Data migration is a significant challenge for any business, but it’s a crucial step in business transformation. That said, it’s rarely straightforward. A poorly executed migration can have serious consequences, and the process has traditionally been manual, time-consuming, and full of potential pitfalls. Add to that the constant pressure to keep business operations running smoothly, and it’s easy to see why implementing changes – even necessary ones – can be difficult.
This is where the right expertise, strategy, and tools make a significant difference. With advancements in technology like automation, data migration no longer needs to be a painfully slow, manual process.
In this blog, Frantisek Slepanek, Team Lead of Delivery & Service Development at Columbus, shares his insights on data migration. We’ll explore the common pitfalls, challenges, and – most importantly – show how businesses can plan ahead for a smooth, drama-free migration.
Challenges with data migration
Frantisek highlights the historical burden many companies feel when approaching data migration, which can make the challenges seem overwhelming. "Many organizations hesitate to tackle messy data continuously in time because it has built up over decades. But migration to a new ERP system is a rare opportunity to clean up, archive, and optimize how data is managed in the future.
"I always tell customers: if you're switching ERPs, see it as a fresh start. Don’t carry unnecessary historical baggage – reduce as much as you can. Be prepared for change, and choose the right migration approach to make sure your hard work pays off in the new system." One of the biggest challenges Frantisek sees when speaking to customers is that data often feels abstract, making it difficult for people to fully grasp what they’re dealing with.
"A simple discussion about data can lead to different interpretations for technical and non-technical people. Take the word ‘apple’ – for an English speaker, it’s just an apple, but for someone else, it might mean something entirely different. A simple example is when we tested this with AI-generated images and saw how meanings can vary widely – each Gen AI will generate an image of ‘apple’ in a different way, and sometimes not even correctly. This is exactly the challenge we explain to customers starting their data migration journey.
"From a business perspective, data is critical to run the system – but it’s very often not the priority when the new system implementation starts. Companies are looking for a new innovative system, new advanced processes, and the latest ERP features, yet very often they say, ‘just use the data we have.’ And that’s where the real challenge lies. Data and structure need to be considered in the whole picture as well."
Frantisek also points out that he reminds customers of the simple truth of ‘garbage in, garbage out’ when it comes to migration. Poor-quality data – whether outdated, inconsistent, or full of duplicates – can lead to reporting errors, operational inefficiencies, and misguided business decisions. If businesses migrate messy data without addressing these issues upfront, they risk carrying over the same inefficiencies and limiting the benefits of their new system.
“You can have the most sophisticated ERP in the world and invest huge amounts of money into it. But if you don’t pay attention to your data, you’ll never realize its full potential, especially in the modern era of data mining, process intelligence, automation and use of AI,” says Frantisek.
Another challenge companies face is finding the right time to clean up their data and migrate properly. Frantisek points out that businesses are always focused on keeping operations running, so the ideal time to address data issues never seems to arrive. As a result, it’s often pushed aside and left unresolved.
"Many companies know what they should do, but they lack the time or capacity to push through major changes. What we show them is that while there’s never a perfect time to clean up their data, the best time is during migration."
One of the reasons data migration often gets deprioritized is that, historically, it was a manual, spreadsheet-heavy process. This approach was tedious, highly dependent on human effort, and prone to errors. "Many of the IT leaders and specialists we work with were consultants before – they’ve done migrations the old way. They remember it as a painful, time-consuming process where there’s little time to manage the migration and data improvements simultaneously," says Frantisek.
But with automation, data migration is now more efficient than ever before. "What I’ve learned from the customers I speak to is that they see the value in data quality, but they’ve witnessed so many outdated approaches in the past – what we call ‘the spreadsheet way’. It’s a process they associate with exhaustion: endless Excel sheets, manual corrections, multiple stakeholders, and a high margin for error," explains Frantisek.
"What we do is provide an alternative approach. We show them that data migration doesn’t have to be painful. It can be structured based on specific data logic, automated, and scalable, making the process smoother, more efficient, and far less stressful."
Best practices for successful data migration
Frantisek says that in the many success stories he’s been a part of, one key factor was prioritizing activities for data preparation from the start. "If a complex process like a new system implementation is to be successful, businesses need to work with their real data from day one – data they’re familiar with. When you work with something you know, you can immediately tell if something isn’t behaving as expected in the new system."
