Product affinity

This AI-powered tactic enables targeting based on your customers' relationship with a certain product or products.

When you use the Product affinity tactic, advanced machine learning models will examine your customers' behavior. Based on their purchases, it will present building blocks you can use to generate customer groups with different levels of affinity to a specified product or products, product category, or brand.

The results are approximations and come with recommendations on how to use them.

How it works

Approach A: Using SKUs for specific products and variants

Go to Targeting tool, select Products on the side menu, and then Product affinity.
From your product database, get the SKUs of the products you're interested in (one or several).
Enter these SKUs in the field, separated by commas, then select Generate.
The product affinity score graph will appear. Adjust the slider to configure your desired segment size and add it to the composer using the Include, Exclude, and Restrict buttons.

The list of SKUs entered could, for example, contain all SKUs for a specific brand or category, or all variants of one product (such as different sizes or colors). All SKUs given are taken into consideration when calculating product affinity.

Approach B: Using brands and categories

Instead of entering SKUs, you can configure Engage to let you target groups of customers using drop-downs. The fields available depend on your setup, but typically include product metadata like brand, category, and sub-category — the same as those available in the Specific article filtering tool.

Product affinity dialog showing brand and category drop-down fields

You can also combine these fields along with SKUs as shown in Approach A.

For example, entering "boxer shorts" and "briefs" in the Categories field will make Engage examine all your contacts for their affinity to those two categories, effectively building an imaginary "underwear" category. This is functionally the same as adding all the underlying SKUs, and works as an OR statement (a union), creating a group larger than either individual category.

If you also add "Nike" and "Adidas" in the Brands field and select "sneakers" in Categories, Engage queries the model for SKUs that are sneakers from Nike or Adidas only. This corresponds to an AND statement (intersection) and results in a smaller group.

Reach out to your Voyado representative if you need more fields.

Product affinity score graph

Recent improvements make the experience more transparent, flexible, and actionable by replacing the previous fixed segments (Fans, Lookalikes, Detractors) with a fully configurable graph-based interface.

The graph shows the distribution of all customers based on their affinity scores. You can use a slider to:

Define a narrow, high-affinity segment (top scorers only)
Create a broader audience by lowering the threshold
Preview how many contacts match your settings

This gives you precision and flexibility when building segments tailored to specific campaign goals.

Combine customers with high affinity with the Recurring customer lifecycle group or the Active buyers purchase persona to further increase the probability of a conversion. It might be worth investing extra in this group for the specified products in your online ad spend.

Example use cases

Promote a specific product or category to the most relevant customers
Limit an email campaign to your most engaged contacts
Avoid over-communication by excluding low-affinity contacts
Run A/B tests with different affinity thresholds

Don't target customers in the low affinity section of the graph for this product or brand. Instead, consider promoting complementary or alternative products that might align better with their preferences.

To learn more about working with Product affinity, see the dedicated FAQ.

Using and interpreting the score graph

Below are some generalized examples of histogram score graph outputs and possible ways of interpreting them. Keep in mind that your product catalog and purchase history are the baseline for the results. These general examples may not represent the underlying reasons for your specific outcome.

Scenario 1: Majority of contacts around high affinity

In this scenario, a majority of your contacts have high affinity to your queried products.

Histogram showing contact scores distributed heavily to the right, indicating high affinity

If your graph shows scores concentrated to the right with a high average affinity, this could be due to:

Your query targets very popular products, linking many contacts' purchases to the queried products
Your query includes "super categories", "super brands", or popular products that are broad and include many common items
Your product catalog is relatively small, meaning many contacts have purchased a large proportion of it, linking your query to many contacts
For low-volume products, this pattern may indicate they were purchased by a highly active group of contacts, whose purchase history connects the queried product to many similar contacts with high affinity

Scenario 2: Majority of contacts around middle or zero affinity

Most contacts are neither strongly close nor far from your queried products.

Histogram showing contact scores concentrated around zero affinity

If your graph shows scores clustered around zero affinity, this could be due to:

Your queried products are relatively unknown to the model — they may be new or infrequently purchased, making it hard to identify contacts with strong high or low affinity
Your queried products are very popular but your regular purchase patterns split contacts into clusters, with the queried products acting as a bridge between clusters with no strong affinity in either direction
A majority of contacts have few purchases ("cold contacts"), so clear affinity patterns haven't yet formed. This should resolve over time as more purchase history is incorporated — it's a common pattern in newer environments without imported purchase history

Scenario 3: Majority of contacts around low affinity

A majority of contacts' scores gravitate toward strong negative affinity.

Histogram showing contact scores heavily concentrated at the low affinity end

A histogram showing strong separation between the queried products and the majority of contacts could be due to:

The queried products are purchased infrequently and represent a niche cluster
The contacts who would score high for the queried products are a distinct subset of your contact base, not representative of the majority
Your queried products are older, with few relationships to current and active products

Scenario 4: Less spiky distribution of contacts

Contacts are distributed more smoothly, covering large parts of the scale.

Histogram showing a smooth, even distribution of contact scores across the affinity scale

If your graph shows a smooth, evenly distributed pattern, this could be related to:

Your query contains "contradicting" products, giving every contact some affinity but no strong signal in either direction for the majority
Your query contains weakly connected or loosely correlated products
Seasonality may be a factor: your queried products are highly seasonal, so few non-seasonal products are strongly connected, but contacts are still linked through other purchases and spread across the full spectrum

Scenario 5: Two or more peaks of contacts

Contacts are distributed into two or more distinct peaks or clusters.

Histogram showing two distinct peaks of contact scores at opposite ends of the affinity scale

If your graph shows two or more distinct peaks or clusters, this could be due to:

A strongly connected cluster and a weakly connected cluster exist in your contact base — for example, contacts divided into communities with different purchase patterns such as premium vs. budget buyers
You've queried a combination of products where each product individually has distinct peaks, and the combination further strengthens those peaks
Your query contains substitute or mutually exclusive products, meaning some contacts have strong affinity while others have distinctly low affinity (buying from a competing brand or substituting the queried products)
You've queried older or non-recent products, resulting in older contacts having high affinity while newer purchasers have lower affinity

Search