How Softsell Modernised JustSA's Legacy Retirement Planning Platform
About JustSA
JustSA is a leading retirement planning services provider based in South Africa, serving thousands of pensions and retirement fund members across the country. Operating within a highly regulated financial services environment, JustSA is committed to digital excellence and sought to modernise its technology estate by aligning all enterprise applications with its Microsoft-centric technology strategy.
Customer Overview
- Customer: JustSA
- Industry: Financial Services — Retirement Planning
- Solution: Legacy Platform Modernisation
- Customer Location: South Africa
- Project Duration: 9 Months
- Scope: Legacy Application Modernisation, Cloud Migration, Automation Testing, Support & Maintenance
Executive Summary
Challenge: JustSA’s Retirement Planning Portal had outgrown its PHP foundations. Aging architecture, split technology stacks across teams, undocumented integrations, and a growing maintenance burden were limiting the organization’s ability to scale, secure, and evolve the platform.
Solution: Softsell executed a full modernisation of the platform — migrating from PHP to .NET, restructuring the data layer to Azure Table Storage, resolving critical adviser session and proxy account architectural blockers, and embedding automated QA from Sprint 1.
Impact:
- Legacy PHP platform successfully migrated to a scalable, maintainable .NET application.
- Adviser session and proxy account architectural blockers fully resolved, enabling smooth user flows across all integrated applications
- Passed independent penetration testing with commendation from senior JustSA IT leads and verified by external Pentesting provider
- Eliminated the overhead of two separate technology stacks, consolidating onto a single Microsoft ecosystem
- Future-ready, cloud-native architecture enabling faster delivery of new business features
The Challenge: A Platform Holding Back a Business
JustSA’s Retirement Planning Portal has served the organisation well for several years. But by the time the modernisation initiative began, it had become a constraint rather than an asset.
The platform ran on PHP — the only major application in JustSA’s estate not built on the Microsoft technology stack. That single divergence had compounding consequences: two separate development teams, two support structures, and a growing knowledge gap as the rest of the organisation moved forward on .NET.
Beyond the technology divide, the platform itself was showing its age. Key integration points with enterprise tooling, CRM and messaging infrastructure, and Azure components were undocumented and deeply coupled. Security vulnerabilities had accumulated in the aging codebase. And the data layer — built for a simpler era — was not designed to scale.
JustSA needed more than a rewrite. They needed the platform rebuilt on a foundation that could grow with the business, integrated cleanly with the rest of their technology estate, and secure enough to meet the compliance expectations of a regulated financial services environment.
The Solution: Four Workstreams, One Unified Platform
Softsell structured the engagement around four parallel workstreams, each addressing a distinct dimension of the modernisation challenge.
PHP-to-.NET Migration
The PHP-based platform was migrated module by module to .NET. Each component was individually assessed, mapped, and re-implemented to ensure functional parity before sign-off through both automated regression testing and UAT. The migration preserved all existing business rules and user experience patterns — users experienced continuity; the codebase underneath was transformed.
Cloud Data Platform Redesign
The existing SQL-based data model was restructured and migrated to Azure Table Storage. This was not a lift-and-shift. It required a deliberate redesign of the data architecture — including partition key strategy and entity schema — to leverage cloud-native NoSQL capabilities and maintain query performance at scale. Moving this early in the project paid dividends in reliability throughout.
Architecture Discovery & Integration Mapping
Before any code changes were made, the team conducted a structured deep-dive to map all upstream and downstream integrations — covering messaging infrastructure, CRM systems, SSO, and Azure components. This investment was not optional: without it, the team was operating blind to the dependencies that would later prove most critical. It proved foundational to everything that followed.
Embedded QA & Automation from Sprint 1
A Selenium-based test automation framework was built in parallel with development from the very first sprint. QA engineers were embedded throughout, providing continuous regression coverage and catching migration-related issues before they reached production. For a project of this nature — where business logic is being re-implemented rather than built fresh — regression coverage from day one was not optional.
Learnings: What This Project Taught Us
This project surfaced challenges we did not fully anticipate at the outset. The following learnings reflect real problems we encountered, how we worked through them, and what we would do differently next time.
Undocumented Architecture Hides Critical Complexity
What Failed
When we entered the project, no formal architecture documentation existed — only milestone-level tickets were available. As the team explored the codebase, they uncovered deep and largely undocumented dependencies across the Retirement Planning Portal, integrated enterprise tooling, CRM and service bus infrastructure, and multiple Azure components.
