What is RFM segmentation and how can a D2C brand use it?

RFM segmentation explained simply for Indian D2C brands: how to score recency, frequency and monetary value and turn segments into WhatsApp and email actions.

Anadoris Digital · · 4 min read

Stunning colorful flower bed
Photo: Henry Lorenzatto / Unsplash (CC0)

Sending the same campaign to every customer is the most common reason D2C messages feel like spam. RFM is a simple, well-established method to avoid that. It needs nothing more than order history, which you already have in your store, and it works in a spreadsheet before you ever need a CRM to automate it.

This guide explains how to calculate it, how to name the segments, and what to do with each one.

The three scores

  • Recency: days since the customer’s most recent order. Lower is better.
  • Frequency: number of orders in a chosen period, such as the last twelve months.
  • Monetary: total spend in the same period, after discounts and excluding returns.

Each customer gets a score, usually 1 to 5, on each dimension. The simplest method is to sort customers and split them into five equal groups for each measure. A customer who bought last week, buys often and spends well might score 5-5-5; someone who bought once, long ago, for a small amount scores 1-1-1.

Set recency around your repurchase cycle

Equal groups are a starting point, but recency is more useful when tied to how your products are actually used. A face serum that lasts about six weeks, a protein tub that lasts a month and a handbag bought once a year need very different recency bands. Look at the typical gap between first and second orders in your own data and set bands such as ‘within one cycle’, ‘one to two cycles’, ‘two to four cycles’ and ‘longer’.

Turn scores into named segments

Scores are hard to act on; names make them usable by the team. A practical small-brand set:

  • Champions: recent, frequent, high spend. Thank them, ask for reviews and referrals, give early access.
  • Loyal regulars: frequent buyers with moderate spend. Suggest bundles and subscriptions.
  • Promising new: recent first-time buyers. Focus on usage tips and the second order.
  • Needs attention: were regular, now slipping past their usual cycle. Send a reminder, not a discount.
  • At risk: high past value, long gap. Personal outreach or a considered win-back offer.
  • Lapsed one-timers: bought once, long ago. Low-frequency campaigns only, or let them rest.

Resist the urge to create every possible combination of scores. Five values on three measures give over a hundred cells, which nobody can act on. Six or seven named groups are plenty for a small team, and each one should have an owner and a default action written down next to it.

Using RFM on WhatsApp and email

Each segment should map to a channel and a frequency, not just a message. Champions may welcome a WhatsApp note about a launch; lapsed one-timers are better reached by occasional email. Keep promotional WhatsApp messages for customers who have opted in and who are likely to find them useful, because over-messaging on WhatsApp can lead to blocks and reports that affect your account.

A worked illustration

Imagine a haircare brand whose main shampoo typically lasts about five to six weeks. A customer who first bought four months ago, has ordered three times since, and spends more than most would sit in Champions. A customer who bought twice, but whose last order was ten weeks ago, has slipped past two cycles and belongs in Needs attention. A customer who bought a single trial kit seven months ago is a Lapsed one-timer.

Each gets a different next step. The Champion might receive a WhatsApp note about a new conditioner with an invitation to try it first. The slipping customer gets a simple reminder that their bottle has probably run out, with a reorder link and no discount. The lapsed trial buyer gets an occasional email about the full-size range. None of them receives a message that ignores what they have actually done.

The illustration uses round numbers for clarity; your own bands should come from your order data.

Common mistakes

  • Including cancelled and returned orders in monetary value.
  • Using one recency rule for products with very different usage cycles.
  • Recalculating once and never again; segments should refresh at least monthly.
  • Giving discounts to Champions who would have bought anyway.
  • Treating marketplace and wholesale buyers the same as D2C shoppers.

RFM uses past behaviour only. It will not tell you why someone stopped buying, so pair it with a short feedback question for at-risk customers.

Where to start

  • Export twelve months of orders with customer number, date and net value.
  • Calculate recency, frequency and monetary per customer in a sheet.
  • Assign the six named segments and check that the groups look sensible.
  • Pick one action for two segments, such as a second-order nudge and an at-risk call list.
  • Move the logic into your CRM so segments update automatically.

Anadoris Digital can set up RFM segments inside your CRM so they refresh on their own and feed your WhatsApp and email journeys.

Related service: CRM Solutions · More on this topic: CRM Solutions articles

Common questions

How many customers do I need for RFM to be useful?

There is no strict minimum, but with only a few hundred customers the five-way split becomes noisy. In that case use three bands per measure, or simple rules such as ‘bought in the last cycle’ and ‘bought more than once’, which give similar practical value.

Should marketplace orders count in RFM?

Only if you can reliably link them to the same customer. Most marketplace buyers cannot be identified, so including them adds noise. Run RFM on your own site and directly identified customers, and track marketplace performance separately.

How often should RFM segments be recalculated?

Monthly is a sensible minimum, and weekly or daily is better if your CRM can do it automatically. Recency changes every day, so stale segments mean you nudge people who have already reordered or miss those who just slipped.

Is RFM the same as customer lifetime value?

No. RFM describes recent behaviour to decide what to do next. Lifetime value estimates the total value of a customer relationship over time. They use the same order data and work well together, but answer different questions.

Your next chapter

Let’s build what comes next.

A new website. A better system. A simpler way to work. Tell us where you want to begin.

Discuss your project