Visualisations are the preferred way of communicating with everybody, not only data experts. Visualisations can be used to start a conversation. They can show the situation as it is, where difficulties are situated, and how policy decisions can change and improve the situation.
Visualizations are well-known in the context of monitoring and reporting. Once a (policy) decision is made visualizations can be used to monitor the changes (or lack of changes) that occur after the implementation of the decision. Visualizations also aid in communicating the results of the decision to peers that were not or less involved in the decision but are interested in the results.
But visualizations can also be used to define a question, make a question more precise, and can be used to thoroughly analyse the results. To be able to support a policy question with data, it is necessary to translate the policy question to a data question. This transformation can lead to new insights and help make the original question more precise. Later on, when the results have been obtained, visualizations can help understand the results and analyse them further.
Because of this it is important that visualization tools are easy to use, generic, flexible and able to integrate different data types. Domain experts should be able to understand the question and interpret the results based on the visualization. If a visualization methods is generic it can be used for different questions, posed by different departments or even cities. Visualizations are also a work in progress.
Visualizations should also be flexible. They can continuously be improved to ensure that they are easy to comprehend and reflect the situation of the policy problem without bias. Constructing the perfect visualisation is a challenging and time consuming task. The visualisations should not be too complex and “only” based on the most relevant data. Visualisations can continuously be improved until a final form is reached, and regardless of whether they are in their final form or still being improved, they can always be used to aid and guide the conversation about a policy problem.
Finally, visualizations tools should be able to integrate different data types. Data is gathered through a variety of tools and methods, each generating output in their own way. A generic, flexible and easy to use visualization tool should take this into account and be able to integrate different data types.