Bluecore’s powerful product recommendation engine can now incorporate the buying and browsing behaviors of your shoppers to add an additional layer of personalization to the products you recommend to your customers.
In this Hands On With Bluecore session you'll learn:
- Everything you need to know about Bluecore’s new recommendation algorithms
- How to select the best algorithm to meet your goals
- Best practices from Bluecore’s customers using recommendation algorithms
- How to begin testing with your current campaigns
3. terminology Look What You Forgot…
Input Products
• Typically featured in first content area
• Fed into the Bluecore algorithms that power
recommendations
• Set of products a customer engaged with
• Includes browsed, carted or purchased items
How They Are Used
Output Products
• Set of products generated from the
recommendation engine
4. terminology
Product Attribute
• Product Name
• Product Price
• Division
• Category
• Sub-Category
• Brand
• Stock Status
A characteristic that defines a certain product
Out of box, Bluecore’s integration will pick
up these product attributes:
7. Attributebasedrecommendations:Keytakeaways
1. Best Programs for Attribute Based Recommendations
Attribute driven recommendations may yield higher conversion for higher funnel
programs, where customers haven’t yet decided what to buy, and where less customer
behavioral data is available.
2. Test & Optimize
Shopping behavior can vary greatly by brand and industry, so test into the best strategy
for your programs.
3. Data Integrity
Attribute driven recommendations are only as good as the product data available on your
ecommerce site. This is the best approach for sites with structured data that is consistent
and prevalent across all products.
8. Recommendationstrategyoption#2:
collaborativebasedrecommendations
Definition: Collaborative based algorithms use collective wisdom of customers to identify which
products tend to show up in the same session, e.g. which products tend to be viewed together or
which products tend to be bought together.
Strategies We Will Walk Through Today with Use Cases:
1. Co-View
2. Co-Cart
3. Co-Purchase
4. Best Sellers
10. Bestsellers
Definition
The Best Seller strategy shows either site-wide best sellers or category specific best sellers.
• Site-wide is great when you do
not have a lot of customer
browse data available.
• Category specific is great for
product notification triggers
that are driven by changes in
the catalog.
Where To Use
11. Co-view
Input Product
Output Products
Recommendations
Definition
As the name indicates this algorithm recommends products that tend to be viewed in the same session with the input products. This was
popularized by Amazon's "customers that viewed these items also viewed ...".
This algorithm is a good choice for product abandon emails, especially for partners with less consistent onsite data structure.
Where To Use
12. Co-purchase
Input Product Output Products
Recommendations
This is similar to the Co-View/Cart algorithms except that we're recommending products that tend to be bought with the input
products. This data set can be enhanced by feeding offline purchase data to Bluecore
Definition
Where To Use
We typically recommend this algorithm for post-purchase emails where cross-sell is a key strategy.
13. Co-cart
This algorithm recommends products that tend to be carted in the same session with the input products.
Input Product
Output Products
Recommendations
Definition
This algorithm is a good choice for partners that are unable to pass Bluecore purchase data or have low sales volume.
Where To Use