Customer segmentation for small businesses: 8 segments to start with
Eight practical customer segments for a UK café, shop or salon loyalty programme, from lapsed regulars to nearly-next-tier members, with the message to send each one.
Key takeaways
- Segment by behaviour (visits, spend, recency) before demographics.
- Start with a few segments that each have an obvious message.
- Dynamic segments update themselves as customers buy.
- Compare each segment’s return rate against a group that didn’t receive the message.
Why segment at all?
Sending every offer to every customer wastes rewards on people who would have come anyway and annoys people the offer doesn’t suit. Segmentation means choosing who receives each message based on what they actually do. Even a small business can do this with a loyalty programme, because every purchase is linked to a member.
In Unity, segments are built from rules, such as days since last visit, total spend or loyalty tier, combined with AND or OR. They update automatically as customers buy.
Eight segments worth creating
Each of these has a clear purpose and an obvious message. Adjust the numbers to your business’s normal visit cycle.
| Segment | Rule idea | Message |
|---|---|---|
| New members | Joined in the last 14 days | Welcome and how to earn; nudge a second visit |
| Lapsed regulars | Visited 5+ times, but not in the last 30 days | A personal “we miss you” with a reason to return |
| Nearly next tier | Spend just below a tier threshold | How close they are and what the next tier gives |
| Big balances | Points balance above a set amount | Remind them their reward is waiting |
| Top customers | Highest tier | Thanks, early access, an invitation |
| Referrers | Invited at least one friend | Thank you; remind them of the referral reward |
| Joined by referral | Has a referrer | Help them make it a habit |
| Location regulars | Main location is a specific site | News and offers for that site only |
Behaviour beats demographics
Age and gender filters exist, but they are rarely the best place to start. What a customer does, such as how recently and how often they visit and how much they spend, predicts what they will respond to far better than who they are. Use demographic filters only when the offer genuinely depends on them, and be careful not to make assumptions.
Keep a comparison group
To know whether a campaign worked, compare the segment that received it with similar customers who didn’t. The simplest way is to hold back a small random part of the segment. If both groups return at the same rate, the message didn’t change anything, however good the open rate looked.
Read more in measuring loyalty programme ROI.
How to build a segment from rules
A segment is just a set of conditions. “Lapsed regulars” might be: visit frequency of five or more in the last six months AND last visit more than 30 days ago. “Nearly next tier” might be: total spending between £120 and £149 when the next tier starts at £150. In Unity, you choose a rule, an operator such as “greater than” or “between”, and a value, then combine rules with AND or OR.
Name segments after their purpose, not their rules, so that anyone in the team understands them: “Lapsed regulars – 30 days” is clearer than “VF≥5 & LV>30”. Check the size of each segment before you send. A segment of six people isn’t worth a campaign; a segment of half your members probably isn’t targeted enough.
Put segments to work
Start with the lapsed regulars segment: it usually has the clearest return. Our win-back guide walks through the campaign step by step. Segmentation is part of Unity’s Grow and Scale plans; see the customer notifications page for the full list of rules.
For current guidance on customer information and marketing preferences, see the ICO’s direct marketing resources and our Privacy & GDPR policy.
