Customer segmentation: how to split your base into groups and offer each its own approach
Sales7 min read
Customer segmentation in plain words: why split your base, the main criteria and RFM analysis, example segments and how to turn them into campaigns and tasks.

Customer segmentation is dividing a client base into groups by shared characteristics, so you can offer each group what suits it and spend managers' attention where it pays off most. The same mailing to the whole base annoys half the recipients. A personal offer for a segment works many times better.
In this article: Segmentation criteria · RFM analysis · Example segments for different industries · How to turn segments into actions · RFM segmentation example: a nail studio · Segmentation in B2B · Common mistakes · Walkthrough: segments and actions for a base of 1,200 clients · How many segments are enough · 30-day plan
Segmentation in a minute
| Question | Answer |
|---|---|
| What is it? | Dividing the base into groups |
| Why? | Own offers and priorities for each group |
| Main criteria | Recency, frequency, amount, interests, source |
| Popular method | RFM analysis |
| How many segments? | 4–8 working segments |
| What next? | An action for each segment |
Segmentation criteria
| Criterion | Example segments |
|---|---|
| Value | Key, mid-level, one-time |
| Purchase recency | Bought this month, six months ago, over a year ago |
| Frequency | Regular, occasional, one-time |
| Interests | New builds and resale; haircut and coloring |
| Source | Listing platforms, website, referral |
| Client type | Individuals, companies |
| Geography | City, district, region |
| Stage | New lead, client, lapsed client |
RFM analysis
RFM divides clients by three parameters:
| Letter | Parameter | Question |
|---|---|---|
| R — Recency | Recency | How long ago was the last purchase? |
| F — Frequency | Frequency | How many times have they bought? |
| M — Monetary | Money | How much have they paid in total? |
Each parameter is scored from 1 to 3, which gives these groups:
| Segment | R | F | M | What to do |
|---|---|---|---|---|
| Best clients | 3 | 3 | 3 | Personal attention, privileges |
| Loyal | 2–3 | 3 | 2 | Loyalty program, upsells |
| New | 3 | 1 | 1–2 | Second purchase, introduction to the range |
| Slipping away | 1–2 | 2–3 | 2–3 | Win back: personal message, offer |
| Lost | 1 | 1 | 1 | Occasional reminders or exclude |
Example segments for different industries
| Industry | Segments and actions |
|---|---|
| Beauty salon | Haven't visited for over 2 months → reminder and booking; regulars → membership |
| Clinic | Missed a follow-up visit → call; families → family program |
| Real estate | Bought 3–5 years ago → offer to sell or trade up |
| Education | Finished a course → next level; dropped out → talk about the reasons |
| Wholesale | Reduced purchases → find out why; new dealers → training |
How to turn segments into actions
- Choose 4–8 segments. More than that and the team won't work with them.
- A goal for each. Second purchase, win-back, larger order value.
- An action for each. Mailing, call, personal offer.
- Assign owners. Key segments to a personal manager, mass segments by mailing.
- Measure the result. How many clients in the segment responded and bought.
For how to retain clients, see the article on client retention. For how to label your base with tags, see the article on tags in a CRM.
RFM segmentation example: a nail studio
A nail studio with a base of 1,200 clients split it in a single evening:
| Segment | Clients | Action | Result after a month |
|---|---|---|---|
| Come every 3–4 weeks | 310 | Membership with a discount | 74 memberships |
| Haven't been for 2–3 months | 280 | Personal message with a convenient time | 96 bookings |
| Came once | 420 | Offer of a second treatment | 61 bookings |
| Haven't been for over six months | 190 | One message, then a pause | 18 bookings |
The figures are illustrative, but they show the main point: a personal message for each group works better than one mailing to the whole base.
Segmentation in B2B
For corporate clients, additional criteria apply:
- company size and number of branches;
- the client's industry;
- purchase volume and its trend over the year;
- who makes the decision;
- contract end date.
Key B2B clients need a separate approach — more in the article on key clients.
Common mistakes
- Segments for the sake of segments. The base was split, but nothing is done with it.
- Too many segments. Each of 20 segments has 15 clients.
- No data. Purchases, source and interests are not recorded.
- Segments are not updated. The client became a regular long ago but is still listed as "new".
- The same message for every segment. The point of segmentation is lost.

Walkthrough: segments and actions for a base of 1,200 clients
An illustrative example. After RFM analysis, the base is divided into five segments:
| Segment | Clients | Share of base | Share of revenue | Action |
|---|---|---|---|---|
| Champions | 90 | 7.5% | 38% | Personal manager, special terms |
| Loyal | 210 | 17.5% | 31% | Loyalty program, upsells |
| Growing | 300 | 25% | 17% | Offers matched to needs, reminders |
| At risk | 250 | 20.8% | 9% | Reactivation: personal contact, offer |
| Dormant | 350 | 29.2% | 5% | Automatic mailing, base cleanup |
| Total | 1,200 | 100% | 100% |
You can see that champions and loyal clients are only 25% of clients but bring in about 69% of revenue. This is where personal attention should go. There are many dormant clients, but they bring little revenue: automated scenarios suit them.
How many segments are enough
Start with 4–6 segments. With more, the team stops telling them apart and remembering the action for each. A good segment is one for which you have a separate action.
30-day plan
| Week | What to do |
|---|---|
| 1 | Export clients: last purchase date, number of purchases, amount |
| 2 | Assign scores for recency, frequency and amount, and combine them into segments |
| 3 | Define an action and an owner for each segment |
| 4 | Launch the first campaigns, measure the response and adjust |
How it works in Mestra
- Client filters. The client list can be filtered by fields, tags, source and responsible person.
- Payment history. The card shows all purchases and amounts — the basis for RFM.
- Client source. You can see where each client came from.
- Mailings. You can send a message to selected clients in messengers.
- Win-back reminders. Birthdays, contract end, repeat visit — Mestra prepares the text.
- Churn prediction. AI flags clients who may leave.
Frequently asked questions
What is customer segmentation?
Dividing a client base into groups by shared characteristics, so you can offer each group what suits it and allocate attention properly.
Which criteria should I use for segmentation?
Recency and frequency of purchases, amount spent, interests, source, client type and stage of the relationship with the company.
What is RFM analysis?
A segmentation method based on the recency of the last purchase, the frequency of purchases and the amount the client has paid.
How many segments do I need?
Usually 4–8 working segments, each with a specific action.
How often should I update segments?
Once a month or quarter: clients move from group to group.
Summary
Segmentation turns a client base from a list of contacts into a set of groups with clear actions: win back those who are slipping away, retain the best, turn new clients into regulars. What matters is a small number of segments and an action for each.
In Mestra, clients are filtered by fields, tags and sources, payment history is visible in the card, and AI flags clients who may leave. Try Mestra free for 15 days at mestra.chat.


