How to Build a Shopper Affinity Profile in Seconds

Personalization makes all the difference for the shopper. Learn how you can start personalizing your offerings to consumers during the same browsing session – in less than five clicks!

One of the holy grails in online retailing both now and over the past several decades is how to effectively personalize the shopping experience in a way that leads to higher conversions, larger cart size, and the creation of an affinity profile that allows for better personalization for future visits.

Why Personalization is Important

There is no shortage of vendors attempting to optimize personalization to, say, recommend the correct lamp to accompany an end table, or an eyeshadow to compliment a blush. And it’s no wonder. Epsilon Research has found that 80% of consumers are more likely to make a purchase when brands offer personalized experiences. On average, 71% of customers express some level of frustration when their shopping experience is impersonal, according to Instapage.

Today’s shopper, whether they’re truly aware of it or not, expects a personalized experience from start to finish if they have any intention of following through with converting, and this can be quite challenging. When it comes to matching shoppers to the products they really want, the traditional approaches aren’t as effective as they once were. Unique selections and strategic pricing or promotions just don’t do the trick anymore.

Differentiation, however, or taking a personalized approach in which a unique experience is created and tailored to individual customers, is making an impact on the success of e-commerce. Not only does it drive top-line growth, but it also builds customer loyalty. In fact, according to Instapage, 44% of consumers say they’re likely to become repeat buyers after a personalized shopping experience with a particular company.  

Where Personalization Has Missed the Mark in the Past

Instapage also says that 75% of consumers are more likely to buy from a retailer that recognizes them by name; who recommends options based on past purchases, or who knows their purchase history. However, less than 10% of tier 1 retailers believe they are highly effective at personalization. And nearly one-third of tier 1 retailers report having limited or no capabilities to support personalization efforts at all. 

While many retailers want to and claim to “personalize” a shopper’s experience, many have missed the mark time and time again. In fact, according to Accenture, retailers lost $756 billion globally in 2017 alone due to poor personalization and a lack of trust. 

When recommendations are poor and aren’t based on the attributes consumers care about, they tend to get lost in the generic pitfalls and overly broad segments of experiences. With limited insights related to products, especially how different products and attributes compliment each other, the ability to effectively personalize decreases.

Segmentation vs. Personalization

Although often used interchangeably, segmentation and personalization are quite different, independent techniques. While there is a time and place, traditional segmentation in shopping refers to treating people as one unit, or “ensemble.” This method involves focusing on a group of people with similar interests and tastes and categorizing, labeling, and boxing them. Rather than treating people as individuals with their own specific style, preferences, and needs. For instance, rather than mapping an individual’s needs, a retailer might focus on a one-size-fits-all campaign for those who fit a certain criteria.

For example, if 3 out of 10 shoppers add dress #7 to their cart, a teal velvet dress with a sweetheart neckline, then a retailer might target these shoppers with ads of dress #7 specifically and/or dresses visually similar to it. However, let’s say each shopper clicked on the dress in the first place for unrelated reasons. One shopper may have clicked on every velvet dress, while the other may have clicked on dresses in varying shades of blue, while the other was looking for dresses with a sweetheart neckline.

In this instance, if the retailer has a unique shopper affinity profile built for each consumer, then better 1:1 personalization could have occurred, as the shopper would have been better understood as a whole and correct predictions could have been made. Rather than generalized segmentation, each shopper could be matched closer to what they’re looking actually for.

What is a Traditional Shopper Affinity Profile?

A traditional shopper affinity profile has been a proverbial pot of gold at the end of the retail rainbow, built to help retailers capture what a shopper is actually looking for. Shopper affinity profiles capture visual affinities (e.g., color, pattern, shape) and non-visual affinities as well (e.g., brands, categories, price) in order to build a more comprehensive understanding of each individual shopper. 

These profiles are made up of psychographic attributes and are the core driver of purchasing behavior NOT represented in traditional customer profiles. Yes, demographic data, transactional data, and behavioral cues play a part, but what truly makes up a shopper affinity profile is: 

  • Lifestyle
  • Concerns
  • Opinions
  • Values
  • Fears
  • Interests
  • Personality
  • Attitudes

How to Make Personalization Even Better With an Intent-Driven Shopper Affinity Profile

Enhancing this traditional personalization for e-commerce doesn’t have to be complex – and one of the best ways to do it is by utilizing AI and machine learning. Here are a few key ways AI helps to build a better shopper affinity profile in seconds:

  • It helps you to gain a better understanding of who your audience is.
  • It makes better predictions about your customers and purchasing patterns.
  • It allows for a truly personalized experience, based on shopper intent.
  • It does away with bland co-occurrence and similarity models.
  • It works with the available data to make it leverageable.

When it comes to intent-driven personalization, instead of relying on common approaches for product recommendations such as “sold-with” or “most popular,” Lily AI’s customer intent platform gives you the ability to address a segment of one – the shopper in the moment, based on their unique style, occasion, or preferences. Efforts can then be better spent building off a consumer’s psychographic shopper affinity profile in order to help recommend the products they’re most likely to be interested in.

The Lily AI Shopper Affinity Profile



Real-Time, In-Session Predictive Analytics

Let’s take a look at how to do it better, and to do it in real time. Take Allison, for example, who is looking for a leather handbag. When Allison searches for “Leather handbags,” the predictive analytics attached to her personal psychographic profile helps to filter out the the products she has no interest in (i.e., clutches). It can then simultaneously recommend handbags that do fit the criteria she’s looking for, as in “green” and “cross-body” style handbags, which she does have an affinity for. 

Rather than delivering up irrelevant items that could result in her taking her search somewhere else, she instead is recommended items specifically tailored to her liking, helping to increase the chances of conversion. Even more importantly, the retailer is building this affinity profile on Allison from the moment she arrives on the site and starts browsing. In five clicks, we have a strong understanding of what she does and doesn’t like, and we can deliver an experience expressly tailored for her and her preferences.

Lily AI’s Personalization Potential

Lily AI uses its extensive product attribution data to train its AI models, which means we know more about what shoppers are actually browsing and buying – and why. Our generated unique shopper profiles show each shopper’s affinity toward different styles, colors, fits, and fabrics.  

With Lily AI’s 15,000+ product attribution taxonomy, retailers can deliver better product recommendations in seconds with visually-similar matching and algorithms. Our AI models can be customized to optimize for revenue, conversion, or any other metric that fits a retailer’s business needs. All in all, Lily AI helps retailers personalize the shopping experience in fewer clicks (less than five) and delivers robust product recommendations that convert with the help of optimized and personalized shopper affinity profiles.

Personalization is Powerful

Providing relevant and meaningful personalized experiences to consumers is key to helping them feel excited and appreciated. Differentiate your brand and retain your highest-value customers with better personalization practices and capabilities from Lily AI.
Learn More about Lily
A styled outfit put together