customers (historically high F and M but low recent R) need a reactivation campaign with a specific offer. New customers (high R, low F) need an onboarding sequence that builds the relationship.
Automating dynamic segmentation
Manual segmentation — creating static lists and assigning contacts by hand — does not scale. Automated dynamic segmentation updates segments in real time based on behaviour: when a contact clicks a link on a specific topic, they automatically enter the interest segment for that topic and begin receiving the corresponding nurturing sequence.
The platforms that support dynamic segmentation natively: ActiveCampaign (the most complete for SMEs), Klaviyo (specialised in e-commerce), HubSpot (integrated with the CRM). Mailchimp on its higher-tier plans also offers basic behavioural segmentation.