The above models can be used to solve both tactical and strategic tasks. It is not necessary to use all of them. In communications settings and personalized applications, segmentation by life cycle and RFM or RFM+P models are most often used.
Segmentation methods
To build the main models, it is necessary that the data be accumulated for at least one year. Such vp quality email lists models include RFM, RFM+P, LTV segmentation. Segmentation by the customer life cycle can first be built on business rules, and then revised as data accumulates.
Technically, segmentation is carried out by analysts using various software and machine learning. Clustering is becoming increasingly popular: when the total mass of clients is automatically divided into clusters with similar characteristics, and refining the characteristics of the segments is the task of analysts.
Segmentation cannot be static. Customer behavior changes, so some models need to be recalculated on a monthly basis - lifecycle segmentation and RFM. The results of the analysis for other models are less subject to change, so they should be reviewed no more than once a year.
In most cases, segmentation brings advantages:
understanding strategically attractive market segments
understanding the overall health of the base and business: for example, determining the percentage of customer churn
identifying the reasons for the decline in key indicators - for example, a reduction in turnover due to a decrease in the frequency of purchases or the average bill
the ability to focus efforts on profitable segments
the ability to save on discounts for customers who are active and ready to make purchases without additional motivation
allows you to differentiate strategies for positioning, promotion, assortment management and pricing
- Board index
- All times are UTC
- Delete cookies
- Contact us