Executing a successful data migration and cutover

Often, companies don’t reflect upon data migration until the time has come to perform this task. Too little attention is given to the preparatory phases of data transition. Given its complexity, doing proper preparation work is fundamental.

When executing an IT transformation initiative, be sure to give some early thought to your data migration and cutover phases. This will save you a lot of stress at later project stages.

The actual design of system functionalities, extensive testing, and proper definition of requirements are crucial elements of a successful go-live. In fact, having a good grip on the data that will be migrated is the be-all and end-all of such a project.

Go-live deadlines are often tight, making it critical for everyone to closely stick to the schedule, which is why this schedule needs to be defined properly.

Our data migration team has experience with many different approaches to the implementation process. Here are 9 examples of ideal practice during data migration and cutover activities. By following this advice, you will be able to complete your migration project successfully – on time, within scope and on budget:
 

• Any good preparation requires a determination of the scope and a common understanding of the predefined goals:
  • Identify the resources which you will need during the migration as early as possible.
  • Select an experienced data owner who can support you at any stage of your data migration and cutover and who understands your connection to (and dependencies on) other teams within the project.
  • Prepare an estimated timeline for project completion and plan the budget accordingly.
  • Identify all data entities required in the target system, determine the data sources and select the proper migration tool.
  • Define the acceptance criteria and document them. Clear and fully determined acceptance criteria are a prerequisite for validating the extracted data accordingly.
     
• Data cleansing is an important part of any data migration plan and should happen after the initial data assessment is put together. While most data cleansing can be automated, it nonetheless needs to be monitored and reviewed for any inconsistencies. The data’s quality - as well as pitfalls and gaps - should be analyzed at an earlier stage in the process in order to develop a detailed data quality plan. Doing this preparatory work properly will then enable automation wherever this makes sense.
  
  
• For key users to validate data, it is extracted from the selected legacy systems. Ensure that an extraction template and criteria are created in advance in order to extract the data in a structured way.
  
  
• In the data validation phase, it is essential to involve the key users early on. The more key users are kept in the loop, the more ownership they feel they have, and therefore, the better the validation will be. This step grossly increases your chances of better-quality data at the time of the migration because stakeholders have been given a heads-up and can prepare. Don’t leave the key users alone on this path – help them with the validation, guide them if necessary, especially when data volume is extensive.
  
  
• After the data has been validated, master and transaction data are loaded into the target system. Monitor the load and analyze any errors immediately.
  
  
• When the data are uploaded to the target system, they must be validated once again for accuracy, completeness, and the data acceptance criteria determined at the beginning.
  
  
• Be sure to plan the cutover early enough. In the best-case scenario, perform two or three data and cutover rehearsals before you go live. Especially when the project is large, decentralized and involves different stakeholders in different time zones – plan each step thoughtfully, review the plan with all participants and then rehearse.
  
  
• Remember to clearly document all data migration validations and the statuses of the cutover tasks. Documenting these tasks well gives you the chance to detect discrepancies between the source and the target systems, improving your likelihood of achieving a seamless cutover. It will also be necessary for the post-project audit that will likely take place after the hypercare phase ends.
  
  
• Do not underestimate the hypercare phase and the effort it requires. It is important to ensure that end users can use the system for their daily work, and the key users and technical teams have sufficient resources to support raised issues quickly.

If you want more information on this topic, get in touch with us. We’re happy to provide you with even more pointers ahead of your next data migration project.