The idea
The inspiration of the financial statement visualisation using Sankey comes from an idea of having an intuitive view of the flows in the financial statements. During my previous work, there was much time and efforts spent on understanding and analysing the flows in financial statements analysis merely based on numeric figures. In this situation, what an intutitive view of the flows in financial statements provides is to improve the work efficiency and make the deliverables more interactive and attractive to clients.
For my own interest, the one I usually put much attention on is the income statement to evaluate the profitability of a business, where I can witness the inflow of assest into the business and the outflow incurred to generate the revenue. Hence, the visualisation here starts with the visualisation on the income statement.
Additionally, I aim to focus on the performance by geographic segments for the sample business selected here. Therefore, when it comes to the operating income inflow, it was splited into the operating income from various locations and regions.
The process
- Data Source
The data source comes from UOB Group 1H22/2Q22 Results. And the data mainly comes from two related financial documents:
- Design Process
- Tool and platform selections
- Data Visualisation
I started the visualisation by using the Google Sheet for the purpose of automation and convenience of sharing. To achieve this, the income statement was imported into the Google Sheet and an extension, ChartExpo was installed and used here. By using ChartExpo, it can be functioned like the pivot table with selection of metrics and columns and this extension can deliver the Sankey diagram automatically.
However, the issue with visualisation in Google Sheet using ChartExpo is that the metric can only select one column, which means that some key figures cannot be splitted twice. Additionally, the negative values cannot be visualised and the extension takes all figures as a sum-up.
After encoutering the issues that will have effects on the key functions of the visualization, Figma is selected here for the visualisation. Even though it cannot transport the figures directly into the Sankey diagram, i.e. drawing the diagram is needed, it does ensure that the visualisation output meets the expection with quantity and quality.
Besides a financial statement analysis followed the regular routine, as mentioned, one focus of the visualisation will be the performance by geography. Hence, the operating income is composed of the operating incomes from difference locations and regions so that the readers are able to compare and evaluate the performance in various locations.
Moreover, two key figures are selected here to give additional summarized information to readers. The one is oversea contribution and the other is operating profit from Singapore. This splits the operating profit into two portions, Singapore and overseas, considering the headquarter of United Overseas Bank (UOB) is in Singapore.
To Improve
The visualisation of the financial statements here is to provide a efficient way for the readers especially the beginners to financial statement analysis to understand the flows and to evaluate the performance in a straightforward way.
What can be improved next are:
- Figures inside the visualisation.
The figures now used in the visualisation are those imported from the financial statement in a specific period. For better evaluation, it is also common to have a comparison cross years or quarters. Therefore, the figures can be updated to year-on-year figures or quarter-on-quarter figures to visualize the movements during the financial reporting period.
- Automation of the visualisation.
As mentioned above, the Google Sheet was abanboned eventually for the visualisation because of some functions in ChartExpo to be improved. Hence, for now, the visualisation need to be drawn and designed mannually. This gives the flexiability of adjusting the design template with the changes of the business models, however, it will be more efficient and concise if there are tools to help implement the data input.