A customer who bought yesterday for the first time needs a different conversation from a regular whose visits have slowed. Purchase history can help you notice that difference before you send both people the same offer.

RFM is a way to look at three signals: recency, frequency and monetary value. In plain language, that means how recently a customer bought, how often they bought and how much they spent. It gives you questions to investigate, rather than a promise about what someone will do next.

Start with three questions

  • How recently did they buy? Look at the date of the last recorded purchase. Compare it with the buying pattern that makes sense for your business.
  • How often did they buy? Count purchases over a chosen period. A customer's first order and their fifth order tell different stories.
  • How much did they spend? Review spending over the period you have defined. Decide how refunds and discounts affect the amount, and use that definition consistently.

Check that purchases belong to the right customer and that the records cover the sales channels you want to understand. An incomplete history can make a regular look like a first-time buyer. Spending also differs from profit: a large order can carry a large discount or cost.

Make the groups fit your business

There is no single number of days that makes every customer inactive. Someone may buy coffee several times a week and furniture only occasionally. Seasonality, the product and a customer's usual routine all affect what a gap between purchases means.

A customer group, sometimes called a segment, brings together people who meet chosen conditions. For example, you might look at first-time buyers who have not made a second purchase, or regulars whose visits have slowed. Choose the period and conditions deliberately, then inspect a few purchase histories to check whether the group makes sense.

A score or rank can help summarize records, but always ask how it was calculated. A customer can rank above others and still have gone longer than usual without buying. A rank within your list and a rule based on elapsed time answer different questions.

Give each group a useful next step

Imagine a homeware shop reviewing two groups. This is a hypothetical example, not a Perksmith customer result or a recommended spending rule.

A first-time buyer might benefit from information about using their purchase, or an eligible offer for a second purchase. A regular whose visits have slowed might need a different invitation. Neither group automatically needs a bigger discount. Start with the customer's situation, the benefit you can afford and the terms you can explain.

Customer insights can help you explore purchase history and the customer groups available in your plan. An automated follow-up sequence connects a starting moment, timing and action—for example, a recorded first purchase followed by a welcome message. Confirm the information, tools and rules available for your setup before using it.

Only send messages to customers with permission for that kind of communication on the channel you choose. Check offer eligibility, expiry and budget as well. Customer messaging explains the supported channels and permission requirements.

Check what changed, including the cost

A customer buying after a message does not prove the message caused the purchase. They might have bought anyway. Review purchases, refunds, message charges and reward costs together.

Where a suitable comparison is possible, compare similar customers who receive the campaign reward with a group that does not receive it over the same period. Consider the amount of data and other changes affecting sales before drawing a conclusion. Read the guide to measuring extra business for a worked example and its limitations.

FAQ

What does RFM stand for? Recency, frequency and monetary value: how recently a customer bought, how often they bought and how much they spent. These signals describe recorded purchase history.

Can I use the same customer-group rules for every business? No. A café and a furniture shop have different buying patterns. Choose a period and rules that fit your business, then check the underlying purchases before acting.

Does an RFM group tell me who will buy next? No. Purchase history can help you choose a relevant follow-up, but it does not reveal a customer's intentions or prove that a reward will cause another purchase.

Request a demo