Update type.md #61
Update type.md #61
Conversation
| @@ -39,7 +39,7 @@ These recommendation types are social-proof driven to help shoppers find what ot | |||
| |---|---| | |||
| |Viewed this, viewed that|Recommends products that shoppers view disproportionately more often with the currently viewed product.<br/><br/>**Where used:**<br/>- Product detail<br/>- Cart<br/>- Confirmation<br/><br/>**Suggested labels:**<br/>- Customers who viewed this product also viewed (PDP)| | |||
| |Viewed this, bought that|Recommends products that shoppers tend to buy disproportionately more often after viewing the current product. Helps guide shoppers to discover products that they might not have otherwise noticed.<br/><br/>**Where used:**<br/>- Product detail<br/>- Cart<br/>- Confirmation<br/><br/>**Suggested labels:**<br/>- Customers who viewed this ultimate bought<br/>- Customers ultimately purchased<br/>- What do others buy after viewing this product?| | |||
| |Bought this, bought that|Recommends products that shoppers buy disproportionately more often with the currently viewed product. Most often used on the cart or product detail page to increase exposure of related cross-sell product to increase average order value. Displays highly relevant products shoppers can add to their cart by aggregating what other shoppers have bought with the current product.<br/><br/>**Where used:**<br/>- Product detail<br/>- Cart<br/>- Confirmation<br/><br/>**Suggested labels:**<br/>- Get everything that you need<br/>- Don't forget these<br/>- Frequently bought together| | |||
| |Bought this, bought that|Recommends products that are infrequently purchased compared to the other products. This type will take into account the logical relations between products and will aim to recommend more interesting combinations versus top-purchased combinations. Displays highly relevant products shoppers can add to their cart by aggregating what other shoppers have bought with the current product.<br/><br/>**Where used:**<br/>- Product detail<br/>- Cart<br/>- Confirmation<br/><br/>**Suggested labels:**<br/>- Get everything that you need<br/>- Don't forget these<br/>- Frequently bought together| | |||
There was a problem hiding this comment.
I think "infrequently purchased" sends the wrong idea that we're ordering them to only return infrequently purchased SKUs. If the goal is to convey that this is not a simple co-occurence metric but a more sophisticated collaborative filtering algorithm, we should just say that. I suggest - Recommends products that shoppers buy disproportionately more often with the currently viewed product. Note this is not a simple-occurrence metric but a more sophisticated collaborative-filtering machine-learning algorithm that looks for "interesting similarities" that are not skewed towards popular products. Displays highly relevant products shoppers can add to their cart by aggregating what other shoppers have bought with the current product.
There was a problem hiding this comment.
This should be added to the other collaborative filtering types as well and not just bought-bought
There was a problem hiding this comment.
the problem we have is that by reading this description it is difficult to understand what the end result merchants will get. @krithikachandran can we add an example or easy way for people to understand this complex behaviour?
What Bulk has in mind now is that our algorithm returns "low sales volume products" which from your description this is not always the case
There was a problem hiding this comment.
I know this PR is closed but I wanted to respond to Sandra's comment
@sgonzalezmangana - I agree that the language might not be easy to follow. I am not sure how to provide an example without explaining LLR or probabilities which will only make this more complex. The simplest way we can differentiate is just specifying that this is based on shopper aggregate behavior across sessions and that if they are looking for in-session behavior then this rec type is not suitable for them.
|
@devops-devdocs import the pr |
|
Migrated to internal repo. Thanks @shiftedreality and @krithikachandran! |
|
Something went wrong |
Purpose of the pull request
This pull request (PR) ...
Affected pages