The Smarter Devon programme have invested in User Experience upskilling for the workforce so we can ensure we are creating data that is clear to understand for every skill. We put the end-user at the front of every design decision we make. We look at how a user feels when using a product, system or service. By designing the experience positively, we can increase user engagement, and build more effective services. We have captured an outline of our process below to get you thinking about how you could apply User Experience to your data projects.
The Government Data Quality Framework says that understanding user needs is essential to ensuring that data is fit for purpose. User Experience (UX) for analytics has helped us at Devon County Council to fulfil the data quality framework’s requirement to research and understand your users’ needs, then prioritise efforts on the data which is most critical.
The first stage of the process is to have a focus group. This group should include anybody who is likely to use the report you are building (external, or internal). It is a good way to set the scene for what you are trying to do and explain that you want to work collaboratively with the report’s users so they get what they want.
You want to understand current processes and issues, as well as aspirations.
Define
Some people struggle to speak in large groups, so this is a good opportunity to conduct some individual interviews also giving you real insight to the purpose of the project you are wanting to start. Conducting a one-to-one interview with a user can help you define which business questions they need answered. In the session, it is good to also talk through any existing reporting they have, so that you can understand any gaps and what they like or don’t like. We want to minimise a user’s ‘cognitive load’ (or effort exerted during a task) and decision-making time. Users will stop looking at a Power BI report or digital product if it takes up too much of their time.
Top Tip – At the end of the session, ask the interviewee if they would be prepared to help with testing a prototype report to see if their original captures have changed.
Design and validate the prototype
Once we have an understanding of the purpose of the project, the data translators create a mock up of a Power BI report using PowerPoint. Once a mock-up is complete we tend to test it with a few individuals. At this stage, small changes can be made to create and ideal design. Once we have a prototype, we are all in agreement, and it has potential to answer the objectives, we do a final show and tell to the original focus group.
It is very important that you explain to people that any mock up is an illustration not a promise!
Build and test
Usability test sessions usually take up to one hour, depending on the number and complexity of the tasks we want users to attempt. During the interview we give control of the reports to the user, whilst we walk them through a series of tasks based on the business questions defined during the requirements stage. The development team will watch the participants looking to complete specific tasks using the reports. Asking the users to “think aloud” as they move through the reports helps understand what the users are doing, thinking, and feeling – all of which will enable feedback that helps improve the reporting. Any changes to the report must go through the testing process again before sign off.
During these test sessions you are able to see the user working their way around the report. This highlights design flaws or where extra context is required to accurately interpret the data. Some visualisations may look pretty, but are quite difficult to use! Conducting this build and test phase ensures we have a report that delivers the project’s objectives ensuring the data is fit for purpose.