Data visualisation

Some data visualisation that I designed.

Disclaimer: Due to non-disclosure agreements with BNP Paribas, I am limited in the amount of work I can show.

Introduction

As I worked in a data science team, data visualisation is one of the key deliverables. I used the tools like Tableau and PowerBI to build dashboards. I also use prototyping tool to define new visualisations based on different user needs.

My Role

In this kind of projects, I worked as both UX designer and implementer.

Process

I worked with the data scientists who know the data very well and business analysts who gather the business needs of our clients.

Normally, I first need to know who use our solution, and what are the data that are needed. As in a ToB project, it is usually hard to talk to the real users (mostly are sponsors, and clients), so we need to compromise between the client's goal and the users' needs.

It is suggested to have something visual to present so that we are sure not to waste the business users' time.

Due to the confidential issue, I will only show some pieces of data visualisation with the fake data through different use cases.

Information architecture & workflow

Leasing Inside Information architecture

Information architecture of the use case Leasing Inside, a visualisation tool that seeks to increase leasing business through more efficient commercial targets based on cross-shared data. It hinges on a prospect scoring model and an engagement amount estimation model that give more direct visibility to leasing sales.

1/3

Data visualisation UI

Alignment metrics for sectors	Alignment metrics for sectors	Social media data Overview
Landing page Portfolio alignment Dashboard	Data Quality Dashboard Comparison between two data sources

Take away

Instead of conducting usability testing after finishing the mockup, we need to present to the internal clients regularly the results of the machine learning models so that we can optimise the model. In this case, I learnt that the users of the machine learning models are more interested the data information and the data quality rather than the visual representation.

Therefore, my work was to focus on how to represent the data in a more understandable and intuitive way instead of how to make it just beautiful. Besides, it is also important for a UX design of data to understand the logics of different data sources and the meanings behind the data.