The BlueConic CDP offers several tools to create personalized product recommendations. To create product recommendations in BlueConic, you use the Product Recommendations toolbar plugin in the toolbar of the content editor. This lets that you insert personalized product placements on your pages or in an app (note that you'll need to also set up a Product Collector).
Inside a BlueConic dialogue, select the content editor to insert a content placement based on a recommendation algorithm of your choosing at the place of the cursor.
To add a recommendations placement, edit an interaction and select Insert object > Product Recommendations from the content editor toolbar.
When you select "Product Recommendations" BlueConic adds a placeholder placement to the content editor. Hovering over the placeholder makes it light up in blue. Click and edit or double click the placeholder to open the product recommendation pop-up for configuring the products that get recommended and how they appear to customers.
Configuring product recommendations for personalization
Collector: Select a Collector, which serves as a source from which recommended items are chosen. You define this by configuring a Product Collector Connection.
Layout: Click the dropdown menu to choose a template to display your recommendations -- for example, in a list of links, or a list with images, etc. You can also choose or edit your own template for recommendations.
Frequency cap: You can choose to exclude items after the user has seen them a set number of times and not clicked on them.
Recommendation sets: Choose the number of products you'd like to include in the recommendations that are displayed. Add new sets of recommendations with different algorithms and filters that select the recommended content.
You can have multiple sets of recommendations. Double-click the recommendation set to edit or adjust its algorithms and filters. Click copy to duplicate a recommendation set. If you have more than one, you can reorder the recommendation sets by dragging and dropping the bar up or down. Click delete
to remove a recommendation set.
Enable fallback: You can set a fallback algorithm so if the recommendation set(s) you've defined don't deliver enough results to fill a set, you can fill the empty space with other items. If you choose to enable fallback recommendations, products that don't already appear in the recommendation set and also match the configured algorithms and filtering options are used to fill any empty spots in the recommendations area.
Product recommendations algorithms for personalization
To refine the algorithms that choose recommended products, select the algorithm box in a recommendation set.
The algorithms tab opens, where you can adjust the algorithms that select a unique product set for each individual user. As you make changes here, you should be able to see them reflected in the placement in your editor window.
To return to the main recommendations window, click the arrow in the upper left corner.
Choosing your product recommendation algorithms
You can choose from a number of algorithms to boost the prevalence of products with certain characteristics as described below. For each algorithm added, you can control whether it is used at all, or to what degree it is incorporated into the overall selection.
There are three types of algorithms you can apply: aggregate stats-based, product-based, and profile-based.
Aggregate stats-based algorithms for personalization
- Recent high CTRs: Products that have been clicked most within recommendation placements served by BlueConic.
- Top products: Products that have been engaged with the most across the site over the last several hours or days (configurable in product collector).
- Viral products: Products that have been popular as an entry point or landing page to the site over the last several hours or days (configurable in product collector).
Product-based algorithms for personalization
- Look-alike products: Products that have a similar textual makeup in the description to the current product.
- Same category: Products in the same category as the current product.
- Recency: Products with a recent release or availability date.
- Seen products: Products the user has already browsed.
- Carted: Products this user has already placed in the shopping cart (and which are still there).
- Products carted most often: Products most often placed in the shopping cart during the chosen time frame.
- Bought: Products this user recently bought.
- Products bought most often: Products most often bought during the chosen time frame.
Note that for product-based algorithms, you need to use a Product Collector to create a collection of product data (called a product store) to feed recommendations.
Profile-based algorithms
The following profile-based algorithms also take a "ramp up speed" setting, which allows BlueConic to react more quickly to any information that is being populated in the user profile.
- Collaborative filtering: Products engaged with by other users similar to the current user.
- Same interest: Products in the same categories as those the user has shown the most interest in.
See BlueConic recommendation algorithms for details on how each algorithm operates.
Filters
In the Filters tab, you can choose to include or exclude products based on metadata or other options.
- Seen products: Products the user has already browsed.
- Carted: Products the user has already placed in the shopping cart (and which are still there).
- Bought: Products the user recently bought.
- Hide out-of-stock products: Include only products that are in stock, based on an in-stock indicator configured in the product collector.
Metadata filtering for personalization
Metadata scraped from the product collector can be used to filter the product selection:
In this way, you can include or exclude products based on their relationship to metadata within the current product being browsed, or within the user profile:
Below is an example where an explicitly defined value, "clearance", being excluded from a selection of products:
You can also match personalized recommendations to a user's categories of interest, as in this example:
See your most recommended products
With any personalization strategy, it's important to measure your results. To see which products are recommended most often, use the top recommended items insight.