Data Discovery Overview
Data discovery is the process to understand the customer’s data and how the data is used in their business processess.
What is the purpuse of the data discovery? » Data Modeling » Data Schema » Customer Identity » Activation
Data Analysis Overview
» Process of cleaning, transforming, and modeling data
» Discover useful information for business decision-making
» Extract useful information to make informed decisions
Data Analysis Best Practices
- Determine if the data being analyzed is Profile data, Event data, or Lookup?
- Determine if all the data columns provided need to be loaded to Experience Platform? Which ones are required to satisfy the defined use cases?
- Are there any fields that have garbage or unreadable values? Could these be encrypted fields in the source system?
- Are there any data fields that are not populated or not present in the data source that would be required to fulfill the use case? It is especially important to do this exercise with key fields or identity fields.
- Date columns should be reviewed for far future or past dates, and date format (in flat file or JSON) should be reviewed in detail.
- How much volume or history of this data is required to satisfy scoped use cases?
- What are the key identity fields in the data being analyzed? How do they relate to other data sources that have already been analyzed?
- Determine how the ingested data will be kept in sync with the source system?
- Delta File
- Streaming Ingestion
- Other
- Which is the best tool to use to analyze this data or data sample?
- Excel, Large text viewer can be done on sample flat files and JSON files
- In some cases, it might be better to load sample data into a temporary table. This will enable SQL queries to be run and possibly visualization tools.
