Addressing the Challenges of Data Migration
Data migration—transferring valuable information from one system to another as a company expands—is an everyday fact of life. It is often misperceived as a routine, “copy and paste” task and many businesses underestimate the complexity of data migration, including the time, cost, and labor involved. Awareness and understanding of several key factors when undertaking a data migration can lead to a more efficient, effective, and smoother process.
Want More Tech News? Subscribe to ComputingEdge Newsletter Today!
Failure to plan is planning to fail
In the simplest of terms, data migration is about moving data from one system to another. There are many variables to consider, cost is only one. Examples of data migration also include moving all employee e-mail addresses from one provider to another or utilizing a new tax preparation software for a business, making it necessary to transfer financial records to that new platform. Sometimes it’s necessary to relocate data to a software program that can scale up to more efficiently as a company grows.
Having a plan, or a set of tasks in place ensures that data migration is robust and easily available while eliminating excessive downtime and minimizing its impact on the business. Customers don’t want to hear that a vendor’s online services will be down for hours or more as a data transfer takes place, especially if they are trying to access their own stored information. Planning should also prepare for future migrations based on forecasting of a company expansion, based on the history of that growth.
Poor planning increases the timeline and can lead to costly delays, especially for companies involved in online commerce. For example, Uber and Amazon have data migration projects occurring on a regular basis, avoiding significant downtime. Even when the migration occurs far from the customer interface platform, on the “back end” for example, it can impact all services to some extent.
“Copy and paste” doesn’t always work
In short, a simple copy-and-paste operation is not realistic. While it sounds simple enough, every new system that data is transferred to has its own set of nuances and limitations that must be analyzed and addressed. Therefore, it is essential to understand the end-to-end differences of the product being considered the next best step for a company. What type of maintenance and reporting tools are featured in the new system? Is it more labor-intensive to make that switch at this time? What is the return on investment, and when will it be achieved? To address these issues, it is helpful to conduct beta testing before committing to the switch.
Six factors to consider in a strategic data migration plan
Time, effort, and dedication to data migration are often underestimated by some enterprises, leaving them scrambling when the switch takes place. Designating a project leader who “owns” that migration from start to finish is a good start. Having the resources in place, including time, personnel and money are necessary.
When looking to enhance company performance and become more competitive, businesses should consider the following six factors.
1. Legacy system modernization
Before replacing a legacy system, companies should ask themselves several questions:
- Why upgrade from the “old” database to a new one?
- What are the advantages of system modernization?
- Is the new system more secure in this era of cyberattacks?
- Is the new system faster, or is the legacy system no longer supported by the industry?
- What features might be lost if any?
2. The costs associated with third-party technology
Outside parties claiming they can assist with data migration can end up costing more than if the transfer was managed in-house. If considering a third-party vendor, carefully determine which tools are appropriate for your organization. Costs may be direct or indirect. In recent years, data migration software packages from tech giants such as Samsung and Microsoft have crashed, often requiring fixes. An online search of “data migration crashes” shows that third-party systems are not bulletproof.
3. Multiple databases can create fragmentation, impacting business operations
Companies that have numerous divisions and hundreds of teams are always looking to scale up (think Amazon again). Large enterprises often store customer data in a myriad of databases that speak different computer languages. Some of those databases may have been created by key players long gone from the company. That’s where standardization across multiple platforms can ensure that when data migration becomes necessary, it becomes a more rote, simpler task.
4. Mergers and acquisitions impact data migration/integration
Because businesses grow through mergers and acquisitions, it is important to do upfront work that establishes commonality in software languages and future objectives. This helps make transferring data from a recently acquired company to the “parent” system less labor-intensive and less costly.
5. Scaling as the business expands
Forecasting and planning for the arc of company growth are essential. From a data-migration standpoint, this involves ensuring that any new computer system or database being considered has the capacity to manage that anticipated expansion as it occurs.
6. Migrating data to the cloud
Rather than storing and managing data in-house, which can become very expensive and difficult to scale, cloud software solutions, such as Google, Amazon or Microsoft, may be viable options. By taking administration and maintenance costs off their plate, businesses can focus on product development and future growth. Increasingly, companies are looking to the cloud to manage data management and scaling when appropriate.
Best practices in data migration initiatives
Data compliance and industry regulation issues—including HIPAA health record protections, the European Union’s GDPR standard for data protection, and the Payment Card Industry Data Security Standard (PCI DSS)—are other factors to consider when making a data migration, or at any other time for that matter. Are encryption standards robust enough? Assess security firewalls and make any necessary upgrades before embarking on a data migration project.
Similar to product development, data migration requires advance planning and an allocation of resources, including time, personnel, and budget. Identify priorities and come to an agreement on how the phases of any data migration project will take place. Get the “buy-in” needed throughout the corporate structure before undertaking such an operation. Advise customers in advance about any potential service interruptions.
Nearly every data migration project is more complex than anticipated. To overcome common challenges, create a robust and strategic data migration plan that also takes future expansion into consideration. To ensure seamless migrations, ensure that data migration goals are in alignment with organizational goals, creating a template for future success.
About the Author
Parul Bharadwaj is a Sr. Technical Program Manager for DoorDash with extensive experience managing large-scale impact across AWS Global Partner Operations, AWS Business Technology Solutions, and AWS DynamoDB at Amazon. Most recently, she is leading engineering initiatives for data infrastructure, experimentation, and real-time streaming projects at DoorDash. Ms. Bharadwaj holds a master’s degree in Information Systems Management from Carnegie Mellon University and can be reached at email@example.com or LinkedIn.