The most significant blockers emerged around two specific areas: adviser–client session handling in Azure and the management of proxy adviser accounts. Both caused session loss and inconsistent navigation across interconnected applications. Advisers operating on behalf of clients could not reliably maintain session state, and proxy account routing was producing incorrect navigation paths — causing broken user flows that were difficult to reproduce and trace.
Why It Happened
The session and proxy logic had evolved organically over years without being formally documented or centralised. The behaviour was implicitly encoded across multiple services and integration points. Without upfront system mapping, the team began making changes without fully understanding the downstream impact on session state and cross-application routing — which led to rework.
How It Was Fixed
The team paused active development and conducted a structured discovery exercise, mapping all upstream and downstream systems — including CRM messaging infrastructure, the service bus layer, SSO flows, and Azure session management — before resuming changes.
We implemented a stable adviser–client session model in Azure capable of persisting multiple concurrent client sessions per adviser. For proxy accounts, we designed explicit routing rules for default and proxy-based navigation values, ensuring consistent behaviour across the platform and all integrated applications while maintaining authentication integrity throughout.
Outcome
All adviser session and proxy account blockers were fully resolved. The solution subsequently passed independent penetration testing, with senior JustSA IT leads specifically commending the robustness and security of the implementation — requiring only minor adjustments before sign-off.
Key Takeaway
Deep-dive system mapping of all upstream and downstream dependencies — messaging infrastructure, SSO, CRM integrations, and session management — must happen before any code changes are made, not after issues surface. Aligning proxy scenarios, SSO behaviour, and cross-platform navigation in the design phase would have prevented the rework we encountered mid-project.
Cloud Data Migration Requires Upfront Architecture Investment
Migrating from SQL to Azure Table Storage was not a lift-and-shift exercise. It required a deliberate redesign of the data model — including partition key strategy and entity schema — to avoid cloud storage performance anti-patterns and maintain query efficiency at scale. Teams that approach this as a technical task rather than an architectural decision risk building a cloud-hosted system that performs worse than the one it replaced. Investing in data architecture planning early was critical to a smooth and performant migration.
Early QA Involvement Safeguards Business Continuity
Embedding QA engineers from Sprint 1 and building Selenium automation alongside feature development proved invaluable. The automated regression suite caught migration-related issues early — particularly around business rule preservation — reducing the risk of functional breakage reaching production and giving stakeholders confidence throughout each release. Projects of this nature, where business logic is being re-implemented rather than built fresh, require regression coverage from day one, not as a final phase activity.
Results
- Legacy PHP platform successfully modernised into a scalable, maintainable .NET application.
- Eliminated the overhead of two separate technology stacks, reducing operational complexity across development and support teams
- Resolved critical adviser session and proxy account architectural blockers, enabling smooth and reliable user flows across all integrated applications
- Passed independent penetration testing with commendation from senior JustSA IT leads, requiring only minor post-test adjustments
- Improved application security and compliance with modern development and coding standards
- Established a clean, extensible architecture enabling faster delivery of future business features
- Enhanced scalability and cloud readiness through Azure-based services and Azure Table Storage integration
- Meaningfully reduced long-term maintenance effort and operational overhead through full technology consolidation
- Provided JustSA with a future-ready platform fully aligned to its enterprise Microsoft technology strategy
Conclusion
The JustSA modernisation engagement was more than a technology migration. It was an architectural reset — one that cleared years of accumulated complexity and positioned the organisation to move faster, scale more confidently, and operate more securely.
By consolidating onto a single Microsoft technology stack, resolving deeply embedded integration challenges that had never been formally documented, and embedding quality engineering from the very start, JustSA gained a platform capable of supporting the business’s evolving needs — not just today’s requirements.
The results speak directly to what matters in a regulated financial services environment: a system that is secure, reliable, maintainable, and ready for what comes next. Softsell is proud to have delivered that outcome for JustSA and remains a committed partner in their ongoing technology journey.
Tech Stack:
- Platform: .NET
- Frontend: CSHTML, JavaScript
- Database: SQL Server, Azure SQL
- Cloud Platform: Microsoft Azure
- Data Storage: Azure Table Storage
- Test Automation: Selenium
- Quality Assurance: Manual & Automated Testing

