Data Visualisation: Why is it an important skill for a data architect?
Several months ago we took a look at what it means to be a data architect. Today we’d like to take a closer look at one aspect of a data architect’s skill set that can often be overlooked - data visualisation.
What is data visualisation?
Data visualisation is the presentation of data, often in graphical form, that makes it easy to understand, explore and communicate. Whilst this might be thought of as more under the purview of analysts, it is also important for data architects to understand. This is primarily for two reasons: data architects need to understand how data is used in order to best design the systems that will store, process and publish this data, and data architects need to communicate their own designs and processes to a wide range of stakeholders, often in the form of visualisation.
Why is it an important skill?
Understanding and designing outputs
In our previous blog (see above), we discussed how data architects need to be able to see the big picture and understand all aspects of the process by which data is used within their organisation or project. This includes the visualisation of data. Data architects need to understand how data will be displayed and communicated in order to best design their systems. Will the data be primarily presented in tables, will it be graphs, or will it be in the form of a varied and detailed dashboard. Knowing the eventual outputs will help considerably in good design of the system. Given the somewhat fuzzy boundaries of what a data architect does, there may even be projects where we need to design such user focused visualisations ourselves, and as such, knowledge of them is crucial. A recent project we worked on primarily involved designing a data dashboard to communicate data insights to other members of the organisation. Thus knowledge of user focused visualisations is important for data architects, either for understanding how the data in their systems will eventually be used, or for directly designing such outputs themselves.
Communicating designs
The second reason data visualisation skills are important for data architects is to communicate their data architecture designs to a wider team. Any data architect should be able to build an Entity-Relationship Diagram (ERD) to communicate the structure of data in tables and the links between them. Data flow diagrams (DFDs) are important for communicating how the data will flow through the systems, how it will be transformed and where it will sit. Heatmaps are a very important tool for understanding and visualising existing data, where data is concentrated and illustrating usage patterns. Sankey diagrams are useful for showing how data points flow through a system. We have made use of these ourselves in several recent projects. For example, in our SatsWales work we used ERDs to communicate our database designs and heatmaps to examine the ‘fullness’ of the data, this helped us to demonstrate to other members of the team our understanding of the data and our reasoning and process behind our designs, thus helping to facilitate good service design. Whilst in our work with MoJ, Sankey diagrams were used to examine the flow of applications and answer key questions. As such, we consider all these to be useful for data architects to communicate their ideas and designs to a wider team, and to help them better understand existing data storage and structures.
Summary
In summary, data visualisation is an important tool for data architects both to understand how data will be used in the systems they are designing and to effectively communicate their designs to a wider audience.
Author
Tags: