1. Bringing it all together
Hey there! Throughout this course, you've explored the datasets and built impressive data visualizations. Now it's time to assemble everything into a neat and actionable dashboard, but before we do that, let's study some important concepts!
2. Dashboarding vs. storytelling
Data and insights which we skilfully turned into powerful visualizations can only bring value when exposed to the outside world: to our stakeholders. An interactive dashboard is one of the typical deliverables of any data analyst skilled in data visualization. It is meant to be self-explanatory and serves as a self-service analytics tool for non-data profiles. Rich with filters and buttons, it is excellent for any exploratory data analysis or browsing through large volumes of data.
A Data Story is a different type of deliverable in which attention is put on insights, often drawing focus to concrete data points. Enhanced with a narrative, such as visible on the slide, stories are equivalents of typical slide decks or presentations and contain the essence, the key findings.
3. From worksheets...
It all starts with individual charts or tables prepared in separate worksheets. Individual graphs are then assembled into
4. From worksheets to dashboards...
thematic dashboards, often containing information that complements each other thanks to the interactivity elements such as filters and actions. For example, if we now enabled filtering on the map presented at the top of the dashboard, we would have simultaneously triggered a filter on all related dashboard objects.
5. From worksheets to dashboards and data stories.
Stories are used to highlight the most interesting data points. They act like handy bookmarks, keeping all selections ready for the presentation but remaining fully interactive, allowing us to filter, zoom in, hover over to discover the information in the tooltip, and so on.
6. Looks matter - pretty and simple catches the eye!
Designing a good dashboard or a story is, in fact, a time-consuming exercise in which every detail matters. Appealing design, taking into consideration company branding, but also an efficient use of space and attention to detail, such as adding self-explanatory labels and titles, can make or break even the most carefully prepared dashboard.
How to not get overwhelmed? Consider starting small, incorporating just a few objects and sheets to your dashboard, which you can, in steps, consult with your non-data stakeholders. Test your filters and parameters meticulously and provide clear documentation. Remember that even one confusing combination of selections and filters can render a lot of questions from future users.
7. Design for any device
Even the prettiest dashboard will suffer if displayed on a screen in the wrong resolution. As dashboard builders, we need to account for a situation in which a user opens our dashboard on a different device than ours, for example, on a mobile phone.
Preparing the dashboard for multiple resolutions and screen types (for example, tablets, smartphones, and laptops) is an extra step that a careful dashboard designer takes to ensure the optimal user experience. An alternative approach would be to design the dashboard to fit the worst-case available space. Consider setting a fixed-sized dashboard, for example, fitting the size of a PowerPoint presentation or a generic laptop screen resolution.
8. Don't lose your audience!
The visual aspect is not the only thing we need to consider when developing a dashboard. We need to keep a close eye on the performance. Generally, any load time exceeding ten seconds is considered too long. To mitigate it, try to reduce the number of data marks in your visualizations by pre-filtering them. Another option is to create a data extract. If you want to know more about these topics, check out the DataCamp course Connecting Data in Tableau!
9. Let's practice!
Ready to roll with the practical part?