Digitalization of consulting processes

What is still considered a vision of the future in many banks became reality in a compact, eight-week project. One of our largest clients decided to test the potential of digitalization in the advisory process – with a clear objective: “Buy or make?”

Initial situation

Personal advice remains a central element in banking – at the same time, it is one of the last major areas with untapped digitalization potential. To address this, a small, experienced team of software developers and product owners was formed. ONLU was allowed to provide an additional person to support this team. The aim was to develop a functional digital consulting process within just eight weeks.

Result: Surprisingly positive

The implementation was carried out with high productivity, minimal disruption and very short communication channels. Despite the ambitious timeframe, there was no negative stress. Particularly noteworthy: even initial skeptics were convinced – so much so that a larger follow-up project was decided shortly after completion.

The most important steps in the project:

A central goal of the project was to use technologies that guarantee long-term flexibility. For this reason, open source-based and standardized solutions were deliberately chosen in order to avoid vendor lock-in

1. requirements analysis & technical understanding

Some discussions with specialists took place before the project started. Specific questions were asked during implementation. The focus was on a specific form of consulting – with a clear scope and without any claim to completeness of all possible use cases.

2. system architecture

The architecture was deliberately kept simple:

• Spring Boot-based web application as a specialist application

• Second application as Camunda wrapper, with specialized APIs

• Angular frontend, connected to REST interfaces

• Use of 1-2 databases (MongoDB & MySQL)
  This architecture was ideal for a proof of concept, but asynchronous processes would be given greater consideration for productive operation.

3. backend development

• Technologies: Java 17, Spring Boot, Camunda

• Implementation using the API-first approach with clearly defined REST interfaces

4. front-end development

• Use of the customer’s existing design systems and Angular components

• Result: Massive time savings – around a factor of 3 according to the front-end team

5. database design

• MongoDB for the specialist application: no fixed schemas necessary, one document per consulting case was sufficient

• MySQL for Camunda: Existing Helm setup (incl. statefulset, PVC, deployment) could be reused from previous project

6. testing & quality assurance

• Unit tests with JUnit/Mockito

• System tests with karate

• Goal: Ensure stability despite high speed

7. DevOps & CI/CD

• CI/CD via GitLab pipelines

• Deployment on OpenShift via Helm charts

• Already in use from day 1 – no major challenges

Conclusion

This project impressively demonstrates how much can be achieved in a short space of time – when an experienced, focused team works together with clear objectives and the right tools. The pragmatic architecture, the targeted use of modern technologies and the agile way of working not only led to a successful proof of concept, but also paved the way for a long-term, larger follow-up project.