Expand my Community achievements bar.

Join us for an upcoming in-person Adobe Target Skill Builders event ~~> We're hosting these live learning opportunities to equip you with the knowledge and skills to leverage Target successfully. Learn more to see if we'll be coming to a city near you!

Able to recommend products as soon as they are added to the catalog of Adobe Target Recommendations


Level 2


Description -

Why is this feature important to you - Reducing the delay will allow promote newly launched products

How would you like the feature to work - being able to promote / recommend Adobe Target Recommendations entities as soon as they launch on application for user.

Current Behaviour - 15 Delay is present for product to be added to the catalog, and depending on criteria look-up the product can be recommended post next run of the algorithm which is anywhere between 12 - 48 hours.




Community Advisor


@neham25403375  - we have used Recommendations APIs which are often faster than uploading data through the interface. I will suggest you try https://developers.adobetarget.com/api/recommendations/#tag/Entities/operation/saveEntities API and check that it meets your requirements





Level 1


Hello, @Rajneesh_Gautam_.

Thanks for the information and for sharing your experience.
I´m afraid, by using the API, you solve the insertion of items in the collection but not the delay in the execution of the algorithm, which can take between 12 and 48 hours. Consider that you have a media site. You need the algorithm to run more frequently because if you have a breaking news item, it should be shown in the recommendation immediately (or as urgently as possible). Imagine a serious event in the locality or in the country or even in the world (an attack, a serious alert...), how can a media site wait 12 hours for this information to appear in the recommendation? Would the user understand this disinformation?





Community Advisor


Hi @elenam40808951  - thanks for the clarification. As per Adobe documentation, algorithms are executed every 12 hours, which I agree, can be made quicker.


In the meantime, there is one way to make this window shorter for popularity-based algorithms

- use a shorter lookback window: when its 6 hours, the algorithm runs every 3-6 hours