Balancing User Needs and Technology after System Migration

System migration isn't easy, but it can be made easier

Posted on August 11, 2025 in Blog Posts

Just like any major transition, a system migration comes with its fair share of challenges. It’s rarely easy, but the outcome—improving systems and resources for users—can be incredibly rewarding. If you read my previous article, “Balancing User Needs and Technology During Migration Planning,” I hope it offered some guidance for those in the early stages of this process. But as anyone who’s been through it knows, the real work begins after the planning phase. The actual migration and implementation can feel overwhelming, even with the most detailed checklists and preparation. As I said before, “purpose-driven migrations start with people, the users,” and I still stand by that. But I’d like to add that the first actionable steps in any successful migration are taken by the dedicated librarians and information professionals who make it all happen. No matter the scale, a migration is a major undertaking, so take a moment to recognize the effort and be proud of the work you’ve done.

To quickly recap, in Part 1, I outlined nine key categories for migration planning:

1. User Research and Assessment of Needs
2. Contextual Considerations
3. Digital Library Analysis
4. Content Analysis
5. Metadata Analysis
6. Normalization of Metadata
7. Content Migration
8. Verification and Quality Control
9. Migration Resources

These overarching categories are a good way to visualize these key parts of a migration as you prepare for the next steps, but when it comes to beginning the actual migration, no matter if vendor-supported migration services are included or not, these categories will seemingly blur, especially when factoring in technology, users, stakeholders, and many other considerations. By no means do they occur chronologically, especially as you begin the actual migration and implementation of a new system. The planning period before a migration brings a great sense of reflection and aspiration to ensure every detail is considered, from new, desirable features to ensuring minimal risk of moving the metadata and objects to a new system. 

I will warn you—the last sentence above is so eloquently and positively worded, making migrations sound like a blissful and relaxing day at the spa. Just to bring us back to reality, not every detail will be able to be considered, negotiating among priorities will happen, some desirable features will need to be on the back burner, and complex issues will need to be resolved to ensure minimal risk of data loss. From my experience post-migration, there were many things that I had explicitly prepared for with the utmost detail that were quickly derailed, and an endless number of unexpected twists and turns that I could not have predicted. Not to say these aspects from migration planning were not helpful, because I can confidently say that without this planning process, I would have felt much more overwhelmed, but the unpredictable nature of a migration still ultimately prevailed. 

Because of this, I hope my experience and these insights will help others in the field and shed light on the migration process for anyone considering this mighty and exciting transition at their own organization. Those nine categories are still extremely relevant and will shape how I discuss the migration process, with the addition of new aspects not previously mentioned. 

---

🔥 Stay up-to-date with LibTech Insights by signing up for our free newsletter. Just one weekly email with our new blog posts, top tech news stories, and other bonus content. Check out some posts from our archive:

• Social Media and Scholarship as Two Sides of the Same Coin
• An Open Knowledge Partnership: Wiki Movimento Brasil and Libraries
• Giving Up the Good Fight?: Librarians and Information Literacy

🌟 Register now for our free email micro-course on AI literacy!

---

User Research and Assessment of Needs

The use cases gathered during our initial research and needs assessment were essential in guiding the selection of our new platform. This user data also helped shape decisions about which features to implement, how content should be displayed, and which viewers would best support different formats. While this information was incredibly valuable, it had its limits because no two platforms are exactly the same. 

To address the remaining questions, we are planning user experience and usability studies on the new system. These will help us answer questions such as whether users prefer one viewer over another for still images or a different viewer entirely for PDFs. Rather than relying on assumptions, we want to make informed decisions based on real feedback. 

The same approach applies to internal workflows. Our new platform includes a broader range of project management and workflow tools. Instead of assuming full adoption, we are conducting a testing period that includes demos, discussions, and feedback sessions. This allows us to evaluate how well these tools meet our needs and ensures that both user-facing and internal decisions are grounded in evidence.

Contextual Considerations

Collection and user needs were expected to be central contextual considerations, like the size of our digital collections, staffing levels, digital literacy, access policies, and administrative requirements. What surprised me, however, was how often I had to explain or defend these considerations throughout the migration process. These factors frequently became reference points when evaluating whether certain features were appropriate for our environment, even if those features were objectively valuable. 

For example, I was asked why we were not implementing more robust digital preservation features in the new digital library platform. While those features are undoubtedly beneficial, they were not necessary for us because we already have a dedicated digital preservation program, including multiple storage strategies and workflows managed by our Digital Preservation Librarian. 

Another question I received was about why some objects had multiple file types attached, such as a single image with associated TIFF, JP2, and JPEG files. In our previous system, derivatives were automatically generated upon ingest, which enhanced user access and download options. We chose not to delete these files during migration, as they served as backups and we wanted to avoid accidental data loss. Although these files no longer serve the same purpose in the new system, their presence now gives us an opportunity to reassess our storage strategy and better understand how the new platform handles analytics and digital storage metrics. This situation, while unexpected, became a valuable learning experience and a use case for future decision-making.

Digital Library Analysis, Content Analysis, and Metadata Analysis

Analyzing our digital library during the migration was initially challenging due to the significant differences between our previous system and the new one. We faced a steep learning curve just trying to understand the terminology and data model, which were drastically different. Although intimidating at first, this shift ultimately provided a fresh opportunity to engage deeply with the new platform. With so much unfamiliar territory, I had no choice but to dive in, ask questions constantly, and rely on trial and error to test workflows, configurations, and processes. Fortunately, the time I spent learning our previous system inside and out during the pre-migration phase paid off. It made conversations with the vendor much more efficient, allowing me to quickly provide details, like metadata record counts and object types, without delay. 

