BlueConic leverages your first-party customer data to deliver personalized recommendations at scale. By tailoring content suggestions based on individual behavior and preferences, you can enhance engagement, drive conversions, and build customer loyalty.
BlueConic offers a range of tools to customize personalized content recommendations. It uses contextual filters and algorithms to tailor recommendations to the current page, offers rotating variants for optimization, prevents repetitive content with exclusion rules, and ensures consistent delivery with fallback options.
Before you begin
Set up a Content Collector Connection to gather the content from your channels that will be used for recommendations.
Make sure you have the Content Recommendations Toolbar Plugin added to your tenant.
Add content recommendations
Content recommendations can be added to your channels in the form of BlueConic Dialogues:
Open the dialogue you want to use for content recommendations.
To add a content recommendations section, edit an interaction and click "Insert object" from the content editor toolbar.
Select "Content Recommendations" to add a placeholder to the content editor.
Click and edit or double-click the placeholder to open the Content recommendations pop-up where you can configure the content placement.
Configure your content recommendations
Once you've added a placeholder for your content recommendations, you can configure the following settings:
Select a Content Collector from your tenant, which serves as a source from which recommended items will be chosen.
Click the dropdown menu to choose a template to display your recommendations. You can also choose to edit your own template for recommendations.
(Optional) Set a frequency cap to exclude articles after the user has seen the article a set number of times and not clicked.
Specify the number of displayed recommendations. The default setting is 4 items per set, with one algorithm and filter. Add sets for varied recommendations.
Double-click on a recommendation set to edit its algorithms and filters.
Click Copy to duplicate a recommendation set.
Reorder recommendation sets by dragging and dropping the recommendation bar up or down.
Click Delete to remove a recommendation set.
(Optional) Use fallback algorithms to fill recommendation gaps when defined sets are insufficient. Fallback items, matching your filters and algorithms, populate the remaining slots.
Save your settings.
Define or edit your content recommendation algorithms
To refine the algorithms that choose recommended items:
Click on the algorithm box in a recommendation set to open an edit window.
In the Algorithms tab, you can control which algorithms are used to generate content recommendations.
BlueConic offers recommendation algorithms that are profile-based, aggregate usage stat-based, and content-based. For details on how each algorithm operates, see BlueConic content recommendation algorithms.
In the Filters tab, you can include or exclude content based on its metadata or other settings, like whether or not a user has already seen a piece of content.
When you click the Add filter button, you can construct a new metadata filter tailored to the data your content collector gathers. For example, if your content consists of articles tagged by the categories Sports or Entertainment, you can filter for articles that exactly match those categories.
As you make changes, they will be reflected in the content placement in your editor window.
Save your settings.
Next steps
Refine your content recommendation algorithms with A/B testing.
FAQs
I removed an article from my page. When will it be removed from the recommended items?
A content item will be added to a queue to be deleted when the required fields can no longer be scraped in the browser of the visitor, even if a visitor views the item -- for example, if the article has been deleted or if the publication date is no longer available.
I cannot see item X in my personalization recommendations. What could cause this?
Items or articles are added to the recommendations queue when a customer or visitor views the item. Articles that have no views from customers or visitors won’t be added to the queue.
Items in the recommendations queue are evaluated and if the required fields become valid, those articles are added to the content store for personalization.
How does indexing work with personalized content recommendations?
Indexing items may take some time while when the collector is still collecting items Depending on how much traffic your channel has, there might be a short delay indexing content items when the collector is still actively collecting items.
When will my item be added to the queue to be evaluated?
Items are added to the queue for the first time when a visitor views an article. Also, when an item or content article has been clicked on twice, but no view follows the click, the system checks for all required fields and re-evaluates whether to include or delete the item.
How can I view my top recommended items?
To see which content items are recommended most often, use the Top recommended items insight.
How do I add content recommendations to emails?
Using the Open-Time Email Recommendations feature, you can deliver dynamic, individualized content or product recommendations via email based on up-to-the-minute customer data. These recommendations are updated the moment your customers open their email.
Why are some of my product recommendation links pointing to the homepage instead of the correct product pages?
This typically happens when products with variants (grouped by Item Group ID) are imported via SFTP and the product file isn't properly sorted. BlueConic uses the URL from the first item in each product group, so if that item has a homepage or placeholder URL, all variants may link incorrectly. To fix this, ensure the product file is sorted by Item Group ID, and the first product in each group contains the correct URL.
Why can't I filter content recommendations by "contains" in BlueConic, and what can I do instead?
BlueConic only supports exact matching for metadata filters in content recommendations, meaning there is no built-in support for partial matches or "contains" logic. If you need to filter content based on metadata that contains multiple values for example (like "download" and "webinar"), you can use the 'categories' field to store multi-value metadata, which allows for more flexible filtering based on exact values. For complex metadata, such as custom fields like 'Content Type', you’ll need to store these values in the 'categories' field to ensure BlueConic can filter content recommendations based on multiple tags. If your metadata includes multiple values (e.g., "download," "webinar"), ensure those terms are consistently stored in the 'categories' field for filtering. This way, you can apply filters based on exact matches to values stored in that field.
If this solution still doesn’t fully meet your needs, you may want to consider adjusting your metadata structure to make filtering more straightforward, ensuring all relevant values are represented within the 'categories' field.