Data migration, while a simple task in theory, can evolve into an incredibly complex and time-consuming process if handled improperly. In fact, 83% of data migration projects either fail or exceed their budgets and schedules.
Small issues often become amplified during the process of migration, resulting in increased downtime and a negative impact to your bottom line. Without the proper procedures and precautions in place, migrations can become more of a nuisance than a helpful procedure. Here are four common data migration missteps to take heed of before you begin the transfer process.
Misstep #1: Failing to Inform All Relevant Parties
In the early planning stages of your data migration, it’s crucial to make sure anyone and everyone who will be affected by the migration is made aware of the project’s details. No matter how small the data set you’re arranging to migrate, there’s a team member on the other side of that data who needs to be informed of the upcoming changes you plan to make.
As Netapp recommends, “Track them [relevant employees] down and explain the need for this project and the impact on them before you get going on the task. If you don’t, you’ll certainly hear from them at some stage, and chances are good that they’ll disrupt your timeline.” To avoid further complicating a process that already demands a certain level of expediency, communicate with the soon-to-be-affected parties early-on and often.
Misstep #2: Running a Project Without a Governance Framework
The creation of an information governance process to guide your data migration project will save you time, effort, and frustration in the long run, as well as drastically reduce the amount of risk attributed with the project as a whole. According to Enterprise Systems Journal, some keys to successful data migration governance include:
- Having trust and confidence in the process/investment made
- Eliminating unpredictability within the project
- Providing better project visibility and management
Without a proper governance framework in place prior to launching the data migration process, you may experience time delays, added project costs, and poor user adoption, among other things. Sit down with the personnel involved with the migration and lay the groundwork early-on, before you experience negative performance indicators or costly interruptions.
Misstep #3: Migrating “Messy” Data
This tip is especially relevant if your migration involves a shift from a legacy system to a new system. Before beginning any migration process, sort through and clean your data to make the entire migration easier and more efficient. Archive data that doesn’t serve a relevant purpose to your organization – don’t just delete it or ignore it. Data profiling, in which you identify whether historical data is suitable for new systems, is another crucial step in organizing your data prior to migration. ETL Solutions recommends creating a retirement plan for data that is no longer relevant or useful: “Any data to be retired should be recorded, along with a description of what replaces it or what it can be removed. The data that is no longer needed may have to be archived for tax purposes or to meet the requirements of an industry’s governing bodies.”
Misstep #4 – Skipping Post-Migration Checks
Though tempting to rely solely on reports that provide basic quantitative checks, such as how many records were migrated and how long the process took, this kind of information does not deliver useful insight into any problems or errors associated with those records. While manual inspection and ad hoc queries can be useful for inspecting small amounts of data, applying these methods to large data sets can be incredibly time-consuming.
Larger data sets lend themselves well to a process called post-migration testing, otherwise known as validation testing. In this process, a user will launch a series of predefined queries to inspect the content of the source and target databases, which ultimately provides a discrepancy report of errors in the data migration process. For an even more comprehensive method of verification, reconciliation testing allows users to explore data correctness at the individual record level by retrieving and comparing each record in the source and target databases. Rather than simply describing overall error discrepancy, reconciliation testing goes a step further, pinpointing the specific records that appear faulty.
Have Questions? We Can Help.
If you’re in the process of planning a data migration or have recently completed a data migration project and you have questions/concerns, the experts at Envision can help. Whether it’s helping you create a governance framework, managing validation testing post-migration, or helping you to plan for a successful migration in some other way, we have a solution for your data needs. Contact us today to get started.
Explore our data analytics and business intelligence services to find out how you can leverage your data in powerful, connective ways.