Presenters
Source
Revolutionizing Disability Benefits: How MongoDB Powers a More Accessible PIP Service 🚀✨
In the UK, millions rely on the Personal Independence Payment (PIP) to navigate life with disabilities or health conditions. Historically, accessing this vital benefit meant a daunting phone call, lengthy paper forms, and a process that often added stress to already challenging circumstances. But a transformative journey is underway, spearheaded by DWP Digital and Capgemini Invent, leveraging the power of MongoDB to create a more accessible, user-friendly, and scalable digital service.
The Human Story Behind the Tech 🧑💻❤️
Before diving into the technical marvels, it’s crucial to understand who this service is for. Kathryn Sylvester, a Product Lead at DWP Digital, shared a poignant quote from a claimant: “You have to be in the right headspace or work up the confidence to begin the application.” This single statement encapsulates the core challenge.
The Old World: A System Designed for Difficulty ☎️⏳
- The Front Door: A phone call, only available during working hours.
- The Headspace Hurdle: Requiring significant mental energy to even initiate contact.
- The Time Sink: Approximately 20-25 minutes to start a claim over the phone.
- The Paper Deluge: A 50-page paper form mailed out, taking about 2 weeks to arrive.
- Lost Momentum: By the time the form arrives, the claimant’s initial motivation often wanes, especially for those who struggle with momentum.
The Scale of the Challenge 📈🌍
- Around 4 million working-age people in the UK claim disability or health benefits, with PIP being the largest.
- The department spends approximately £60 billion annually on these benefits.
- PIP receives 20,000 to 22,000 new claims per week.
- This service supports 4 million vulnerable individuals who depend on it monthly.
The Catalyst for Change: COVID-19 🦠➡️💡
The pandemic accelerated a long-held ambition to modernize PIP. With face-to-face services closing overnight and support networks disrupted, the need for a digital alternative became urgent. What began as a crisis response in 2020 has evolved into a multi-year plan to rebuild and repurpose how people apply for, evidence, and track their benefits digitally, at a national scale.
The Product Vision: Empowering Users to Tell Their Story 🗣️✍️
The PIP application is structured around 12 daily activities (e.g., preparing food, managing treatments). Applicants are asked to describe how their condition affects them. The overwhelming feedback from users was clear: they wanted to tell their story in their own words, at their own pace, without the constraints of checkboxes or dropdowns. This desire for personal expression is the guiding principle for the digital service.
The Engineering Backbone: MongoDB’s Role in Scaling 💾🛠️
Ibrahim Tamam, a Technical Lead at Capgemini Invent, explained how MongoDB makes this user-centric approach technically feasible, especially when scaling to millions.
The Problem with Rigid Schemas 🧱❌
Traditional database architectures often force users to “flatten” their complex needs to fit a rigid schema. This can exclude individuals with unique circumstances, like multiple overlapping health conditions and diverse evidence. Ibrahim emphasized, “A schema that can’t bend itself, that can’t be flexible, is a schema that’s excluding its users.”
MongoDB’s Flexible Document Model 📄✨
MongoDB’s document model keeps the entire application, with all its activities, events, and evidence, as a single, coherent object. This eliminates the need for complex joins and allows for:
- Seamless Data Representation: The application’s structure mirrors the user’s story.
- Agile Policy Changes: Government policy changes, which are frequent, can be implemented in days or weeks without lengthy migration programs.
Scaling Strategies with MongoDB 📈🌐
MongoDB provides key patterns for scaling this critical national service:
- Horizontal Sharding ➡️🔄: The application itself serves as a natural shard key, meaning data is distributed efficiently. Each application is self-contained, minimizing cross-shard penalties on the “hot path” (frequently accessed data).
- Multi-Availability Zone Resilience 🛡️⚡: For a service supporting over 4 million users, downtime is unacceptable. MongoDB offers out-of-the-box automatic failover with replica sets, ensuring no lost time and eliminating the need for manual firefighting.
- Schema Versioning (Document Level) 📜⬆️: When schema changes are necessary, they happen at the document level. Each application can understand the version of the schema it was written under, allowing new code to read older data seamlessly. This means no “flag day” migrations where users are informed of downtime. The system simply adds a version to the document, and the software adapts.
The Strangler Fig Pattern: Evolving Without Disruption 🌳➡️🏢
Instead of a disruptive “big bang” replacement of the legacy system, the Strangler Fig pattern is employed. The new digital service grows around the old one, piece by piece, claim by claim, region by region. This pattern, combined with MongoDB’s document model, event-driven integration, and strong observability, makes the transformation traceable and enables:
- No Downtime for Claimants: Users on the legacy system can remain there while new users transition to the digital service.
- Gradual Migration: The full migration can occur over several years as the program concludes.
A Powerful Partnership 🤝💼
This transformation is a testament to the collaboration between Capgemini Invent, DWP Digital, and MongoDB. Capgemini’s expertise in legacy-to-MongoDB migrations, coupled with their strategic partnership in MongoDB’s AI ecosystem, brings invaluable experience and public sector delivery muscle.
The New World: Accessibility and Dignity 🏠☀️
Returning to the claimant’s story, Kathryn painted a picture of the new reality:
- Anytime, Anywhere Access: The claimant can open the service online from their own kitchen, at any time of day or night, on a day of their choosing.
- Flexibility and Comfort: If they get tired, they can close their laptop and return later when they feel better.
- Support System Integration: A daughter can help with parts of the application, uploading photographs and documents.
- Personalized Storytelling: The online form allows them to describe their activities in their own words, in the way they choose.
On the invisible backend, this entire human interaction is captured as a single document in a MongoDB database, preserving the entire shape of the claimant’s story without flattening or loss.
Key Takeaways for Building Critical Services 🎯💡
Ibrahim and Kathryn summarized three crucial lessons:
- Database Choice is an Accessibility Decision: When building services people rely on for their livelihoods, the database choice isn’t just a technical concern; it’s fundamentally about accessibility.
- Modernization is an Evolution, Not a Big Bang: Critical national services are best modernized using a Strangler Fig pattern, evolving alongside the legacy system to ensure zero downtime and no negative impact on users.
- Focus on Human Impact: Technical metrics like throughput or cost per transaction are important, but the ultimate measure of success is building something that allows a person, on a difficult day, to tell their own story. This is the true meaning of scaling a service like PIP.
This project represents a proud achievement, demonstrating how technology, when thoughtfully applied, can profoundly improve the lives of millions.