How does a Dashboard differ from a Visualization?
Data Visualization
Data Visualization aka dataviz is basically presenting raw data in a graphic or visual form. It makes grasping information as well as decision-making easier by presenting data in a well laid-out visual format, making communication more effective, powerful and impactful. Instead of looking at boring tabular data, visuals quickly and clearly give us a sense of data, helping humans understand patterns, trends, relationships and outliers in the dataset.
Data Visualization is an art of showing data in a visual or graphical fashion that makes it easier to digest complex information and make sense of it. It can aid in storytelling. Good data visualizations provide insights and clarify actionables. These can be a guiding factor while formulating strategic decisions.
Benefits of Data Visualization
Visualizations can ultimately be considered data assets that can help an organization grow and scale. The more the employees of the organization understand the data that resulted from their operations/actions, the easier it is for them to make necessary course corrections. Few benefits include:
- Visuals are more easy for the human mind to comprehend and recall. Good visualizations reveal outliers, trends and hidden patterns in data.
- It is not easy to look at rows and columns of information in Excel or tabular formats to see patterns or gather insights. Visuals like graphs and various forms of charts like boxplots, histograms, scatter plots, etc. make it easier to see the data at a glance and understand the gist.
- Interactive visualizations take the power of visuals to a whole new level, by allowing one to drill up and down the dataset, pausing and playing animations and comparing time series data. This can be a great way to show correlation.
- Infographics are yet another form of visualizations that show a ton of information in a one-pager making it easy and fast to gather key takeaways that would have otherwise taken some reading of raw information from myriad sources. Infographics communicate the essence of the topic easily.
Dashboards
Dashboards take data visualization to the next level, possibly adding interactivity and furthering our understanding of the insights we can gain from the basic visualizations. They have become a norm for business communications, organizational decision-making and boardroom discussions alike. Kaplan Norton Balanced Scorecard is based on a similar concept of creating a hierarchy of KPIs, however, more in a textual fashion.
How do Dashboards & Visualizations Differ?
In my opinion a data visualization and a dashboard differ on the following grounds:
- Frequency of updates: A data visualization is static while a dashboard is regularly updated, sometimes in real-time.
- Interactivity: While there could be static dashboards, the well-designed ones are interactive, allowing the audience to see the relationships within data points and explore the dataset themselves.
- Important information vs all-encompassing: A data visualization or infographic may try to squeeze in as much information about the subject or topic. However, the goal of dashboards is to relay the most important information in a cascading style. They are well suited to communicating quantitative data.
- Consolidated information: Dashboards communicate key information that is typically consolidated in a single screen that leverages the power of seeing information and relationships at a glance that would otherwise not have been available in scattered dataset views. Interactivity features help drilling up and down the data hierarchy to see details that are not provided at first glance. They eliminate excessive details that clutter and simplify communication.
- Call for actions or Monitoring: Dashboards primarily serve as a tool for monitoring trends. They are intended to elicit a call for action. This could be both for course correction on downward trends and exploring reasons for positive trends. Drilling down into information provides the specific details.
Does a dashboard have to be a single screen?
Stephen Few describes a dashboard as “A dashboard is a visual display of the most important information needed to achieve one or more objectives, consolidated and arranged on a single screen so the information can be monitored at a glance.” He asserts that:
- dashboards are meant to consolidate multiple pieces of information to a computer screen,
- they need not be comprehensive,
- should narrowly limit information with the primary intent to provide an overview of data.
Some thought leaders say that dashboards should not have any scrolling. While I do agree that left to right scrolling is a no-no, I am not averse to top-to-down scrolling.
There are differing views on single-screen at-a-glance dashboards vs multiple screen dashboards. A simple answer that applies in the majority cases is having a single dashboard. Isn’t that what differentiates it from a report?! Most dashboard best practices and conventional wisdom stress on showing only relevant data, help piece-out important insights, cut confusion, and be effective. Too often, I come across many dashboards that follow the rule of one-page and jam too many small graphs on a screen that are hard to read, leaving no room to show data connections or explanations.
In my opinion, each has its own place with pros and cons associated with each approach to leverage visual real estate. For example, a CEO can find a single dashboard giving a complete global overview and picture of the business revenue, profits and losses combined. While this information can be drilled down to specific lines of business (LOB), there is value in each LOB having its own linked dashboard with KPIs that might be relevant to that vertical or business. Nowadays, we also see platform dashboards on Facebook, Twitter, LinkedIn, YouTube, Amazon, etc. There are dashboards that show multiple KPIs to those that show a single number or two (for example, health apps). In some cases dashboards have to limit the size of a mobile app (Uber, gaming apps, etc.)
Designing Dashboards Key Elements to Consider
The design of the dashboard (single or multiple) needs to be framed based on:
- the data context,
- audience,
- intent of visualization,
- purpose it will serve,
- frequency of communication,
- Level of interactivity/responsiveness,
- type (strategic, operational, tactical, analytical, informational),
- and the depth of information it will communicate.
Multiple screens work well for analysis to derive gold nuggets of information that could then be explained in a single-screen dashboard to Executives, who do not have time to sift through multiple screens of information. Based on the audience, multiple screens may sometimes not be effective if the audience can’t follow the relationships, data connections and fail to draw conclusions or make comparisons. Ultimately, what is needed to do the job is to be qualified and addressed. Dashboard experience will determine if the consumers continue using them or not.
