Query Service Data Distiller

[ribhubanerjee]

Has anyone used Data Distiller capabilities and built any use cases around it? I want to understand what the benefit of it is and how it can be used or implemented? any suggestions on certain use cases will be helpful

[Anil_Umachigi · Accepted solution]

@ribhubanerjee There are quite a few use cases,in short every time you need a way to manipulate, subset and in some cases fix data issues data distiller can come in handy.. some examples include 

1. Custom calculated dimension/metrics: A eCom company would like to populate values for certain metrics into a field in schema ( or use it purely for reporting), and this needs to happen every night ( churn rate, birthday month etc..), one could use to do this within the platform and use the output for desired purpose 
2. Data Subset: A insurance company has loyalty data for all brands flowing into a single dataset, the business wants to only enable certain fields to profile, a subset of data can be created using data distiller 
3. Scheduled Query: Ability to schedule a query for time bound usecases
4. Data Fixes: Ability to fix data issues, not really updating the dataset but use create a new dataset after all error are fixed. 

Hope that helps 

Anil

[SaurabhM]

The best explanation for this is here: 

https://data-distiller.all-stuff-data.com/unit-1-prerequisites/prereq-101-why-was-data-distiller-bui...

I am hoping to bring all of the Data Distiller material into this community soon. 

In essence, when it comes to any batch data processing for Real-Time Customer Profile, Audiences, Personalization attributes, and even Customer Journey Analytics, Data Distiller will help you clean, shape, and manipulate that data. It also supports creation of reports and creating feature engineering pipelines for AI/ML workflows. 

Here is what is underneath the hood and it is a very good read of how it actually works: 

https://data-distiller.all-stuff-data.com/unit-1-prerequisites/prereq-102-key-topics-overview-archit...

[ribhubanerjee]

Thanks for sharing this. This definitely helps in better understanding.