At the core of effective marketing is the ability to engage with customers in meaningful ways. But it can be hard to know how to scale the personalization needed to support this strategy.
Segmentation is an important pillar of automating marketing personalization, but you can’t rely on static demographic information alone to inform your campaigns. That’s why more retailers are now leveraging AI and machine learning models like RFM analysis as a tactic to support a successful marketing strategy.
What is an RFM Model?
First things first…what is RFM?
RFM stands for:
- Recency: When a customer last purchased from your brand.
- Frequency: The number of purchases the customer has made from you in total or during a specific timeframe.
- Monetary value: The amount of money the customer spends in a transaction.
An RFM model is a statistical technique that analyzes dynamic consumer behaviors and segments them into different categories for targeted marketing campaigns.
It gives you insights into how to adapt your marketing strategy based on how customers interact with your brand.
Skypoint Cloud RFM Machine Learning Model: Turning Insights Into Action
RFM analysis sounds good in theory, but how can you turn the insights into action at scale?
Skypoint Cloud can now connect the dots: We recently launched the AI-powered RFM machine learning model to help our customers automate their segmentation strategy and act on the insights with the click of a button.
Data Input
All you need to get started is a customer’s transaction data, which you can export from a platform you already use, such as Shopify or Square.
- The first parameter in the model is recency. Upload a table or CSV file with the transaction date.
- The second parameter is frequency, which is calculated from the transaction_date.
- The third parameter is monetary value. Provide the model with a table that includes the order_amount field.
Select Your Configuration
Once the model has received all the required values (i.e. transaction_date and order_amount), you can choose between two configurations:
- Default: The entire result set is divided into 20 percentile increments. For recency, the lesser the value, the higher the percentile. For frequency and monetary value, more value equals a higher percentile.
- Custom: You can set the percentile values based on your needs. For example, breaking the dataset into the top 30 percentile, then 20, 10, 25, and 15.
Scoring and Segmentation
The model will associate each percentile bucket with an RFM score ranging from 1 to 5.
Next, the model will put customers into different categories, from which you can create an audience segment.
Finally, the “quick audience” section makes it a breeze to set up a dedicated campaign using popular tools such as Klaviyo, Campaign Monitor, SendGrid, HubSpot, etc.
Engage Customers and Drive Sales With RFM and Product Recommendation Models
Providing personalized product recommendations based on past purchases helps increase engagement, especially with customers identified as “need attention” or “about to sleep” by the RFM model. Skypoint Cloud allows you to generate a list of recommended products for each customer automatically.
To link the RFM entity with the product recommendation entity, create a relationship based on the Skypoint ID. Then, use the relationship to generate a specific audience within Skypoint Cloud Studio, such as “RecommendProductsForAboutToSleepCustomers.”
You’ll receive a table of customer email addresses and a list of recommended products for each person. You can export the resulting entities into your preferred campaign tools to efficiently engage with customers and boost sales.
Increase Customer Retention with RFM Models and Customer Churn Models
Churn is one of the costliest issues for your organization to deal with, so it’s important to be proactive in your customer retention efforts. Skypoint Cloud’s customer churn prediction model helps you identify customers at risk of churn by analyzing RFM data and customer behaviors.
The insights will allow you to take action before they decide to churn, such as sending personalized campaigns with relevant promotions and incentives to entice them to return to your business.
Maximize Customer Lifetime Value with RFM and Customer Lifetime Value Models
Our RFM model can also be paired with a customer lifetime value (CLV) model to maximize your revenue by:
- Identifying high-value customers: Segment customers based on purchase behaviors and identify those most likely to become high-value customers with Skypoint Cloud’s built-in CLV model. Then, prioritize them for targeted retention strategies or upsell/cross-sell opportunities.
- Optimizing acquisition strategies: Predict the potential lifetime value of new customers with a CLV model, then optimize your acquisition strategies to focus on those most likely to become high-value customers.
Tap Into the Power of RFM Models
Skypoint Cloud allows you to combine RFM model output with other prediction models to generate in-depth customer insights and improve business outcomes.
Our platform integrates with various tools, including Databricks SQL, to support a robust data infrastructure. You can also connect Skypoint Cloud with your favorite campaign tools to streamline workflows, automate analytics, and gain actionable customer insights to improve your marketing strategy.
Schedule a demo today to learn more about Skypoint Cloud for retail and hospitality.
