How RFM Analysis Improved Customer Segmentation and Retention at Kick Pleat

What Business Questions Guided the RFM Analysis for Kick Pleat?

The RFM model analysis for Kick Pleat required a nuanced understanding of a customer base shaped by both seasonal shopping behavior and a high share of one-time purchasers. As a retailer driven by curated fashion collections and distinct seasonal demand patterns, Kick Pleat needed deeper visibility into how customers engage across both eCommerce and store channels, and where the strongest opportunities for retention and growth exist. 
With a significant portion of shoppers purchasing only once, understanding how to convert initial interest into long-term loyalty became especially important.

The key questions were:

• Who are Kick Pleat’s highest-value customers, and how can they be retained?
• What behavioural patterns distinguish each customer segment?
• How can marketing strategies be refined to increase repeat purchases and customer lifetime value?
• How can Kick Pleat better align paid media, messaging, and content with specific customer segments to improve engagement and conversion?
• How can the brand reduce reliance on one-time purchasers and build stronger long-term customer relationships?

To address these challenges, we conducted an RFM (Recency, Frequency, Monetary) Analysis and a Cohort Analysis to segment customers, uncover purchasing trends, and provide actionable marketing insights.

How Did We Approach Customer Segmentation Using RFM Analysis?

The methodology behind this RFM model, developed to enhance customer messaging and improve retention, combined structured data extraction, advanced analytics, and machine learning to generate actionable insights. Below is an overview of the process and the role data science played at each stage.

• Data Preparation: To build a complete view of customer behavior, we first extracted data from Shopify, Kick Pleat’s eCommerce and POS platform. This phase also involved essential data cleaning and preprocessing steps, including excluding ineligible orders and standardizing formats to ensure consistency and accuracy.
• RFM Analysis: We then calculated Recency, Frequency, and Monetary scores for each customer to evaluate purchasing behavior and build a clear segmentation framework. Using machine learning techniques, we developed a model that automatically grouped customers into four key segments: loyal shoppers, occasional high spenders, casual shoppers, and lost shoppers.
• Cohort Analysis: To complement the RFM model, we conducted cohort analysis to examine how customer groups behaved over time, based on the month of their first purchase. This allowed us to identify retention trends across different cohorts and measure how purchase frequency changed following the initial transaction.

The RFM model analysis revealed four primary customer clusters represented below:

Kick Pleat’s sales distribution closely follows the Pareto principle, where a relatively small portion of customers generates the majority of revenue. The top 10% of customers by lifetime value account for more than half of total sales, highlighting the significant role that high-value shoppers play in the brand’s overall performance. Within this group, Loyal Spenders and Occasional Big Spenders are the primary drivers of revenue, demonstrating the importance of nurturing and retaining these high-value relationships.
Beyond this top tier, the remaining revenue is generated primarily by Casual Spenders and Dormant Low Spenders. While their individual contributions are smaller, their large numbers collectively represent an important portion of total sales. Additionally, a significant share of these customers have only interacted with the brand once or infrequently, highlighting a clear opportunity for retention, re-engagement, and strategies designed to encourage repeat purchases.

Seasonality plays an important role in Kick Pleat’s customer acquisition, with noticeable peaks occurring in December–January and again in June–August, while April and October tend to experience slower acquisition periods. These fluctuations likely align with the launch of new seasonal collections, which drive heightened customer interest and engagement.
Customer value continues to build steadily over time. On average, a cohort takes roughly 2.5 years to generate revenue equivalent to what it produced in its first month, effectively doubling its initial contribution within that timeframe. As cohorts mature, they continue generating meaningful revenue, with long-term value extending well beyond the first few years of acquisition.

Customer Retention Strategy

Something Powerful

Launch win-back email campaigns designed to re-engage one-time buyers and encourage repeat purchases.
Tailor messaging by customer segment, using distinct tones, structures, and offers rather than applying the same communication approach across all audiences.
Use paid media to retarget lower-value segments with personalized product recommendations and upper-funnel campaigns that rebuild interest and consideration.
Activate paid media campaigns for higher-value segments with engagement and conversion objectives focused on deepening loyalty and driving repeat revenue.

Customer Acquisition Strategies

Build lookalike audiences in paid media based on high-value customer segments to increase the likelihood of acquiring new customers with stronger long-term value, rather than relying solely on broad audience targeting.
Strengthen SEO and content marketing efforts to attract qualified organic traffic and support sustained customer acquisition.
Introduce referral programs that encourage existing customers to recommend the brand to others, helping drive more high-intent customer growth.

Ongoing RFM Monitoring and Advanced Insights

Conduct RFM analysis ahead of each season to refresh customer segments and track how customers move between clusters over time, helping assess shifts in value and engagement.
Explore more product- and brand-specific segmentations to compare behavioural patterns within each cluster and uncover more tailored opportunities based on customer interests.
Personalize product recommendations using past purchase behaviour to increase relevance, strengthen engagement, and encourage repeat purchases.