Melissa Ursi, chief executive of Quoris, discusses the hidden bottleneck in NHS enterprise resource planning implementations.
As NHS organisations advance major digital transformation programs, such as enterprise resource planning (EPR) deployments and optimisations, one challenge is increasingly creating a bottleneck for implementation timelines and outcomes: migrating mountains of documents and metadata from disparate legacy systems to create one unified patient record.
While platform selection, interoperability, and clinical adoption are usually top of mind when discussing digital transformation, many organisations experience a very different operational reality once projects begin. Extracting, transforming, validating, and preparing legacy clinical content is one of the most underestimated risks in any transition. When overlooked, legacy systems don’t just slow implementations; they create governance challenges, derail timelines, and quietly drive up project costs. In many cases, what’s treated as a downstream task becomes the very thing that determines whether a project stays on track at all.
Problems with legacy systems
Legacy data migrations vary widely across departments based on a multitude of factors, such as how many management platforms were used, which speciality-specific applications need to be addressed, or whether we’re working through scanned and paper archives. Document conversion and metadata mapping can become a gating factor when converting unstructured or loosely structured documents and data into formats usable within EPR systems. Complicating things further, GDPR and retention requirements can create headaches about what needs to be migrated, archived, or destroyed.
In so many cases, extracting usable data from these environments can quickly shift from a simple transfer to a full-blown reconstruction. While technology and AI have advanced significantly in automating portions of this work, meaningful outcomes still require human oversight. Ongoing, iterative refinement is essential to ensure clinical usability, data integrity, and alignment with evolving compliance requirements.
Issues during the data conversion process rarely stay contained. They create cascading downstream effects that delay migration timelines and jeopardise go-live readiness. To address this, organisations must elevate migration planning from a supporting activity to a core workstream, with the same level of visibility, rigour and ownership as the broader implementation.
At this stage, many feel like they’re on track; they’ve completed discovery, the migration plans are defined, and timelines are crystal clear. It’s at this moment that the bottleneck typically arrives.
Legacy providers are usually the first place customers turn when they want their data extracted. With high demand for extractions and few alternatives to get the project done, legacy providers are often the most expensive option, and their inability to start the project due to deep wait lists or backlogs is causing significant delays for hundreds of clients. Implementation keeps moving forward, but the go-live date keeps getting pushed further away while data waits on conversion work beyond the team’s control. Meanwhile, organisations wait in line and end up paying for unnecessary maintenance costs.
The lesson is clear: if extraction and conversion are secured early, with the right expertise and capacity, you’ll have a more accurate timeline that keeps things on schedule.
Scale changes the maths
The scale and condition of your legacy data can significantly impact your ability to meet the target completion date. In the discovery phase, it’s important to note and catalogue factors such as volume, variety, sensitivity, and the condition of the legacy data before committing to timelines. This early investment will contribute to a smoother overall project and avoid costly rework later.
A minor metadata issue across a few hundred records is frustrating, but still manageable. Across millions of documents, it becomes a milestone-level adjustment. The most successful programs tend to decouple migrations from the critical path whenever possible. This is done through phased migration strategies, categorising high-value clinical data, and creating hybrid access models that keep legacy systems accessible to care teams for a defined period. This helps instil confidence and gives the team the time needed to become comfortable with the new system, ensuring reconciliation and quality control are maintained. Having the right experience and expertise from both technical and clinical leaders can significantly improve the efficiency of the process and accuracy of the forthcoming migration efforts.
Lessons learned
Across large-scale digital transformation projects, there are lessons that are becoming increasingly clear. First, poor data quality doesn’t resolve itself during a migration process; it has to be actively managed throughout and after the decommissioning of the system. Second, there is no success without early and frequent IT and clinical engagement, not only for adoption but also for validating meaningful and useful improvements to the data. Third, don’t underestimate extraction and conversion as a scheduling risk. They can create clinical, operational, and costly risks.
As technology continues to advance and NHS organisations accelerate digital transformation, those that recognise legacy data migration as a strategic priority rather than a technical afterthought will be better positioned to deliver on-time implementations.
Organisations that secure specialised extraction and conversion expertise early are better positioned to move on the timeline they’ve set. Those who don’t often find themselves waiting in line, falling behind, and absorbing unnecessary cost along the way.
