Successfully reported this slideshow.

RFM Segmentation

120

Share

Loading in …3
×
1 of 25
1 of 25

RFM Segmentation

120

Share

Download to read offline

RFM Segmentation is the easiest and most frequently used form of database segmentation. It is based on three key metrics: Recency, Frequency and Monetary Value of customer activity. RFM is often used with transactional history in e-commerce, but can also work for Social Media interactions, online gaming or discussion boards. Based on calculated segments a marketer can prepare cross-sell, up-sell, retention and reactivation capampaigns. This deck provides a simple introduction to the RFM Segmentation methodology.

RFM Segmentation is the easiest and most frequently used form of database segmentation. It is based on three key metrics: Recency, Frequency and Monetary Value of customer activity. RFM is often used with transactional history in e-commerce, but can also work for Social Media interactions, online gaming or discussion boards. Based on calculated segments a marketer can prepare cross-sell, up-sell, retention and reactivation capampaigns. This deck provides a simple introduction to the RFM Segmentation methodology.

More Related Content

Related Books

Free with a 14 day trial from Scribd

See all

RFM Segmentation

  1. 1. RFM CUSTOMER SEGMENTATION
  2. 2. HELLO! Kamil Bartocha Follow me at @WhiteRavenPL or drop me an email: kamil.bartocha@marketingdistillery.com
  3. 3. WHAT IS RFM? RFM stands for Recency, Frequency and Monetary ⊡ It is the easiest form of customer database segmentation ⊡ Often used for reactivation campaigns, high valued customer programs, combating churn etc.
  4. 4. RFM IS BASED ON USER ACTIVITY DATA Anything from actual orders, website visits, app launches etc.
  5. 5. TRANSACTION EXAMPLES E-COMMERCE Orders, visits SOCIAL MEDIA Sharing, liking, engagement GAMING In-app purchases, levels played Purchase history Website visits Social engagement You can use more than one RFM segmentation RFM Segmentation can be applied to activity-related data that has measurable value and is repeatable DISCUSION BOARDS Posting, up-votes LEAD MANAGEMENT Engagement, value
  6. 6. RFM METRICS RECENCY The freshness of customer activity. e.g. time since last activity FREQUENCY The frequency of customer transactions. e.g. the total number of recorded transactions MONETARY The willingness to spend. e.g. the total transaction value RFM Metrics:
  7. 7. TOTALS ● R: Time since last transaction ● F: Total number of transactions ● M: Total transactions value RFM METRICS AVERAGES ● R: Time since last transaction ● F: Average time between transactions ● M: Average transaction value RFM Metrics can have multiple definitions Transactions can only increase customer value in the segmentation Easy to explain Transactions can both increase and decrease customer value in the segmentation Complicates campaigning
  8. 8. RFM TABLE Step 1: Calculate the RFM metrics for each customer Customer Recency Frequency Monetary A 53 days 3 transactions $230 B 120 days 10 transactions $900 C 10 days 1 transaction $20 This is called the RFM Table … and can be easily computed in SQL, R, Spark etc.
  9. 9. RFM SEGMENTATION Step 2: Find the distribution for each metric... ...and define the segmentation... 0 20 40 60 80 90 100 days $800 0 1 2 3 4 5 6 orders $0 $50 $100 $200 $400 $1200 value
  10. 10. RFM SEGMENTATION … by splitting values into bins. 0 20 40 60 80 90 100 days $800 0 1 2 3 4 5 6 orders $0 $50 $100 $200 $400 $1200 value Segment 1 Segment 2 Segment 3 Segment 1 Segment 2 Segment 3 Segment 1 Segment 2 Segment 3
  11. 11. RFM SEGMENTATION The easiest way to split metrics into segments is by using quantiles: ⊡ This gives a starting point for detailed analysis ⊡ 4 segments are easy to grasp and action Segment 1 Segment 2 Segment 3 Segment 4 25th quantile median 75th quantile There are much better ways to choose segmentation points!
  12. 12. ⊡ Use Survival Analysis to cut Recency segments at 25% and 50% of customer churn probability. ⊡ Identify high-valued customers by splitting out the top 10% in Frequency and Monetary. ⊡ Separate one-time buyers from customers with repeat purchase.
  13. 13. RFM TABLE Step 3: Add segment numbers to the RFM Table Customer Recency Frequency Monetary R F M A 53 days 3 tran. $230 2 2 2 B 120 days 10 tran. $900 3 3 2 C 10 days 1 tran. $20 1 1 1 This is called a Segmented RFM Table
  14. 14. RFM SEGMENTATION RFM Segments split your customer base into an imaginary 3D cube It is difficult to visualize! R F M R: Segment 1 F: Segment 2 M: Segment 2
  15. 15. Use a stacked contingency table to count customers in each segment and compute summary statistics STACKED TABLES M R F 1 2 3 4 1 1 2 3 4 2 1 2 3 4 3 1 2 3 4
  16. 16. Use the Recency segmentation to identify customers at risk of churn. This works especially well if you use Survival Analysis for Recency segmentation. RECENCY SEGMENTATION M R F 1 2 3 4 1 1 2 3 4 2 1 2 3 4 3 1 2 3 4 ACTIVE AT RISK CHURNED
  17. 17. RECENCY SEGMENTATION M R F 1 2 3 4 1 1 2 3 4 2 1 2 3 4 3 1 2 3 4 ACTIVE AT RISK CHURNED X/UP-SELL, PROMOTIONAL RETENTION CAMPAIGN REACTIVATION CAMPAIGN Also, use Recency for campaigning
  18. 18. Use the Frequency & Monetary segmentation to estimate customer value. Typical segment names: Premium, Gold, Silver etc. VALUE SEGMENTATION M R F 1 2 3 4 1 1 SILVER SILVER 2 SILVER SILVER GOLD 3 SILVER SILVER GOLD GOLD 4 SILVER GOLD GOLD PREMIUM
  19. 19. VALUE TIERS Each transaction will move customers through Recency and Value tiers. Prospects Freshers Standard Silver Gold Premium 0 Transactions 1 Transaction Freshers Standard Silver Gold Premium Freshers Standard Silver Gold Premium ACTIVE AT RISK CHURNED 1 Transaction 1 Transaction With RFM Metrics based on sums of events, the move can only be towards higher valued segments.
  20. 20. RFM VERIFICATION You can easily verify the power of the RFM segmentation. 1. Split your data into two parts Training Period Test Period 12 months
  21. 21. RFM VERIFICATION You can easily verify the power of the RFM segmentation. 2. Assign customers to RFM Segments using only data from the Training Period R F M Training Period Test Period
  22. 22. RFM VERIFICATION You can easily verify the power of the RFM segmentation. 3. Calculate the average value of customers in each RFM Segment over the Test Period R F M Training Period Test Period
  23. 23. RFM VERIFICATION You can easily verify the power of the RFM segmentation. You should see the average value increase consistently with segmentation i.e. behaviour in the Training Period is a good predictor of value in the Test Period Training Period Test Period
  24. 24. THANKS! Find out more at http://www.marketingdistillery.com
  25. 25. THANKS! This presentation is licensed under Creative Commons BY 4.0, Icons (CC) BY 3.0 from www.flaticon.com by: ● Daniel Bruce, http://www.flaticon.com/authors/daniel-bruce ● Designerz Base, http://www.flaticon.com/authors/designerz-base ● Freepik, http://www.flaticon.com/authors/freepik

×