We are new to Recommendation. We have loaded the product feed into target and created form based recommendation activity for product detail page with criteria of most viewed product and builtin 4x1 html layout.
We are trying to call the Target Recommendation mbox that we created different mbox for recommendation activity and called by getoffer() with the parameter (entity.id) using DTM page load rule.
In staging mode we see in the response have details of recommendation activity, but it doesn't have recommended product data with html code.
Could you please let me know what we are missing?
Is it any issue with recommendation algorithm?
Do we need to activate the test and wait for sometime to get the recommendation suggestion? also could please tell how to debug if everything is working fine?
Thank you so much for your help.
This is a great question so you have uploaded a product feed but still don't see any results in the recommendations for the activity in question. First thing I would test is to see if the algorithm you have built is returning any products. The best way to test this is to do a download/export of the products from the activity in question. You can do that form the edit screen in the top right corner from the three dots. Let us know what you see when you download this Excel file.
a.) This file is blank. (This means there are no products either because they have not been uploaded to the collection or the algo did not qualify them.)
b.) There are products (This means something else is breaking down.)
Looking forward to hearing back from you with more details.
Mihnea Docea | Technical Support Consultant | Customer Experience | Adobe | (:: 1 (800) 497-0335
As suggested by Mihnea, you can check if there are products meeting the criteria by downloading the CSV from the activity. If the CSV file is not blank, you could check the Exclusion rules to confirm that the recommended products are not getting filtered.
You could also run an mboxTrace to troubleshoot the content delivery issues(Troubleshoot content delivery)
Thank you for your response,
i have checked the excel file by downloading i can see only 3 product IDs other than that rest is empty, so something is breaking.
Can you give some reference/documentation to check whether we have uploaded proper data or not.
This does not necessarily mean something is breaking, it also could mean that the algorithm (criteria) that you are using may be a tad restrictive.
For example, if it is a "Most Viewed" or "Top Viewed" algorithm, these products have to fall into the top 500 across the collection, through a rolling duration (for example if you have 2 weeks). Or could be a multitude of other factors that restricts the product matches, such as not uploading the inventory details, but yet restricting to "inventory>1" for example.
Here is a documentation about recommendations Feeds that you may find helpful: Feeds
However, since I sense that the criteria is the culprit, here's some docs covering criteria (algorithms) https://docs.adobe.com/content/help/en/target/using/recommendations/criteria/algorithms.html
If you are seeing some data in the data csv in the activity, it tends to point towards the criteria working, but being too restrictive,
If you would like to delve a little deeper, please consider reaching out to us in customercare, and we can take a deeper dive into the potential culprits.
@siva goutham, Here's the general documentation on planning and implementing recommendations.
Based on the criteria that you are using in your activity, ensure that you are also passing the right entities on the page as mentioned in Create criteria . For example, if you are using current item as recommendation key, then you will need to pass the entity.id value as a parameter in the display mbox.
Also you can run an mbox trace as described here to get additional debugging info.
Using mboxTrace on Recommendations pages: Adding mboxTrace as a query parameter on pages with recommendations replaces the Recommendations design on the page with an mboxTrace details window, which displays in-depth information about your recommendations, including: