Book a demo

Product recommendations

Recommendations – whether from friends, reviews or other encounters – are one of the principal motivators when it comes to making a purchase decision.

Custobar calculates product recommendations for use

These are based on a number of factors, and, importantly, are calculated using all recorded purchases, not just those made online.

Powerful automation for retail

Our system calculates recommendations based upon customers’ sales and browsing data, as well as discerning those products that certain groups of individuals buy. This is based upon the best-performing personal recommendation algorithm we have tested.

Some example product recommendations

  • A recommended product specifically for an individual customer
  • Content-based recommendations, such as ‘Similar products’
  • ‘Customers who bought this also bought…’ – item-based collaborative filtering based on sales data
  • ‘Customers who viewed this also viewed…’ - item-based collaborative filtering based on browsing data

Custobar product recommendation API

Product based recommendations

Format of the API request:
https://[customer_domain].custobar.com/api/recommendations/[product_id]/ [type_of_recommendation]/?[extra_parameter]&[extra_parameter2]

type_of_recommendation
This parameter can have 4 different values: similar, bought, viewed or all. All will return all types.

extra_parameters
ids_only=1 API returns only IDs of the recommended products. Otherwise it will return all product data in Custobar of the recommended products as JSON.

count=x you can specify the number of products you want to receive.

boost=1 -boosts similar products (brand, category, type etc.) higher. Boosted fields can be configured from company settings. Used in bought & viewed products.

category | type | brand can be used for filtering results

Some examples:
Other customers bought also these products (in these examples: Product ID = 0375704027):
GET https://test.custobar.com/api/recommendations/0375704027/bought/?ids_only=1

Filter recommendations with product type:
https://test.custobar.com/api/recommendations/0375704027/bought/?ids_only=1& type=Hard+Cover

Other customers viewed also these products:
https://test.custobar.com/api/recommendations/0375704027/viewed/?ids_only=1

Similar products:
https://test.custobar.com/api/recommendations/0375704027/similar/?ids_only=1

Top Products

Most sold products (returns product ID and count)
https://test.custobar.com/api/products/bought/

https://test.custobar.com/api/products/bought/?category=xxx
https://test.custobar.com/api/products/bought/?brand=yyy
https://test.custobar.com/api/products/bought/?brand=yyy&date=2017-01-01_2018-01-01

Most viewed products (returns product ID and count)
Like the most sold products.
https://test.custobar.com/api/products/viewed/
https://test.custobar.com/api/products/viewed/?category=xxx

You can also include count, i.e:
https://test.custobar.com/api/products/bought/?brand=yyy&count=5
https://test.custobar.com/api/products/bought/?brand=yyy&count=50

Personalized product recommendations for customer

https://test.custobar.com/api/customers/[Customer ID]/recommendations/?ids_only=1&count=10
Example: https://test.custobar.com/api/customers/294168/recommendations/?ids_only=1