Personalise the customer experience

Marketing Analytics use cases

RFM modelling

One of the basic types of segments/cohorts is RFM modeling based on three important inputs:

Recency: When was the customer's last order?
Frequency: How often do they buy?
Monetary: What amount do they spend?

The higher score the customer receives, the more value they bring to your company. Use the RFM model to define your communication tactics that increase value from your customer (spend more/ more frequently). Reward your high-value, active customers with exclusive offers, special privileges, and free shipping, engage your low-value active customers by sending welcome offers, products, or relevant information to get them accustomed to your brand, or reactivate your low-value lapsing customers by sending discounts. In the Intergramma Customer Case we demonstrate how the RFM Model has helped them to gain a complete overview of their customer, automate the scoring of loyalty groups, gain insights into cardholder loyalty, and monitor trends.

Actions to grow your business:

  • Reactivate high-value lapsing customers
  • Grow low-value active customers
  • Reactivate low-value lapsing customers
  • Migrate prospects to become customers
  • Generate traffic to website
  • Migrate guest accounts to become customers
RFM Modelling

Sentiment Monitoring 

Analyse customer service interactions for content and sentiment to adjust marketing messages to improve the experience. With sentiment monitoring, you will improve relationships with your customers by focusing on how they feel about your brand and translating that feeling into your marketing message. Engage frustrated or angry customers through a push to email, media or anything that can be activated based on just an email address.

Sentiment Monitoring

Data-driven segmentation

Use techniques such as clustering to segment customers based on behaviour (example types of products they browse, value and frequency of spend, browsing behaviour, profile information), rather than predetermined business attributes. Address these clusters with a coherent strategy, executing in near real-time via email, media or basically anything that can be activated based on an email address. Start making informed decisions across different customer touch-points. Identify and prioritise groups of consumers you might be missing or neglecting.

Interest-based segmentation

This type of segmentation allows to find people with similar interests. The approach is to:

  1. Derive customer interests based on web data and purchase data (for example categories, discount)
  2. Use cluster analysis to find advanced groups
  3. Find groups that are most similar to each other within each group and distinctive to others between groups.
  4. Score new customer on a daily base in Google Big Query
  5. Create persona’s to stereotype segments
Data-driven Segmentation

Personalisation Engine

Use segmentation and predictions to influence what a customer sees on the site or app. Start to identify cross-sell opportunities on your site and up-sell customers as they engage with you. Personalise your website content and ads using behavioural data from users at scale.

Personalisation Engine

Integrating External Data

Use weather, location and other external data sources to enrich the customer micro-moments. For example, trigger promotions for certain products based on the weather in a city on a particular day for a particular consumer demographic, incorporating creative messaging that resonates and adds value to the customer. Go a step further and incorporate local happenings, events, topics into the messaging for a more relevant and meaningful communication interaction.

Incorporate External Data

Win with a central view of the customer

When data is centralised, you have the foundation to personalise service and sales-related communications.

Learn more