Week 10— My Journey into Data Analytics — DA Minidegree Review — CXL Institute
Welcome to the 10th Week of my Digital analytics journey with CXL. In our previous articles, we learned about Data collection and analysis. This week we dive into “Why data visualization and Data storytelling is the key to actually make a change”
Once you have successfully collected the Data, it is crucial to present it to the main stakeholders to take action. A dashboard, graph, infographics, map, chart, video, slide, etc. all these mediums can be used for visualizing and understanding data. In this article, we are going to understand the importance of presenting Clear, Engaging Data with a Story and What are strategies and Principles to abide by irrespective of your visualization tool to generate compelling Presentations or Slidedoc with the intention of impactful actions.
Understand Brain science of Communication
There are three types of memory that play an integral role in how people will remember your brand.
1. Iconic Memory
2. Short-Term Memory (STM)
3. Long-Term Memory (LTM)
According to Miller’s Law, our short-term memory and absolute judgment are both limited to number 7 on average. We can only remember 7(+/-)2 chunks of information. So the first impressions of your brand should take this into consideration and make sure that the consumers are not overwhelmed with information. To enter someone’s Long Term Memory, your Brand Needs to make an impact either through emotions, storytelling, or Visuals.
Decrease the Cognitive Load
Cognitive load is the amount of active mental processing needed to complete tasks.
There are two main types of cognitive load, intrinsic and extraneous.
- Intrinsic load refers to the complexity of the subject matter or content itself. The more complex the information, the more things need to be processed in working memory.
- Extraneous load refers to how the information is presented. Too fast? Too much jargon? Too dense? This is where data visualizations can help lighten the cognitive load. But it’s also where presenters can become distracting or make things more complex than they need to be.
You may not have control over the intrinsic complexity of the data, but you have complete control over how you present it. Knowing how to use data visualizations effectively can keep you from overloading your audience.
Pie Charts are failing Us
Pie charts look effortless to comprehend but surprisingly make it very difficult for the human eye to estimate the magnitude of angles, thus making visualization not very clear. In addition, pie charts require more labels to get the measurements across, making them not very concise. In short, often they defeat the main purpose of data visualization.
What is a better alternative? A Horizontal Bar Chart. They make the comparison among categories is easier by the length of the bars. Also, Color coding and legends are not required and Direct labeling is easier.
Note: Rather than having different colors for each Bar, Limit it to one color for Background Bars and One Colour for Highlighting data such as Peaks or Present Year Statistics. You can also use Lining and shading as an alternative for color.
Note: In absence of a corporate Palate, use the color wheel
Tips for Horizontal Bar Charts
- The Bars should be wider to avoid confusion ‘
- Avoid stacked bars because no Common 0 Axis makes it difficult to compare data
- Avoid Multiple series in a single Bar Chart
Maximize the data-Pixel (Ink) Ratio
The Data-Ink ratio is a concept introduced by Edward Tufte, the expert whose work has contributed significantly to designing effective data presentations. Tufte refers to data-ink as the non-erasable ink used for the presentation of data. If data-ink would be removed from the image, the graphic would lose the content. Non-Data-Ink is accordingly the ink that does not transport the information but is used for scales, labels, and edges We should Eliminate all extra components such as redundant information, including background color, borders, and grids.
Include a zero baseline
Although a line chart does not have to start at a zero baseline, it should be included if it gives more context for comparison. Sometimes wrong trends or relationships between metrics are reported due to changes in the axis.
Select the right chart form
it is crucial to know the strengths and limits of chart forms before deciding to use them to represent data. Let’s go over some of the common chart forms:
Bar Charts — They Depict nominal data. They are often used to illustrate comparisons. The value axis should always start at zero and use a consistent scale.
Pie Charts — They depict slices of a whole. There should not be more than five slices, and their total value should equal 100%. The largest slice should start at the top and fall to the right.
Scatter Plots — They depict relationships between two variables. They are effective with large datasets and highlight patterns or correlations. Danger: the reader may assume a cause-and-effect relationship between the X and Y axes even when there isn’t one.
Line graphs — They depict patterns over a continuous range. They do not have to start at zero, but the data ranges should be clearly marked and the overall chart shape should retain a 5:8 height and width ratio.
Sparklines — A sparkline is a tiny graph, they give you a clear and compact graphical representation of your raw data adjacent to it. They are best for showing a quick data trend, which can later be explored.
Write a Compelling Data Story
Storytelling with data differs from data visualization because it requires communicators to offer a larger, holistic, view of their message. You must focus first on your audience and structure a larger message before any visuals are rendered. You must identify from the start:
- What do I want my audience to know or do with the data I am presenting?
- How will I structure a narrative that leads to the desired action?
- How is my data helping drive a decision?
A Data without a story lacks attention and the care to take action about it.
Use Mckinsey title for creating Actionable Slides
Mckinsey titles follow a rule i.e The slide should contain 1 Keypoint and 1 Supporting Point. For Eg: If we want to convey about the decrease in bounce rate, rather than Writing vague Titles such as “Bounce Rate comparison according to Device”, we should mention an actionable title i.e 40% Decrease in bounce rate on Mobile. This ensures that the key information is conveyed and can be worked upon by the stakeholders.
Your Presentation or Slidedoc should follow
- Vertical Flow — Each slide has a clear title and the content of the slide only serves to prove and reinforce the point.
- Horizontal Flow — The sequential titles actually fit together and tell a cohesive story.
Lastly, Revise and edit your presentation
The more you design, the more you learn. It is important to give yourself time to design and to revise and edit, over and over again.