Making sure data is prioritized during the process, Frantisek adds, is a shared responsibility between both customers and partners. "There are many reasons why data might not be prioritized. Customers may not provide the right inputs, and consultants may not ask the right questions because they don’t fully understand the legacy systems. But what’s important is that we do know the system the data is being migrated to. That allows us to guide the conversation and ask the right questions.
"The partner should lead the customer through the process, ensuring they ask the right questions and highlight the importance of future data structure. At the same time, the customer needs to dedicate time and resources – and be transparent about potential issues."
He emphasizes that data is just as important as processes and the ERP system itself. That’s why key users, stakeholders, and business owners all need to be involved.
"These are the people who know the data best. They can provide insights into what’s really happening with the data today and how it should work in the future. It doesn’t take much time to involve the right people early on, but doing so will save you months of unexpected issues down the line."
The second key point Frantisek makes is that not all data is the same. "For example, you might be moving data from one system into M3, consolidating data from multiple systems, or dealing with legacy systems that require significant restructuring. Each scenario comes with its own challenges, so it's important to recognize that data migration isn’t just about transferring information – it’s about ensuring general data consistency which fits the new system properly."
This is why, Frantisek explains, businesses need to think beyond just the data and the system itself. The methodology matters just as much. A well-structured process will determine the right tools to use, when to involve people, who should be involved, and what steps need to be taken next.
"That’s why we say everything must start on day one. If you have a strong process in place, it’ll naturally guide you towards the right tools, the right people, and the right approach. The data migration strategy should be just as firm as the business process strategy. Everything must be aligned to guarantee success," says Frantisek.
Three real-world examples to learn from
Frantisek shares real-world examples of how these strategies have led to successful data migrations. One example highlights a global company with operations in more than 30 countries that successfully managed multiple data migration projects simultaneously.
Each region had strict requirements for testing, user acceptance, full-scale testing, and pricing validation before going live. While the company followed a global template, it also had to account for local exceptions.
"Because our approach leverages scripting and automation, we were able to detect commonalities between different migrations, reuse mappings instead of starting from scratch for each country, and run multiple migrations in parallel with high accuracy,” says Frantisek. “This allowed us to complete multiple migrations in a short time without excessive manual input."
Frantisek shares another example of a company that needed an incremental migration for a fast-moving project. "The customer was migrating from a legacy system to M3 on a tight timeline and required incremental loads to continuously test business processes with real data. We broke the migration into three sprints: defining business processes, managing and cleaning up data, and testing and refining the process.
"By using automation and a common platform, we provided transparency to project managers, data owners, and consultants – resulting in better control and increased efficiency."
Frantisek also describes an ongoing project with a large-scale sales organization that’s automating its migration process across 1,000+ locations. "The company’s goal is to migrate 40 divisions every month. By recognizing that every division followed the same structure, we’re creating a copy-paste migration process. This is significantly reducing manual effort while ensuring fast, precise, and cost-effective migrations."
Gain the benefits of data migration
To maximize the value of your data migration project, you need the right expertise, strategy, and tools in place. Not only does this make the process clearer and more manageable, but it also gives you and your team confidence that the data you’re working with is accurate, reliable, and provides the insights you need.
“We’ve spent several years developing our methodology, learning from hundreds of migrations across industries,” says Frantisek. “We provide customers with best practices tailored to their specific challenges. Our approach ensures that every step of migration is well-prepared, making it as smooth as possible.”
Best practice process for a successful data migration
Frantisek explains that the migration process is designed to be straightforward, even for those without a technical background. “We identify the data we’re working with, decide how to transform it, and then determine where it should go. We start with a methodology, execute the process step by step, and apply specific tools to optimize the workflow.
“Regardless of which stage a business is at – whether mid-project or just getting started – we can help to analyze and propose how to adjust steps to meet typically tight deadlines. This may involve fixing specific data blocks or quickly onboard customers into our standard methodology. We don’t take a one-size-fits-all approach; we adapt based on the situation, timeline, and scope of the problem."
If you’d like to learn more about how we can support your business, get in touch by clicking the button below.