2 days Ago
Kevin Campbell is the CEO at Syniti, part of Capgemini. getty Data migration is a critical process for organizations moving to new systems or platforms, whether you're embarking on a system modernization like SAP S/4HANA or working through a merger, acquisition or divestiture. You may have heard about the "lift-and-shift" approach, also referred to as brownfield or selective data transition, which migrates your systems and your data "as is.
" While that seems like a quick and cost-effective solution, I'm here to tell you that it creates more problems than it solves. Let me explain why the lift-and-shift approach is not an optimal choice for any modern data migration project, especially ones that are highly complex, like ERP system transformations. The lift-and-shift approach simply replicates your current system and data in a new environment.
This means: • Legacy inefficiencies, such as poorly structured data and outdated schemas, are carried into the new system. • Duplicate, incomplete or inconsistent data remains unresolved, leading to inaccurate reporting and analytics. • Historical problems like data silos and fragmented datasets are perpetuated.
You're taking your current data and dropping it into the new system without addressing any of the above issues. It's like moving homes. Do you just look around the house you're leaving, pack up literally everything in it and then just drop it all into the new house? Probably not.
You'd measure rooms, clear out that closet full of clothes that don't fit and throw out those roller skates you bought in the late '80s. Lift and shift is akin to taking all of that data (which, let's face it, is probably not at 100% data quality) and just dumping it into your brand-new system. If you don't address data issues before a migration, you miss the opportunity to modernize and optimize your data—which is critical to realizing business benefits from your transformation efforts.
If you've followed me for a while, you know there's nothing I love more than my kids or talking about the importance of data quality . Data quality is crucial for the success of any digital transformation. Period.
Yet with lift and shift: • There's minimal—if any—focus on data cleansing, validation or deduplication. • Inconsistent and redundant data will overwhelm the new system, reducing its performance and utility. • Errors and inaccuracies in legacy data undermine the trust and reliability of the migrated system.
Those last two points can and will kill your project. Poor-quality data is the most common reason we see major transformations not only fail to go live but also fail to deliver the benefits that every transformation is trying to achieve. Like many things in life, trying to fix something later is much more painful.
Trying to address data quality post-migration will be significantly more challenging and costly, requiring additional resources, additional tools and maybe another trip to the procurement office and your CFO to secure even more funding. Modern platforms and technologies are designed to work with optimized data structures. When you lift and shift your legacy, unclean data: • Your business will not be able to leverage the advanced features of these new systems, including things like real-time analytics, AI/ML capabilities, dynamic scalability, etc.
—you know, many of the reasons you took on the system modernization in the first place. • Your new environment becomes a shinier, more expensive replica of your old system with no added business value. • Legacy inefficiencies will drive bottlenecks in the system's performance, which defeats the purpose of the migration.
I will hear people say things like, "Our CIO was let go because the project didn't go live on time!" My response is always the same: Your CIO wasn't fired because of a missed deadline; they were asked to leave because the transformation they oversaw didn't deliver the business benefits that the transformation promised. While lift and shift seem cost-effective initially, there are hidden long-term costs that outweigh any perceived upfront savings: • Ongoing expenses for troubleshooting, cleaning up data and addressing compatibility issues. You did budget for that, right? • Higher storage and compute costs due to unnecessary or redundant data taking up space in the new system.
To paraphrase the late Biggie Smalls, "Mo Data, Mo Storage, Mo Storage, Mo Problems." • The business analysts and data scientists you brought on to help improve reporting are now stuck spending 80% of their time fixing bad data. Instead of finding ways to improve processes and drive growth, they're wasting their time manually cleaning up data errors and inconsistencies.
• Potential business disruptions caused by poor system performance or incorrect analytics based on junk data. In addition to my kids and data quality, I'm also really passionate about taking a data-first approach to any digital transformation . A data-first approach is going to prioritize data quality and ensure you're only migrating the highest-quality data.
While this approach requires more effort upfront, it delivers better long-term results by ensuring the new system operates efficiently, supports organizational growth and, most importantly, delivers business benefits. Hopefully, I've given you some important considerations to keep in mind before you approach your data migration. Let me leave you with a few key questions to ask yourself: • What business benefits are you actually expecting from this transformation? • How good (or bad) is your data quality right now? • Do you have solid data governance, or is it kind of a mess? • What risks are lurking that could throw this whole thing off track? If you can't answer these questions—and most importantly, if your data quality isn't at 99%—then lift and shift is not the move for you.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?.
