Sketching out a dashboard is one of the most important early steps in any dashboard building process. Before jumping into Tableau (or any tool for that matter), taking the time to plan out your structure can save a lot of time and potential rework down the line. It helps you, the creator, figure out what charts to use, where they’ll go, and how users (e.g. clients) will interact with the dashboard overall.
In consulting, this step is especially valuable. Sketches can serve as a communication tool between you and your client, giving them a clear visual idea of what to expect in terms of layout, interactivity, and features before a single chart is built in the software. That means any big changes or feedback can happen early, rather than halfway through development.
Now, let’s be clear, these sketches don’t have to be polished works of art. In fact, they probably won’t be (mine definitely won’t be hanging in the Louvre anytime soon). But they give you a head start, clarifying which elements you’ll need, what the layout might look like, and how the overall flow of the dashboard will feel.
Here below are a few of the initial sketches I drew for this dashboard (they’ll display in a rotating view every 5 seconds)
Disclaimer: definitely not my cleanest work, but hey, they got the job done!
My initial idea was to create a landing page where I can display key information within the dataset at a high level for the target user. It then became quite apparent that the data did not provide enough depth to create all the separate targeted dashboards I had envisioned, so I opted to go for a simpler but cleaner 2 page dashboard (technically 3 but you’ll see why later) design where I build in visual KPIs as well as a sales/profit overview based on the data’s groupings. There would then also be an order table page, where the user could experience a more in depth overview of the dashboard on an OrderID level.
There was also a change in regards to the filter and navigation panes. Originally I envisioned this to be along the top of the dashboard, but as the structure evolved, I shifted this to a sidebar on the left with a filter button that reveals options when clicked. This felt more intuitive for users and allowed for better use of space in the visual layout.
Here’s a mock-up of what the dashboard might have looked like if I had used a tool like Excalidraw at the time to refine the final design (spoiler: I didn’t know about it yet!). Instead, I relied on my pen and paper rough sketches, which worked fine, but if I were to do it again this is where I’d have brought in digital tools to get an even clearer idea of my designs.
Now, post creating the dashboard (and having since been at the Data School for a while), I realised that I had built it in a not so optimal way (using floating tiles rather than containers - another shameless plug check out why we shouldn’t do this here). To rebuild the dashboard more efficiently, I created a Technical Sketch. While a design sketch shows what the dashboard should look like, a technical sketch focuses on how it should be built. It outlines layout structure, interactivity, and functionality in a more detailed way to support development.
And that's it for the sketching part of the dashboard journey. The next blog in this series will look into how I set up the dynamic parameters within the dashboard and built the graphs for the overview page.
Thanks for reading! 😊
Victor