How Qloo’s affinity score measures consumer preferences

Every choice we make, from the brands we trust to the places we visit, is influenced by a web of interconnected preferences. These preferences don’t exist in a silo; they reflect our broader interests, habits, and the way we navigate the world. For businesses, understanding these connections can make all the difference in how you engage with your customers.

This is where Qloo comes in, using its advanced AI technology to uncover the deeper links between culture and behavior. Qloo’s affinity score is like a map, revealing the pathways between individual tastes and the bigger picture of consumer behavior. By tapping into these insights, companies can create experiences that feel personal and relevant, turning data into meaningful connections that truly resonate with their audience.

What is affinity?

Affinity at Qloo measures how closely related one item is to another item. Said differently, it gauges the strength of the correlation between any given input—such as a demographic profile, a specific interest, or a geographic location—to any notable person, place, thing, or interest. This relationship is quantified with a score ranging from 0 to 1. A higher score indicates a stronger affinity, meaning that the input and output are highly connected and that the input is highly relevant to the output.

How are affinity scores created?

Creating an affinity score is a complex calculation that involves advanced statistical analysis, machine learning, and AI models. The process begins with the ingestion of a wide range of anonymized data signals. Qloo’s databases include regular updates from diverse data sources, including transaction data, digital interactions like online reviews and social media activity, geospatial travel patterns, and engagement data from different types of digital content. We also incorporate demographic data, such as age and gender, along with psychographic data that captures lifestyle, interests, identity groups, and values. All of Qloo’s data is completely anonymized, ensuring top-tier privacy and compliance with global regulations.

With this rich proprietary dataset in hand, we deploy sophisticated AI and machine learning models to analyze it. These models are designed to uncover hidden connections between entities by identifying patterns and co-occurrences. For instance, neural networks help us detect relationships between different entities, while statistical models assess tastes in specific locations. Additionally, content models analyze similarities based on attributes, and demographic analysis helps us evaluate the preferences of different population groups.

The final step in this process is the development of the affinity score. Our models identify correlations between entities and produce a score that reflects the strength of these relationships. An affinity score of 1 represents a perfect match, where the input and output are strongly correlated, while a score closer to 0 suggests little to no correlation.

What does the affinity score reveal?

The affinity score is a powerful tool that provides a normalized measure of interest and relevance. For example, if a company wants to know which clothing brands are most strongly correlated with 30-year-old women living in Manhattan, Qloo’s affinity score can provide a clear list of answers. A higher score would indicate that a particular clothing brand is highly relevant to this demographic, allowing businesses to tailor their marketing, product development, and location-based strategies with precision.

In a geospatial context, affinity scores can help businesses understand the level of interest in a particular entity within a specific location compared to other entities in the area and globally. For instance, a restaurant chain might discover through Qloo’s Taste AI that their brand has a higher-than-average affinity in Austin, Texas, compared to other regions. However, when looking at specific neighborhoods within Austin, they may find that one area has a significantly lower affinity score compared to others. This insight allows the restaurant to tailor its offerings and marketing strategies to different neighborhoods within Austin, perhaps by adjusting its menu to better suit local tastes or increasing targeted promotions in areas with lower affinity. By understanding these nuances, the business can optimize its strategies for both local and global audiences, ensuring they resonate more effectively with specific markets.

Making the most of affinity scores

For Qloo’s clients, the affinity score is a versatile tool that can be leveraged in numerous ways. By understanding the affinities of their target audiences, businesses can create personalized marketing campaigns that resonate deeply with specific, unique groups and audience segments. They can also identify which products are likely to appeal to certain demographics, making informed decisions about product development and positioning.

Affinity scores can also enable companies to pinpoint the most receptive geographic locations for market expansion, ensuring that their efforts are targeted and efficient. Additionally, businesses can use these scores to select partners that share strong affinities with their own products or services, creating connections that enhance their overall offerings.