One key takeaway is that no one knows your system, content, and metadata better than you. Migration support teams can assist, but without a strong grasp of your own data, such as mappings, storage structures, analytics, and functionality, simple questions can turn into lengthy and frustrating exchanges.

It’s also important to think ahead about how you plan to sunset your previous platform. Having a clear strategy for removing content and decommissioning the old system early on allows you to stay focused on the migration itself and keeps your timeline front and center, even as other tasks compete for attention. This kind of planning also helps you look beyond the immediate demands of migration. Large projects like this can feel endless, but setting milestones for post-migration cleanup and transition gives you a sense of progress and closure.

Normalization of Metadata

Metadata normalization during migration is rarely as simple as mapping fields from one system to another. Be prepared to learn and apply new transformation methods, often within tight time frames and in unfamiliar environments or editing tools. While migration support teams can assist, they may not be equipped to handle the nuances of your specific metadata. You are the expert on your collections, and your familiarity with legacy metadata structures will be essential throughout this process.

One of the most helpful strategies I discovered, though admittedly a bit late, was to identify and document a set of 10 to 20 representative records in advance. These should include a variety of material types and examples of complex or inconsistent metadata. Having them ready will save time and provide a reliable set of test cases for transformations, display configurations, and troubleshooting. I found myself scrambling to locate these examples during testing, and I would have benefited greatly from preparing them earlier. These records became my go-to references for validating how metadata would appear and behave in the new system, and they helped surface edge cases that might have otherwise gone unnoticed.

Content Migration

No two migrations are ever the same, and even when you think you’ve accounted for every possible scenario, you’re likely to encounter unexpected challenges. That was certainly the case for us, especially given the size of our digital library. With over 160,000 objects, manually re-ingesting content was simply not an option. We also ruled out the idea of extracting a complete list of objects and metadata and manually matching them to files on a physical drive, because it would have exceeded our timeline and capacity.

Instead, we created a copy of our AWS storage and shared it with the vendor, allowing them to extract the metadata, the associated digital objects, and any relational metadata files directly. While this approach helped streamline the process and reduce the risk of data loss, it wasn’t without complications. For example, we had to copy over everything in the storage environment, including content that was no longer intended for the new platform, such as scholarly materials that had already been moved to our institutional repository. This added extra cleanup work post-migration, but it ensured that nothing was lost in the transition.

Ultimately, the key takeaway is to remain flexible and open to alternative migration strategies. Even the most well-planned workflows may need to be adjusted based on system limitations, vendor capabilities, or unforeseen technical constraints. Prioritizing data integrity and continuity, even if it means more work later, is often the best path forward.

Verification and Quality Control

Careful planning of our verification and quality control approach was key to a successful data review phase during the migration. Before the review began, all team members clearly understood their roles and responsibilities, which helped the process run smoothly and efficiently. The most effective method we found was dividing the data review workload as evenly as possible among the team and using a shared spreadsheet to track progress. This spreadsheet included a checklist to monitor how much had been reviewed, along with dedicated tabs for “Errors Discovered” and “Protocols for Reporting Errors.” These additions were essential for maintaining organization and accountability throughout the process.

This phase was undoubtedly one of the most mentally demanding tasks. Reviewing thousands upon thousands of records and objects was both exhausting and time-consuming. However, the payoff was substantial, and identifying and resolving issues early made a significant difference in the overall success of the migration.

Migration Resources

Throughout the migration process, I frequently revisited my list of potential resources, including APIs, community forums, feedback from other organizations, and various tools. But to be completely honest, if you have access to migration support, ask them everything—no question is too small or too specific. As soon as something comes to mind, reach out. They’re there to help, and their guidance can save you time and frustration.

If you don’t have dedicated migration support, I highly recommend connecting with other institutions that have already gone through the process. I did this myself (even with migration support)—special thanks to the incredible team at Tulane University Libraries who work with digital collections. It was incredibly more helpful than documentation alone. Having someone who truly understands the challenges makes a world of difference.

Outside of the Categories

Across all phases of the migration, several themes emerged that extended beyond the original categories: transparency, frequent and open communication (trust me: the endless emails, messages, chats, and meetings bring comfort and beneficial opportunities to all involved), advocacy for your system’s needs, and a willingness to dive into unfamiliar territory. Whether it was learning new tools, defending contextual decisions, or troubleshooting unexpected issues, success often came down to staying curious, adaptable, and collaborative. Even if a task didn’t seem to fall within your traditional role, embracing it with a learner’s mindset made all the difference. 

One of the most valuable lessons I learned was the importance of being both short- and long-sighted. While it’s essential to focus on immediate needs and functionality, it’s equally important to explore features and skills that may not be implemented right away but could offer long-term value. Discussing both short- and long-term goals with colleagues, stakeholders, and even users helped us better understand the broader impact of our migration decisions. I also recommend expanding your conversations beyond your immediate team, talking to outreach staff, marketing colleagues, and user services early on. Begin thinking about how you’ll promote and support the new system before the migration is even complete. This not only builds excitement but also ensures that your new platform is positioned for success from day one.

A migration is not just a technical process; it’s a human one. While it may be exhausting at times, it’s also an opportunity to grow, connect, and build something better for your users and your institution.

We are still finishing up final go-live tasks, and I am proud to present the end product! Check out CWRU’s new digital library platform, Digital Case!

Related Posts