BlueConic recommendation algorithms

BlueConic offers a variety of content and product recommendation algorithms to help you deliver one-to-one content and product recommendations based on individual customers' or visitors' behaviors, interests, and preferences.

To activate recommendations, you use toolbar plugins for product recommendations or content recommendations. To calculate the optimal items to recommend, you can choose the strategy that best fits your content and marketing campaign.

This article provides notes on how the recommendation algorithms work. To add recommendations to your pages or app, follow the steps for Content recommendations or Product recommendations.

Breaking news / Top products (based on the "RECENT_VIEW" field)

Use the breaking news or top products algorithm to boost the articles or products that are most viewed during a timeframe you define. Algorithm notes:

  • Independent of profile
  • Based on statistics of the last X hours (default is 5 hours)

Content items and products are boosted based on the value of the statistics field “view,” which is incremented whenever the article or product page is viewed (independent of recommendations). You can set the algorithm timeframe in the content collector or product collector connection.

Viral news / Viral products ("RECENT_ENTRYPAGE")

Use the viral news or products algorithm to boost the articles or products that are most used as a landing page during the defined time frame.

  • Independent of profile
  • Based on statistics of the last X hours

Items are boosted based on the value of the statistics field “entrypage.” You can set the algorithm timeframe in the content or product collector connection.

Recent high CTRs ("RECENT_CTR”)

Use recent high CTR to boost the most clicked-through articles or products from BlueConic recommendations during the defined time frame.

  • Independent of profile
  • Based on statistics of the last X hours

Items are boosted based on the value of the statistics field “click” (someone clicks a content item or product in a recommendations dialogue) divided by the value of the field “recommendation_view” (an item is recommended). You can set the algorithm timeframe in the content or product collector connection.

Same category (“SAME_CATEGORY”)

Use same category to boost articles or products of the same category as the page the visitor is currently watching.

  • Independent of profile
  • Independent of statistics

Items are boosted based on the values of the “categories” metadata field of the “current item” versus the categories of the other items. If items have multiple categories, the score is higher if they have more categories in common.

Look-alike articles / Look-alike products ("LOOK_ALIKE")

You can use the look-alike articles or products algorithm to boost articles or products that have similar content to the one someone is currently viewing.

  • Independent of profile
  • Independent of statistics

Look-alike algorithms use a TF-IDF weighting factor that measures term frequency and significance to determine which items are most similar to the current item.

This determination is based on the words in the product or content item's “name,” “description,” and “text” metadata fields.

Recency ("RECENCY")

Use the recency algorithm to boost the most recently added articles or products.

  • Independent of profile
  • Independent of statistics

This algorithm uses the value of your content items' or products' “publicationDate” metadata field  to boost items. Newer items score higher than older ones.

Collaborative filtering (“COLLABORATIVE_FILTERING")

Collaborative filtering algorithms make product and content recommendations for individuals based on the behavior of other people who view the same articles or products. Some algorithm notes:

  • Independent of statistics
  • Dependent on the items the individual has viewed, put in a shopping cart, or ordered
  • Weight values: view=1, shopping cart=5, order=25
  • Operates with a ramp-up time, taking time to account for the interests of other viewers

This results in an affinity between profiles and items, which is used in a collaborative filtering model, as described in the Apache Spark API for collaborative filtering.

Same interest ("INTEREST")

The same interest algorithm boosts articles or products that are similar to those the customer or visitor already viewed, based on the values of the item's “categories” metadata fields.

  • Independent of statistics
  • Dependent on the items the customer or visitor has viewed
  • Operates with a ramp-up time

Seen articles / Seen products ("RECENTLY_VIEWED")

This algorithm boosts articles or products the customer or visitor recently viewed.

  • Independent of statistics
  • Dependent on the items the customer or visitor has viewed

The items the individual most recently viewed are boosted the most. You can use this algorithm to create a “last viewed articles” or "last viewed products" overview.

For product recommendations, there are similar algorithms for items the customer recently placed in the shopping cart and items they have recently ordered: "RECENTLY_SHOPPINGCART" and "RECENTLY_BOUGHT"