How are recommendations generated?
Sherpa AI makes use of historical data points of user-item interactions. These data points are fed to the deep learning algorithms accounting for individual user affinity and similar user behavior.
How many recommended products for each customer are generated?
AI Recommendation provided recommended items for each user at max. These 25 items can be different for different users.
Will the past purchased products be recommended again?
Yes. It is possible. Th
Do I get to define the data points to be used to create recommendations?
Yes. All you need is to map the event(s) to some User Actions bucket(s) and add the bucket(s) to the AI recommendations configuration when editing or creating it.
What are the minimum criteria for Recommendations to work?
The followings are the criteria for the recommendation to be usable.
- At least 1 type of user-item interaction event having the relevant data points should be mapped to the User Action
- 1000 user-item interaction events in the last 2 months
- 25 unique users with at least 2 interactions each.
However, the more data, the better the recommendation will result in performance.
What happens to the users with fewer or zero interactions or some newly onboarded users?
When there are very few interactions available for a user they are recommended the popular products.
Is it possible to recommend products from only one set of products?
Yes, please create a catalog having one set of products only. You will need to select this catalog in the 'Create Recommendations' configuration.
I have a use case where my customers purchase or interact with items in my catalog infrequently. Is 'AI recommendations' still a good fit?
Yes, It recommends popular items initially and then quickly personalizes recommendations only after a few interactions. These interactions could be of any type e.g. Product Viewed etc.
How do I enable Recommendations?
Please reach out to your Account Manager today.