Data Migration to SF a few steps

Hi All…This is my first blog post on data Migration. Hope you like the same.

When it comes to Data Migration, people are under assumption that it is straightforward. However, after working one can understand, it is not as simple as it seems.

I am here to share a few thoughts with respect to Migration of Data from Legacy to SF.

As most of you will be aware it consists of 3 steps

  • Extract
  • Transform
  • Load

However, a lot of work is involved for these 3 steps to be executed.

Even before extracting the data, we should be aware of the structure and the fields that will be migrated. For this we take the SF EC workbook, import and export templates as base and prepare an excel which is discussed with client to make them understand the fields in each portlet. They decide which legacy value to be mapped to which field of the portlet. During the extraction there can be fields which are straightforward like First Name. Some fields might involve picklists. The value of these picklist might differ in SF and legacy system. For example, for field “Marital status “they can refer value “Married” as “M” in legacy system and in SF it might be “Married”. These mappings are also discussed and provided by client. The conversion date from which the data must be migrated (The conversion date, which is mostly based on EC go live date, payroll earliest retroactive date. Any changes before this will not be accepted if there is a replication from SF to EC.). The portlets which will contain data for all employees (active and terminated) the portlets which will contain only active employees are also determined before extraction.

Once this is done the extraction step starts. The extraction of data is done by client or the consultant. In our case it was the technical team of client. They used Peoplesoft Queries and extracted the data in the csv format for each portlet basing on the format discussed and shared us the same.

Once the files have been provided, we did a manual data validation to see if all the mandatory fields are provided, if the length of the field is not exceeded, data type of the field is as expected, picklist values are mapped to SF value, the header and sequence of fields is as per the excel provided and few customer specific checks like if the country of company is FRA we will have country of birth, region of birth, place of birth provided in bio info, if a field specific to a country is present or not.

Once this is done the transformation of data from legacy value to SF value is done. The transformation mostly involves changing the legacy value to SF value. It might be changing a legacy company code to SF legal entity, changing the legacy picklist value to SF picklist value etc. This is done either manually or using excel or tools.

After this step is done data load to SF is done either using import or export, import employee data or using internal tools. Also, during load there might be cases where we need to disable few fields to non-mandatory if there is no data provided by client or we need to deactivate some rules so that they do not trigger during the load. After a load is done there might be errors which we face which are discussed with client and left as is of corrected basing on their call, With this we complete the data migration.

To summarize Data Migration consists of below:

  1. Template preparation and discussion with client for mapping and fields.
  2. Extraction of Data using tools.
  3. Transformation of Data either manual, via excel or internal tools.
  4. Load of Data using SF templates or internal tools.

Request you to comment and like the post if you find it adds any value. Also feel free to add on anything missed from my end.