Target recommendations only displays the first recommendation

Avatar

CallMeRob

06-03-2019

As an organization, we are trying to use Adobe Target Premium for recommendations. This is my first time building a recommendation, so I am trying to build a prototype. The issue I am having is only the first recommendation ($entity1.name) properly displays information.   The design is coded to display 3 recommendations (entities). I am new to Recommendations, so my troubleshooting knowledge is limited.

I have created:

  • A sample feed with 14 products (CSV Feed). I confirmed it processed correctly.
  • Multiple sample criterias. I have enabled backup recommendations. No exclusions.
  • A 3 entity design, along with the Target sample designs.   I hardcoded the design, such as $entity1.name, $entity2.name, $entity3.name. It's not part of a function/loop.

While building the activity, the "preview" properly displays information on all 3 entities, however, when activated to production or using QA links only the the first recommendation displays information. The other 2 recommendations are blank.

I have tried testing different criterias and designs, but only the first recommendation displays. I am not sure what I am missing or what is left to troubleshoot.

Accepted Solutions (1)

Accepted Solutions (1)

Avatar

Eric_Vidana

06-03-2019

Not sure if super helpful but here is a go-to design for me to debug if data is coming through a particular criteria.

{

    "recommendations": [

        #set($start = 0)

        #set($end = 9)

        #set($range = [$start..$end])

       

        #foreach($i in $range)

            #if ($entities[$i].id)

                {

                    "id": "$entities[$i].id",

                    "title": "$entities[$i].name"

                }

                #if ($i < $end)

                ,

                #end

            #end

        #end

    ]

}

If you feel like the criteria is to blame, then you can set up an activity with front/back promotions that can be hard coded entities.

It is also worth passing a debug token (Setup -> Implementation -> Debugger Tools) on the url of the request. You will see it run through all the steps of the recommendation engine.