Yes, you can input SKUs from different brands or categories. The system takes all provided SKUs into consideration when analyzing affinity, regardless of whether they belong to the same brand or category or not. However, it's important to understand that the analysis focuses on customer behavior related to all the provided SKUs as a group, and does not guarantee that they have an interest or even purchased any of the individual SKUs presented. A scattered selection of SKUs from different product areas might therefore end up giving you strange results.

Customers, regardless of the group, indeed may have purchased the product. However, the groups are created more based on how closely interested in it they are, as indicated by their behavior.

While it's likely that the person has purchased the product or one of the products indicated, it's really more about purchase patterns across the entire database. A customer may not have purchased a specific product, but could have a purchase pattern similar to other customers who have purchased the product. Meaning they get a high affinity score despite not having purchased the selected product.

These are the least likely individuals in the entire customer database to interact with the product.

Freshness depends on when the model was last trained, and currently it's done every week. There must also be a relevant amount of purchases for a product to be included, meaning customers with different product histories must purchase a product to be included in the model. This means that new products may need time, until their sales volume approaches that of the other products, to become relevant for the model. The model compensates for popularity, weighting bestsellers and popular products less heavily so they're not the sole strong signal.

Currently, product proximity is based only on purchase history—what the customers actually bought. This also includes a comparison to purchase patterns of other customers in the database. If a customer has not bought the product but still gets a high affinity score, it could mean that they have a similar purchase pattern as other customers who have bought the product.

The product affinity results include most contacts, distributed from low to high affinity. There are some exceptions. It only includes active approved members. Also, contacts who have only purchased edge products are not included. Short answer: no, but the majority of contacts are included.

No. Every time the AI recalculates the model, which is currently weekly, that particular group will be regenerated with the same given products, regardless of whether it is part of a segmentation or a target audience.

No. It can be included in groups only after the next model calculation, although it also might not end up in any of the groups.

Yes, this can happen. With frequent recalculation of the AI model, SKUs may come back into the assortment, run out, change, or go on sale, leading to new purchases. These effects apply not only to the listed SKUs but also related ones, so the group size will likely fluctuate.

You'll have to enter all the SKUs included for the product, one for each variant. It's not possible to use just a single variant to represent the product.

If your product database allows you to fetch all SKUs for a specific attribute such as color or size, you can use those SKUs in a product affinity search to find customers with the highest interest in that attribute.

Since high-affinity customers are not guaranteed to have purchased the product, you need to use the article transactions filter and logically combine it with the high-affinity group in your segmentation or personalization setup.

Most likely they have not. But even if they have, they may have purchased many other products from different brands or categories, resulting in a low affinity score.

This often depends on how many purchases the SKU has. New products or new variants may generate too few signals, leading to very large groups. Entering all SKUs for the product or applying additional filters can help. In some cases, frequently bought products such as shipping items can distort results and should be excluded.

A spike-like graph can indicate homogeneous purchase patterns among customers. When behavior is similar, affinity scores cluster closely together. Inventory size and structure can also impact score distribution.

When starting out, use a narrow high-affinity range, such as 0.75–1, as long as the group is large enough. If it’s too small, select the top 25% of contacts. Starting focused makes evaluation easier, and you can gradually expand ranges to balance reach and relevance.