Dashboards are more than just a collection of graphs and tables – they offer a great way to visualize your data for analysis and to gain invaluable insights. However, dashboarding platforms often come with various options for displaying data, along with free reign to structure the dashboard to one’s will. The options can be overwhelming! Yet, the ways in which data is visualized and how the overall dashboard is structured significantly influence end-user adoption and impact the quality of analysis.
This article aims to provide best practices and key considerations for structuring dashboards and visualizing data. While it focuses specifically on dashboarding for performance marketing analytics and Salesforce Marketing Cloud Intelligence, the insights can be applied to other dashboarding or analytics tools as well.
Understanding Your End User
You can construct a dashboard by simply creating graphs and pie charts to display your data; however, to make a great dashboard that guarantees end-user adoption, you need to consider what a typical day in the life of your end user looks like.
Before you build your dashboard, ask your end user what a typical day looks like for them, what key channels they often use or need to log into to access the insights they require, and what processes they follow to get to the information they need. What kind of questions do they need to answer? What questions are they being asked but cannot answer, and why? By understanding these needs, you can help identify the key channels, KPIs, and dashboards that your end user really needs.
For a marketer, this typically looks like having to log into the platforms of key marketing channels to monitor and understand the performance of their ongoing campaigns. Unfortunately, this often means monitoring up to 20 different channels, which also makes it difficult when campaigns span multiple channels. As such, a dashboard should offer a one-stop shop where marketers can just log into a single platform to access all the information they need to monitor and understand their campaign performance, logging into the individual channels only when action is required.
Understanding Your Data
KPIs
In addition to considering the overall workflow of your end user, it is also good practice to understand the KPIs you are visualizing, what they represent, and how they relate to one another.
With this information, you can build a general pipeline of where these KPIs sit in terms of evaluating a campaign’s performance. Metrics such as reach and impressions capture the general measurement of the audience that has seen the post. Metrics like opens and clicks indicate forms of engagement or interest from your audience. The ultimate goal and interaction of a conversion can be captured through click-through rates, purchase conversions, and revenue.
This leaves us with this data structure: reach and impressions, opens, clicks, click-through rate, conversions, and revenue. Then followed by additional performance evaluation metrics, such as media cost, cost per click (CPC), cost per action (CPA), and cost per mile (CPM), as these are derived measurements from the volume metrics captured from the channel source.
It makes more sense to view these derived metrics after understanding the base volume performance. Organizing any visualizations in this data order will help marketers conduct smoother, more concise analysis that fits their workflow.
Additionally, consider whether it makes sense to display a certain measurement as a whole number or a decimal. For derived and rate measurements, such as conversion rates or costs per click, decimal values are useful for capturing minute but relevant changes. However, for volume measurements such as clicks, opens, and impressions, it may not make as much sense (even though some channels do provide them in decimal form). For example, values like “0.25” or “1.8” clicks may not make much sense.
Data Relationships and Hierarchy
Developing a good understanding of the hierarchy of your data and how it relates to one another is crucial to building an effective dashboard. By understanding the hierarchy of your data and how it relates to one another, you can better determine what fields should be available for filtering and even whether that data is better displayed in a flat or pivot table.
While the ability to filter a dashboard is a key feature that many analysts enjoy, it is tempting to overdo it by adding every dimension as an available filter. This may overcomplicate and bombard the analyst with filtering options and may lead to confusion.
A good dashboard should be easily shareable to audiences without needing a complicated explanation for its use. By applying dimensions that are higher on the data hierarchy as filters, you enable analysts to break down these groups into smaller parts in the dashboard, which is more intuitive than requiring analysts to add up and attempt to identify their own groupings of data categories should you apply a filter option that is further down the data hierarchy.
This is also relevant when considering displaying data in a table format, as displaying data against the data’s natural hierarchy will result in tangled and chaotic-looking data. Rather, structuring your data by hierarchy allows you to conduct your analysis on various levels. Displaying your data accordingly should also allow you to group your data in an organized manner.
Looks Do Matter…
As a final touch, consider harmonizing the color palette of your dashboard. Standard dashboard configurations often supply random colors to graphs. Coordinating the colors of your dashboard may seem like a minor task, but it has a significant impact on user experience (UX), adding the final touch – a cherry on top of a well-crafted, user-centric dashboard.
Tools like Salesforce Marketing Cloud Intelligence offer branding options that allow admins to upload their company logo. Intelligence then identifies and applies the matching color palette to the entire dashboard, creating a more unique and personalized view for the end user.
Summary
To build an effective dashboard that guarantees end-user adoption, one must understand how their end users intend to use the dashboard in their daily workflows.
Familiarizing oneself with the data being visualized, including its data hierarchy and relationship with one another, will also aid in deciding how the data should be presented. Lastly, adding those final touches, such as ensuring color consistency, can make a big difference in the user experience.