...

Created Date: 19 March 2024

...

Employment upload process

Identify Duplicates

• Remove duplicate rows from Employment_file.

• Verify unique information for each Contact ID. If there are discrepancies, the file should be reviewed by the requester.

Check for Newer Data

• Check if the primary employment of that Contact was last modified before or after the Employment_file Date collection.

1. Download all the data from the Employment object in an Excel sheet (where AQB__SetAsPrimary__c = TRUE) with Contact ID and Last Modified Date.

2. Create a VLOOKUP column in Employment_file to retrieve the Last Modified Date from Employment object sheet by running a VLOOKUP on Contact IDs.

3. Create an “Action” column with the following criteria:

   If the Last Modified Date is greater than or equal to the Employment_file Date: No Action (remove from your sheet as newer data is recorded).

   If the Last Modified Date is less than the Employment_file Date: Data Matching.

   If the result is #N/A, it indicates no employment exists in the system for that Contact ID: Insert Data.
   
   * Correct date format if needed (dd/mm/yyyy)

Data Cleansing

• For free text fields, correct typos and inconsistent capitalization.

Field Mapping

• Map values from the file to the correct picklist values contained in the specified field.

  • For the Industry field, please map to USYD_Global_Industry__c.

  • Do not forget:
    

    • AQB__SetAsPrimary__c = TRUE

    • AQB__Status__c = Current or Retired (in accordance the person’s circumstances).

    • AQB__StartDate__c = If information is not given by the requester, the field should be populated with the day the data was received followed by AQB__Started__c = On or Before`.

    • AQB__Source__c = File source + year.

    • AQB__EmployerName__c: If it is blank, replace with “Not Provided”

    • AQB__Title__c: If it is blank, replace with “Not Provided”.

    • Started:On, On or Before

Action Folder Instructions

• Insert Employment folder:

1. Create a folder named “Insert Employment.”

2. Filter the Employment_file by “Insert Data” from the Action Column.

3. Save the filtered file as Employment_file_insert_data.

• Data Matching folder:

1. Create a folder named “Data Matching.”

2. Filter the Employment_file by “Data Matching” from the Action Column.

3. Save the filtered file as Employment_file_data_matching.

Data Matching Instructions

1. Upload Data: Request an Analyst or Manager to upload the Employment_file_data_matching to Snowflake as a table.

2. Query for Duplicates: Use the Employment_file_data_matching worksheet in Snowflake to query the table. Ensure you modify the script to include the correct fields and table name based on the uploaded file.

3. Export Results: Download the query results to an Excel sheet for further analysis.

Data Matching Input File (Manual Check)

Filter by Review Flag:

• 100% Match: Use the employment ID to update Jarvis employment data if necessary (e.g., new information from Employment_file_data_matching where the field in Jarvis is blank).
  No further action is needed if the data is already complete.

• > 80% Match or Between 60-80% Match: Conduct a sanity check.

  • If it’s a match, follow the process outlined in step 1.

  • If it’s not a match, add it to Employment_file_insert_data.

• Between 60-80% Match or Isolate for Review: Likely not a match. Manually review to confirm the results and follow the appropriate process as mentioned in step 2.

Employment Name Matching for Employment_file_insert_data

* Employer Name Matching Pathway: Data Visualisation and Reporting\Power BI\Current Power BI Dashboards\Power BI Data Matching\TAP Data Matching\Ad-Hoc Requests\Employment Name Matching

1. Check Employer Account in Jarvis:

   • Before inserting new employment data in Jarvis, verify if the employer has an account in Jarvis to determine the Record Type (Linked Employer or Non-Linked Employer).

2. Input Employer Name:

   • Use the Employment Name Template to input Employer Name from Employment_file_insert_data and Row Number . It’s recommended to use the =UNIQUE function to ensure only unique Employer Names are included.

3. Perform Data Matching:

   • Utilize the Employment Data Matching dashboard to perform data matching. Change the source in the transform data step and refresh the table as per the usual process.

4. Retrieve Data Matching Input File:

   • Obtain the Data Matching input file, which should include the Row Number, Employer Name, Jarvis Account, Jarvis Account ID, Result Comparison, and Record Type ID.

5. Review Matching Results:

   • If the name is not an exact match, the Matching Result column will display “Review”. Filter by this result to determine the appropriate Record Type and Record Type ID.

Getting Information to “Insert Data” and Upload

1. Linked Employer:

   • Use the VLOOKUP function with Employer Name as the key.

   • Retrieve Account ID, and add Linked Employer Record Type ID.
     
     e.g. Data Loader Mapping:

     Image Added

2. Non-Linked Employer:

   • Only include Record Type ID and use Employer Name from Employment_file_insert_data.
     
     e.g. Data Loader Mapping:

     Image Added

     
     

3. Insert (Data Loader):

   • Upload the files separately.

...

De-duplication Rules

Records deduplication involves Deletion and Consolidation activities. In principle, any deletion requires checking all data fields to avoid orphan records or information lost. Below is a table of the rules for the deduplication activities.

...

Jira Ticket