To segment an email list, you split your subscribers into smaller groups using data you already hold, such as sign-up source, purchase history or engagement, and then send each group something that fits what it has told you. The logic is simple. The difficulty is having clean data and enough contacts in each group to make the extra work worthwhile. Start with two or three segments, not twenty.
What a segment actually is
Mailchimp describes segments as filtered groups of similar contacts within an audience, built from conditions (the rules that decide who is included) and and/or logic (whether a contact must meet all conditions or just one) (Mailchimp). Other email platforms use different names, but the mechanics are much the same: a saved rule that re-evaluates as contact data changes.
The same help page lists typical condition types: email engagement such as opening your last email, purchase behaviour, sign-up date, and location. Which conditions are available, and how many you can combine, depends on the platform and plan. Mailchimp, for example, limits regular segments to 10 conditions on its free and Essentials plans and allows nested logic only in advanced segments.
Decide what data you really have
A segment can only be as good as the field behind it. Before designing anything, list what is captured and how reliably.
| Data source | Typical field | Reliability to check |
|---|---|---|
| Sign-up form | Interest or topic chosen, sign-up source | Is the field optional? Many people skip it |
| Shop or booking system | First purchase date, product category, last order | Is it synced automatically or imported by hand? |
| Email platform | Opens, clicks, last engagement | Open data can be distorted by mail privacy features, so lean on clicks |
| Website | Pages viewed, form used | Only available if tracking is connected and consented |
| Manual entry | Location, customer type | Prone to gaps and inconsistent spelling |
If a field is empty for most contacts, a segment built on it will quietly exclude them. Fix the capture point first, then segment.
Segments worth building first
These work for many small organisations because they need little data:
- New subscribers. Contacts who joined recently get a short welcome sequence. Mailchimp’s pre-built New Subscribers segment, for instance, covers people who joined in the past seven days.
- Customers and non-customers. Someone who has bought needs different messages from someone who has only subscribed.
- Engaged and inactive. People who clicked recently versus those who have not engaged in a long time. Define “long” for your own sending frequency rather than copying a benchmark.
- Interest or product category. Based on a form choice or purchase category.
- Location. Useful for events, store openings or shipping cut-offs, less useful otherwise.
Resist demographic segments unless you have a reason. Age or gender assumptions often produce worse copy than a clear interest signal.
A simple build process
- Write the question each segment answers, for example “Who has bought once but not twice?”
- Check the field exists and is filled for enough contacts.
- Build the rule, then preview the member count. A segment of a handful of people is rarely worth a separate campaign.
- Write one message variant per segment, changing the offer or example, not just the greeting.
- Compare results by segment over several sends before drawing conclusions. Small groups produce noisy numbers.
- Review rules quarterly, since products, forms and tracking change.
Consent and Australian rules
Segmenting does not create consent. Under Australian rules for commercial electronic messages, summarised by business.gov.au, you need permission before sending: either express consent, where the person knows and accepts they will receive marketing, or inferred consent, where it is reasonable to believe they would expect it because they gave you their address. The page advises recording who gave permission, when and how, and says unsubscribe options must be clear, free, honoured within 5 working days and kept working for at least 30 days after sending (business.gov.au).
Two practical consequences: do not import purchased or scraped lists into a segment, and make sure an unsubscribe applies across all segments you send to. This is general information, not legal advice.
Limitations and trade-offs
More segments mean more content to produce and more places for errors. A mistaken rule can send a customer a “welcome, buy now” message. Keep a test contact in each segment, and prefer fewer, well-defined groups. Segmentation also pairs naturally with how you think about the customer journey, which is covered in the marketing funnel explained, and with a sustainable publishing rhythm, discussed in how to plan a content calendar. If your list is built partly from site visitors, retargeting visitors who did not convert covers the paid side of the same idea.
Where Radiance Impact fits
Radiance Impact’s Growth division works on strategy, campaigns and search visibility, and the Technology division builds websites and workflow automation that can feed clean data into your email tools. If you want help working out which segments suit your audience, see /growth or get in touch through /contact.
Sources
- Getting Started with Segments, Mailchimp
- Promoting your business by email or text messages, business.gov.au
