The best migrations don’t just move workloads. They move organizations forward.
Migration is often described in technical terms: workloads moved, systems retired, data transferred, users onboarded. Those measures matter. But they do not tell the whole story. The real value of modernization shows up when organizations use migration to rethink how they work, scale and prepare for what comes next.
That idea comes through clearly in two SAS customer stories: Posten Bring, a Nordic postal and logistics company with nearly 400 years of history, and Regeneron, a leading life sciences company building a modern statistical computing environment. Their industries, operating models and modernization paths are different. Yet both arrived at the same conclusion: successful migration requires strategy, discipline, user trust, a clear reason to change and a strong partnership with SAS.
Posten Bring: Move the organization forward, not just the systems
For Posten Bring, modernization was a continuation of a much longer story of adaptation. Founded in 1647, the company has reinvented itself many times. In recent decades, digitalization reduced mail volumes by about 80%, while logistics became a larger part of the business. Data became central to how Posten Bring improved operations, increased production efficiency and competed in a changing market.
SAS has supported Posten Bring since 1999, but the company’s SAS 9.4 on-premises environment had reached its limits. Capacity constraints limited real-time capabilities, weaker API support and rising demand from users who needed faster access to better data made modernization necessary. As the Posten Bring team put it, SAS 9.4 had helped get them there, but it could not take them further. Their next step was to modernize with SAS Viya for streaming data, integrated analytics, a fast modern database, and scalable cloud with SAS Managed Cloud Services.
Warning: This video shows Posten Bring deleting a SAS 9.4 production system during a Microsoft Teams meeting. Please do not try this at home, or at work, unless the migration is complete, the results are validated and maybe a stress ball is nearby.
The work was complex. Posten Bring moved 20 years of legacy code, thousands of ETL jobs across 35 BI solutions, hundreds of SAS Enterprise Guide projects, and tens of terabytes of data from on-premises systems to SAS® Viya® with SAS Managed Cloud Services and SAS SpeedyStore. And the technical work was only part of it. The team also had to upskill hundreds of users and change how people worked so they could take advantage of the new Viya platform.
Along the way, the team worked through data migration challenges, integrations, network and firewall requirements, character and collation issues and the practical question of what should move. The project also showed the strength of the SAS partnership. When serious issues surfaced, the teams worked together to solve them and adapt the software to support Posten Bring’s business needs.
Several lessons extended well beyond the migration itself:
- Require every migrated workload to have a clear owner. If no one owns it, do not move it.
- Start data and integration work early, especially when moving from on-premises to cloud.
- Plan for format, collation, character setting, networking and firewall challenges.
- Onboard users early, even when feedback is tough, so the team can correct course before the migration scales.
The results show both technical progress and organizational change. Legacy ETL workloads ran faster on Viya. Users saw even greater improvements when they optimized their own projects. The company moved to full version control and DevOps-style deployment practices, with 1,100 commits to its production repository in 2026 alone. It also onboarded 1,400 users, more than 10% of the company, to actively use SAS Viya for analytics and reporting.
Posten Bring’s migration mantra is clear: migration is not just about moving systems. It is about moving the organization forward.
Regeneron: Strategy drives execution
Regeneron’s story reinforces the modernization theme from a different perspective. The company set out to “accelerate data to insight” by building a modern, multilingual statistical computing environment around SAS Viya. The goal was not simply to replace SAS 9.4 or SAS Grid. It was to create a broader analytical ecosystem designed for the way regulated life sciences teams need to work today and in the future.
|
|
Regeneron grounded its approach in five principles:
|
In clinical research, governance is not an add-on. It is fundamental to protecting sensitive patient data, supporting auditability, and proving that outputs were produced in a controlled, compliant way.
Regeneron also recognized that users need choice without fragmentation. Its unified analytical platform combines SAS Viya with Posit and a common shared storage layer so programmers and data scientists can work across SAS, R, Python, and future tools without duplicating data or weakening governance. The idea is simple but powerful: one governed data foundation, multiple best-fit tools and a consistent user experience people are willing to use all day.
For Regeneron, AI should not sit off to the side as another disconnected tool. It should work naturally across SAS Viya, Posit and future tools, using secure common protocols so users can ask questions, get answers and receive recommendations without leaving the governed ecosystem.
That strategy-first approach shaped the migration itself. Before moving workloads at scale, Regeneron invested in:
- Stakeholder alignment across IT, business, and statistical programming leadership.
- Rigorous planning, including detailed Gantt charts, risk and pitfall analysis, and mitigation strategies.
- Validation strategy and change management before workloads moved at scale.
- Early user engagement with business SMEs.
- A “test small, scale big” approach using sample workloads in development before production.
- Close coordination across IT, business, storage, CloudOps, SAS, implementation partners and other technology providers.
The technical work was substantial. The migration required changes dataset encoding issues, PROC IML image-generation workflows, Enterprise Guide-to-SAS Studio change management, log-line limits, POSIX permission challenges and validation requirements across AWS, storage, SAS Viya, R packages, and third-party components. In a GxP-compliant environment, every component that data touches must be considered, validated, and governed.
By the numbers, Regeneron had migrated hundreds of users and expected to reach 400 by the end of the journey, with millions of files and roughly 90 terabytes of legacy data in scope. The roadmap began with strategic planning in 2024, moved through non-GxP and GxP platform go-lives in 2025, launched pilot and wave-based migrations in 2026, and is expected to culminate in a modern, multilingual, modular, AI-enabled, cloud-native, and GxP-compliant statistical computing environment heading into 2027.
Regeneron’s migration mantra is equally clear: strategy drives execution. Or put another way, design with intention, execute with discipline and scale with confidence.
What these stories teach us
Posten Bring and Regeneron approached modernization from different starting points. One story centers on resilience, adaptation and bringing a long-standing organization forward. The other centers on strategic platform design, governance, and future readiness in a highly regulated environment.
Together, they point to the same conclusion in that migration is not the finish line. It is the mechanism organizations use to become more adaptive, more governed, more scalable, and ready for what comes next. That requires more than moving code and data. It requires clear ownership, engaged users, executive commitment, disciplined planning, trusted partners, and the willingness to rethink how work gets done.
The organizations that get the most from modernization are not simply asking, “How do we move what we have?” They are asking, “What do we want to become?”
That is why the best migrations don’t just move workloads. They move organizations forward.
