Why
Acceleration In Growth
What
A 360degree AI/ML Enabled Intuitive Customer Platform
How
Machine Learning based recommendations for Cross-sell and Up-sell
Solutions
Old application was non-action oriented which was purely informative, actions were required to be triggered in the respective applications/core systems like the lead generation in CRM. The client wanted to establish new Customer 360 cloud-native platform to provide a solid foundation for creating universal services and integrations which can be reused and leveraged across the bank as building blocks. The platform should support non-technical configuration, should be highly scalable, performant and enable rapid releases.
Team Xebia in collaboration with the client developed a universal service across the bank which integrated many other applications and had a revised, reimagined, advanced Cust360 with analytically derived data points at single location that gave direct insight into actionable analytics. Team created a foundational platform which provides customizable role based 360-degree access of customer information. This includes demographics, psychographics and product portfolio details of bank’s customers. Each service typically implemented a set of distinct features or functionality and communicates across each other through a well-defined, lightweight mechanism to serve a business goal. It provides fault tolerance, load balancing, tracing and logging, and health monitoring.
Key features of the application were:
Xebia’s pre-sales team performed detailed analysis and identified the core components needed to architect the platform and functionality. Hardware capacity planning estimation was carried out based on projected usage of the system for the next five years. Project estimation and capacity plan was drawn up to complete the project. Relevant technical training was provided in order to incept the concept of containers and Kubernetes within the bank. Methodology that was devised to implement the solution were:
- Development ran alongside infrastructure and DevOps setup.
- Development was carried out in Agile fashion.
- UX design ran ahead of core development.
- Microservices architecture style was adopted to run on elastic container orchestrator.
- Big data jobs were designed to provide data for corresponding services.
Development and delivery using Scrum framework
Key focus areas of the development & delivery were around three ceremonies and the key activities have been highlighted for each ceremony that not only involved the Scrum team but also some stakeholders.
 |
Product Backlog Refinement |
During product backlog refinement team performed deep dive of upcoming features to understand the user needs in detail. Wireframe and visual designs were created and finalized after several discussions. User stories were created or further refined (if already defined) along with detailed Acceptance criteria. Technical discussions on user stories considering design, NFR, database, schema, data contract was finalized. Backlog refinement and Estimation was done. After user story review from business stakeholders; final sign-off was taken. Compliance was ensured with Definition of Ready.
 |
Sprint Planning |
Team capacity calculation, Sprint goal draft, Sub-tasks creation and estimation, Sprint backlog finalization and Sprint goal commitment was the major agenda of Sprint planning.
 |
Sprint Review |
Share sprint goal and Sprint backlog, Demo of done functional stories, Demo of platform, DevOps, QA Automation and review of release progress along with next sprint backlog planning were covered in the Sprint review.
The technologies implemented were as follows:
- Setup of on-premise Kubernetes clusters.
- Design & development of Java Microservices based platform.
- Setting up of the DevOps pipeline: Automated Deployment & Release Orchestration on Kubernetes.
- Each microservice can be independently deployed, scaled, and monitored for KPIs across multiple environments.
- Test Cases and Automation of Microservices to enable regression testing.
- Performance Testing using JMeter.
- User experience design via wireframes.
- Responsive application to support multiple browsers.
There were two dimensions to the solution:
- Product Management
- Development & Delivery using Scrum framework
Product Management Approach
Challenges and associated solutions
The client was in banking domain so there were many challenges related to security & compliance. Also, the client was new to the Agile product development so many challenges were on the adoption of agile practices. Following is the summary of key challenges and approach taken to resolve the same.
Challenges |
Solution |
| It was fix bid contract for MVP so; scope, cost & timeline was fixed. But as project was growing, delay in committed timeline started occurring. | At the end of every Version Report (Release Burnup) of JIRA was reviewed and deviations were discussed with the client. Since timeline was fixed, so low priority features were identified which can be descoped for post MVP. |
| Considering banking domain, there were many issues in data sharing and code deployment, due to which there were many external impediments which needs to be resolved quickly to be on time. | Daily alignment call with key stakeholders to resolve these dependencies. A Risk and dependencies tracker created for close monitoring of such issues. |
| Ensure completeness of business requirement of MVP. | A customer dashboard created in JIRA for requirement traceability. |
| Complexity in Platform setup, Working on many new things like K8s, ISTIO for the first time. | A separate team created for Platform and DevOps setup to monitor their impediments and goal more closely. |
| Coding as per banking guidelines and compliances. | Regular code review session planned with the client. |
Results/Key Benefits
Value delivered to Organization |
Value delivered to Business |
|
|