So you are looking to migrate data from one system to another and the business thinks it’s as easy as copying and pasting from one spread sheet to another? The data, in many cases can often be an afterthought at the end of a CRM migration. Trying to explain why a data migration can be complicated and fraught with peril is like trying to bash a square peg into a round hole. Sometimes only failure can open people’s eyes.
However to avoid getting that far the main insight I would share is “if you fail to plan, you are planning to fail”, a phrase coined by Ben Franklin. I’m sure he wasn’t talking about data migrations, but it holds true.
However to look a bit deeper I have tried to list some common issues that organisations run into in the hope that you don’t!
When drawing up a plan include everyone with an interest in the successful conclusion of the migration:
Mapping from system to system based on field names listed in DDLs or ER diagrams is not sufficient, rushing through the analysis may lead you to misidentify sources and skipping profiling may conceal missing content. Instead, the migration team must examine the data on both source and target applications to understand the existing content, rules, relationships and transformations required.
On most IT projects, developers test a new process or interface using test data or a small subset of production data. For a data migration, this practice is a mistake. The team cannot meet its objectives if it is not using the real data to identify any issues with the migration.
You always need to ensure that data will work in the new application. This often requires some form of data cleansing, enhancement, consolidation or transformation.
If your target application implements new or consolidates business functionality, plan to include data quality work in your migration project.
The process of loading data to production must be driven by:
The cutover plan is one of the first things that needs to be agreed. Think about whether any downtime is required, whether data can be cut across in one go or in batches and whether both systems need to run in parallel.
For a CRM data migration, it’s better for data discoveries and their surprises to occur at the beginning of the project when it is easier to react. If you do the data analysis up front, you minimize the need for reworking during the project, limiting the risk of delays and going over budget.
Avoiding this pitfall is simple; involve the data experts, generally the business users. These are the people able to make necessary decisions about when data is “good enough,” what needs to be fixed at the source, what needs to be enhanced and which transformations are working properly. They determine what data to retain and what to archive.
It’s not enough to include business users only in the initial phases, migration projects that succeed involve business users from start to finish.
The appropriate tools to support data migration include all needed processes and don’t force IT to reinvent the wheel repeatedly. Choosing the right tools and knowing how to use them dramatically improve the odds of success for every application data migration.
These are my suggestions to ensure a smooth transition of data, if you have top tips of your own I'd really like to hear them.