Customer segmentation process 

To derive business value from extensive data, several critical steps are necessary. These steps transform raw data into actionable insights and personalised customer experiences.

1. Data collection and cleansing 

The first step in effective customer segmentation is gathering and cleaning data from various sources. This data includes both internal and external sources:

Internal data: Transaction records and CRM databases provide valuable information about customer behaviour. For example, data on whether a customer bought a bag or a dress, the number of transactions, and average spending can offer deep insights.

External data: Tools like Google Analytics and social listening provide additional context. For instance, knowing whether a customer entered the website on a specific product description page (PDP), or the homepage adds another layer of understanding. 

Combining these diverse data types gives a holistic view of customer behaviour and preferences, serving as the base of accurate customer segmentation. 

2. Data analysis and enrichment 

Analysing and enriching the data through advanced analytics is the key next step. While raw data like purchase dates lacks depth, examining transaction intervals uncovers vital insights into buying frequency, crucial for understanding customer behaviour.  

This enriched data facilitates the identification of optimal products and timing for targeted marketing campaigns, crucial for retaining customers. For instance, a segment of the customer base which buy products in the higher price segment, but only during the end-of-sale season, will be offered a promotion during this period. 

3. Segmentation model output analysis 

Building on the process of feature engineering, where variables reflecting customer behaviour are created, an appropriate segmentation model was applied to the data, tailored to the desired business outcome. The next step is to analyse the model's output to determine what differentiates each cluster.  

One cluster could consist of mainly men, with affinity to casual-wear and having a high conversion rate and bag size per order, whereas another is based on women with affinity to fashion forward styles and high cart abandonment. It's crucial not only to identify the unique characteristics of each cluster but also to understand their role within the whole — examining their percentage of the total customer base or revenue.

4. Developing consumer personas 

The final step in the segmentation process is translating data-based segments into relatable consumer personas. This involves giving a name and a face to each segment, transforming abstract data into intuitive personas with recognisable behavioural patterns. 

Suppose a segmentation model identifies a group of customers who consistently buy higher-priced products but often take advantage of discounts. Additionally, this group predominantly consists of individuals aged between 29 and 35. By combining these insights, we can create a persona such as Margaret Markdown, who prefers to buy in the lower price-segment and is sensitive to discounts. These personas help businesses create a deeper understanding of their customer and develop targeted strategies to meet their needs. 

Applications of Customer Segmentation 

By transforming segments into real-life personas, businesses can anticipate future customer behaviour and preferences, leading to several concrete applications: 

1. Targeted marketing campaigns 

By tailoring interactions to individual customer needs and desires, businesses can significantly improve customer satisfaction and loyalty. For example, Margaret Markdown has affinity with products in the lower-price segment and prefers to buy casual-wear products. During the sale season, she will be targeted with text messages about deals, as this is the device she mainly used to interact with the website. This not only benefits her, but also increases the efficiency of the marketing strategies.  

2. Product development 

Businesses can tailor their product offerings to resonate more deeply with their target audience. For instance, if a significant segment of customers prefers eco-friendly products, a fashion retailer can introduce a sustainable clothing line to meet this demand.  

3. Loyalty program optimisation 

By segmenting customers based on loyalty levels, purchase frequency, and engagement patterns, businesses can design targeted rewards and incentives. For instance, frequent buyers get exclusive access to new collections, while occasional shoppers receive discounts to encourage purchases.  

4. Website personalisation

Customer segmentation enables businesses to personalize their website experiences across different stages of the customer journey. For example, Margaret might see a homepage featuring prominent promotions and a product listing page (PLP) sorted by price from low to high. The user experience (UX) helps increase conversion rates and customer satisfaction.

Use cases and proven results 

Customer segmentation has delivered significant results for our clients across various applications:

Improved product margins

One client experienced an estimated 10% improvement in product margins by tailoring their product offerings to specific customer segments. 

Increased member registrations

Another client saw a 300% increase in member registrations year-over-year by implementing targeted marketing campaigns. 

Enhanced newsletter conversion rates

A 0.74 percentage point uplift in newsletter conversion rates was achieved by personalising content based on consumer personas. 

These outcomes highlight the tangible business value of effective customer segmentation, demonstrating how it can drive significant improvements in key performance metrics. 

 We partner with clients to help them better understand their customers and act on these insights. Our approach benefits both customers, who receive a personalised touch, and businesses, which can allocate marketing budgets more effectively, inform product design, enhance loyalty programs, and improve user experiences throughout the customer journey.