Accurate and complete data must be the objective when you are implementing a new system but in practice moving it from one system to another appears to be the most important thing on the Project Teams mind. This is ever more apparent since the advent of Self Service applications and interfaces between key business systems meaning the downtime for the systems must be kept to a minimum.
Compounding this is a lack of budget to employ specialists to take care of the Data Migration, with suppliers all too often leaving clients to their own devices when it comes to populating their data import sheets. The biggest part of any new system implementation is the data migration and import of data into the new system. It is an area that most organisations overlook and generally do not have the internal skillsets to extract the data from their old system or to create the datasets to populate a new system. The client owns the data, they know it better than anyone, and must be involved in the process. All too often a client will get the data extracted by their incumbent system provider and that is passed onto the new supplier to import with little or no data cleansing or validation.
This will lead to months of system misery and frustration and ultimately the users will lose faith in the software and not trust the results that they gain from the system. Reporting will be inaccurate or incomplete, Managers will become frustrated with the system, and it is at this point that individuals then resort to manual processes and the good old excel spreadsheet.
Where do I start?
The starting point should be to review all your data sources and this should include data in your HR system, Payroll system, excel spreadsheets, paper records and any other system that holds people data. Once you have identified all the data sources you will be able to make decisions on what you do with the data and at what point.
How much data should I migrate?
There are several areas of data, for example there is current data and then there is history data. Many suppliers advise only to bring in the current data and suggest that the history data is kept in spreadsheets or their old system.
This needs to be considered as it will restrict your reporting within the new system for the first couple of years. If you leave data in your old system there is a temptation for your staff to fall back to the old system at every opportunity, again this needs to be considered.
Finally, if the data is output to spreadsheets it will be hard to search the data and is potentially not secure outside of a system so should be a last resort option.
Completeness of data
When you have decided on your sources of data and how much you are going to import there is a job pre-migration to validate and update the data so that you are importing complete data records for your employees.
This can be done in one of two ways, you either do the update prior to the migration or an alternative to this could be to get staff to update their records via self-service post go Live. This can be an effective method of getting staff to use the system and save the HR teams a huge amount of time collating and entering the data into your current system.
Either method works but communication with the staff is key to making it a success and the system being populated with complete data.
Correctness of data
Once you have complete records there is then the task of reviewing the data to ensure that it is correct. This requires an eye for detail and a good knowledge of analysing large volumes of data. Due to the vast number of database tables in an HR system that hold employee data it is very difficult to validate data in the same way that you can on a payroll system. In payroll, this is typically dealt with by a parallel running process where you balance numbers in System A with that in System B; this is not quite so easy in an HR system.
As a starting point, some fundamental checks of number of employees in your upload of data can be compared with the data imported into the system that can be carried out for several areas in the system such as absence, etc. This checking doesn’t ensure the actual records are correct and in most cases is something the HR team or the employee should validate.
Data validation for signoff of an HR system is always very much a moving beast and a more pragmatic approach needs to be adopted; it isn’t as cut and dry as a parallel run process for a payroll system.
If you need to do any of the following elements of data migration – cleansing, analysing or updating of mass data – save yourself days of work and involve specialists to assist you.
If you would like to see how LitE Consulting can help you please complete the contact form and we will get back to you