In this article we will cover:
- What RFM Segmentation is and its importance in marketing
- How to create the RFM quantiles in BigQuery
- How and what RFM segments you can create from the RFM quantiles
- Considerations for your own RFM model
*Note — all data used in this article is fictional and generated by myself in BigQuery.
Let’s start with the basics, what even is an RFM Model?
An RFM (Recency, Frequency, Monetary) model is a customer segmentation technique that uses past purchase behaviour to divide customers into distinct groups.
- Recency measures how recently a customer made a purchase
- Frequency assesses how often a customer transacts
- Monetary looks at how much a customer spends
These three segments alone can improve your understanding of your customer base, but you can combine these together to form segments. These segments help you identify which customers are your best customers, those who are slipping away, or others who are engaged but spend less.
Time for an example, and to keep it simple we’ll just focus on the frequency segment for the moment.
Imagine you’re in charge of a large community garden, where hundreds of gardeners come to plant and care for the plants. To understand which gardeners are your most frequent, you decide to log when each gardener visits.
With our new data log of say 100 gardeners, we can count how many times each visited the garden, then split them into five equal parts known as quintiles.
Just so it seems a little more human, we could give each quintile a name:
- First Quintile — The Rare