A statement on social injustice. Click to read more.

Crafting effective data visualizations

By Nate Thompson
Nate Thompson
Data Scientist, Customer Loyalty and Analytics
Aug 7, 2020
4 MIN. READ
The power of well-made charts for customer insights

At first blush, quickly capturing important business information with data visualization seems like a clear-cut illustration of the phrase “A picture is worth a thousand words.” A well-designed visualization can show key insights at a glance that would take a page of prose in a report. Armed with that insight, stakeholders can make informed decisions about what’s right for their business. This is especially true for customer marketing program leaders looking to learn how they can better deliver value to the core business.

But it can often feel like the analyst is communicating data visualizations in a language the audience doesn’t speak. At best, this leads to confusion that requires clarification instead of the visual standing on its own. At worst, it can lead to the sort of miscommunication that happens when a word in one language sounds like a very different word in another language.

This means that an analyst’s ability to communicate in their audience’s “language” is as critical a skill as being able to craft visualizations that, on a technical level, accurately convey the data. Fortunately, modern data analysts have several tools designed to help them build clear and compelling visuals.

When it comes to data visualizations, less is more…

Often, an analyst tries to fit every useful bit of information into one graph. This can result in stakeholders being stuck playing “Where’s Waldo?” with the key insight rather than getting a deeper understanding of the data. When this happens, the analyst has fallen short of their goal: help the client make the choice that’s right to meet the needs of their business. Instead, they would have been better served by a clear visual that draws focus to the important parts and trims anything extra.

When the critical information is a handful of KPIs, a simple table can be clearer than trying to warp the data into a graph. That way, each KPI is given the space it needs to stand out instead of competing for the audience’s attention. Even a table can be stylized into a visual that captures the business’s attention and clearly communicates what they want to know.

For example, take this KPI summary from a theoretical dashboard:

ICF’s global marketing services agency focuses on helping your organization find opportunity in disruption.
Go to ICF Next

These indicators answer a few critical questions: How are we doing on these four KPIs? How does that compare to last year? These questions could have been answered with line charts showing the growth of these metrics over time, but that doesn’t do anything to make it easier to answer those two questions. All a more elaborate graph would add is clutter that stakeholders would need to sort through to answer their questions.

Additionally, the metrics are chosen to support a few key insights. Revenue provides a peek at the overall health of the business, and the other three KPIs illustrate how well the theoretical program has done at bringing in customers and encouraging them to spend at the company.

…except when more is more

Other times, analysts choose to add more—grouping related insights in other ways to reinforce them. For example, take this chart of email deliveries, a key metric for CRM systems:
It adheres to the “less is more” philosophy very well, but in the process loses a fair bit. Deliveries dropped from December to January, but by how much? The lack of clear labels makes it difficult to determine just from the bar sizes. Also, CRM stakeholders have other related metrics that would make sense to pair with this graph. For example, let’s say the open rate has dropped in the last month. If we could pair those numbers with deliveries, we’d be able to better understand whether our existing audience is losing interest (stable deliveries) or if our attempts to reach new customers aren’t going as well as hoped (increased deliveries). Understanding which story is more accurate is critical to helping improve the future direction of the CRM program. Finding a way to bring deliveries and the click-to-open rate into one chart makes it easier to tell that story clearly and compellingly. Let’s combine the two, along with several other related metrics:

The updated title helps point the user toward what they should get out of the graph. Adding labels to the values makes comparison quick and easy. Combining the delivery count with the related rates makes answering the question posed by the title straightforward.

The trick to a chart like this is in the details. Switching the color of the bars from a bold blue to more subdued grey keeps it from drawing too much attention away from the lines. Since this is a dual-axis chart, attention must be paid to scale. And because this chart compares millions of email deliveries to three percentages, it’s easy to give each an appropriate scale that accurately represents the data and the relationship between the two groups.

Visualization as narrative

Data visualization is one of the best tools for translating abstract data into something stakeholders can act on, much as a story conveys its themes with concrete plot and action. A mindful analyst can use compelling visuals to tell a story that is clear and concise in a way that all their audiences can understand. For successful customer marketing programs, that kind of insight is crucial to stakeholder decision making.

By Nate Thompson
Nate Thompson
Data Scientist, Customer Loyalty and Analytics

The latest marketing trends, uncovered.

Subscribe to get insights, commentary, and news sent straight to your inbox.