When moving data between systems, especially when implementing a new HRIS, a significant amount of data transformation is needed. Our handy guide explains the core steps we undertake when migrating data:
Step 1: No two organisations' data needs are the same, and that’s where detailed understanding of your source and destination datasets and the mapping between them is required.
With any HRIS implementation undertaken by Technoivity, we start by reviewing your Master Data documentation. We often find organisations may have out of date documentation and ensuring the correct baseline is understood is key. If required, we can help document the gaps and ensure the As-Is landscape is fully understood before getting too far into a major system change programme.
Step 2: We work closely with the new system’s solution architect to ensure the To-Be datasets are then mapped fully. Over many years we have developed an excellent data mapping template, to enable complex mappings and alignments to be fully and easily understood.
Over many data migration projects, we have seen every possible mapping scenario and have been able to incorporate all. Some interesting and recent examples include: mapping and rounding pay to the nearest shilling and migrating rota data with the exception of the 3rd Wednesday of a month, unless the shift is in Scotland.
Step 3: Data Transformation Tools are where the fun really starts. In the past, we often saw data migration taking place with Excel sheets and lots of formula or copying and pasting, but this is both prone to human error and are time consuming. We have therefore developed our Data Zymphony Transformation tool, which works within Microsoft Fabric to leverage the Azure cloud and AI capabilities to efficiently transform your data for loading into the new system.
Data Zymphony is configured by our team to meet your transformation needs and already outputs data for a number of common HRIS platform configurations, shortcutting the time required to setup the first run.
Your data can be provided from your source system either in data extract reports, or by SQL connection directly to your database.
With built in error handling and substitution rules a high degree of data cleansing and quality improvement can be undertaken within Data Zymphony.
Step 4: Loading and testing the data across multiple iterations to refine and ensure the transformation rules are outputting the correct data scenarios. Depending on the complexity of the organisation and HRIS configuration, we would expect to undertake 3-5 tests to migrate data, review and enhance the transformation tool.
Working with business data owners, we always want to test data in the new system and would prefer a UAT test cycle follow each Data Migration iteration. This way everyone has confidence in the data and new system or can highlight the areas for improvement.
Step 4a: Data cleansing in the source system is an optional step, potentially triggered from the lessons learnt from test loads. Data migration projects will undertake the most invasive review of your data and can often identify quality issues that are not flagged via BAU activities. It is always preferable to migrate clean data, so where possible cleansing rules are added to Data Zymphony, but it is sometimes easier or only suitable to fix these issues at source in the database before extraction.
(Repeat Steps 3 & 4 as required)
Step 5: Migrate the data and cutover to the new system, taking care to document the transformations, and business rules followed, should any future audit require a deeper dive into the migration process.
It is often easy to get caught up in the excitement of the new system and get BAU moving at pace, without fully updating the new Master Data documentation, so any project cutover should ensure comprehensive documentation exists and will continue to be updated over the life of the new system.
The deep understanding of data gleaned from migration, often flags core data integrity concerns that should continue to be monitored in BAU. Data quality monitoring dashboards should be monitored regularly in the new system to ensure the great work undertaken when migrating the data is not eroded.
Conclusion: These steps and best practice principles will put the new system’s data in the best possible place to help you maximise your Business Intelligence analysis and reach your strategic objectives.
For more information about our Data Zymphony Transformation Tool, Data Migration services or Business Intelligence solutions, contact us.
Comentarios