Today's Agent Skill: Customer Segmentation Analyzer
What It Does
Not all customers are equal — some generate 80% of your revenue, some haven't purchased in six months, and some buy frequently at very low margins. Without a clear segmentation of your customer base, you're sending the same marketing messages to everyone and missing huge opportunities to retain your best customers and re-engage the ones slipping away. An AI skill that performs an RFM analysis on your customer data maps your entire base into actionable segments in minutes.
How It Works
The agent takes a customer transaction data export — at minimum, customer name or ID, most recent purchase date, total order count, and total lifetime spend — and groups customers into meaningful segments using the Recency, Frequency, and Monetary (RFM) framework. It labels each segment, describes the profile and recommended strategy, identifies your top customers by lifetime value, and outputs three specific quick-win actions you can take in the next 30 days.
How to Deploy It
Trigger this skill in Claude Code, Gemini CLI, Cursor, or Windsurf by pasting or uploading a customer list and saying "segment my customers by value and behavior." Works best with at least 50 customers and 6+ months of data. Export your customer list from Shopify, Square, QuickBooks, or your CRM as a CSV and paste it directly into the agent. The labeled segments integrate directly with Klaviyo or Mailchimp for automated campaign targeting.
SKILL.md — Ready to Deploy
## Description
Performs an RFM (Recency, Frequency, Monetary) customer segmentation analysis on a provided customer dataset and outputs labeled segments with strategy recommendations, a top-customer list, and prioritized 30-day action items.
## Trigger
Invoke when the user says "segment my customers," "analyze my customer base," "who are my best customers," "help me understand my customers," or provides a customer transaction dataset.
## Input
- Customer data table with at minimum: customer ID or name, date of most recent purchase, total number of purchases, and total revenue from that customer
- Optional: product categories purchased, acquisition source (referral, online, walk-in)
- Date range for the analysis (default: last 12 months)
- Optional: preferred number of segments (default: 5)
## Steps
1. Calculate RFM scores for each customer: Recency (days since last purchase), Frequency (number of purchases in the period), Monetary (total spend in the period).
2. Normalize each dimension to a 1–5 scale relative to the dataset distribution.
3. Assign segment labels based on combined scores: Champions (high R, high F, high M), Loyal Customers (high F, high M, moderate R), PotentialCopy the full SKILL.md and drop it into your agent's skills directory to activate this skill.