The CXL Data Visualisation course covers several different topics around specific concepts, techniques, and even specific visualization types. It highlights the importance of stringing together multiple data visualization with imagery, with text in the right way to decrease cognitive load for the viewers. Every marketer conducts research, analyses data, and presents it to the team in the hope of leading to actions, hence effective data visualization becomes an important part. It is not about making the data pretty but making it understandable, which leads us to the topic below.
Highlighting the importance of "Curse of knowledge" - The curse of knowledge is?a cognitive bias?that occurs when an individual, communicating with other individuals, unknowingly assumes that others have the background to understand.
The curse of knowledge means that the more familiar you are with something, the harder it is to put yourself in the shoes of someone who’s not familiar with that thing. You can’t unlearn what you’ve learned, and you can’t see it with fresh eyes anymore. Plus, you have a much harder time explaining the basics to people who are new to the subject because you can’t remember what questions you had when you were new to the subject.
The Curse of Knowledge can affect your business - Imagine coming across a brand that provides a solution you do not understand, even after being explained you are left wondering what it does, how you would use it, and why you would need it? This is likely a classic case of the curse of knowledge.
How to break the curse of knowledge
- Get a fresh perspective- After working on a product or service for a while, you will find it impossible to have a fresh perspective. The only way to assure that you’re on the right track is to get your idea, wireframe, prototype, and product in front of new users before it's released.?It’s also helpful to get the perspective of fresh eyes encountering your product for the first time. What information do they need to know to understand your offering? What concerns or objections will they have right away?
- Know your audience- Identify the right target customer preferably a customer who is already aware of your industry or company, this will allow you to not water down your language to make the customer understand. Make sure you address the questions and concerns of the less-experienced customers while making it easy for more advanced customers to quickly scan through any information they don’t need.
- Leverage Human insight from new users- To get a non-biased perspective ask any?new employees?in the first two weeks to provide feedback about your product or service.
First-time site visitors?are the perfect fresh eyes for testing out your web copy and your value proposition. Run a remote usability test with people in your target audience, and ask them to:
- Describe what your company does in their own words
- Point out what, if anything, is confusing about your messaging
- Explain any additional information they would need before they were ready to make a purchase or sign up
- The brain size of communication-?Understanding the basics of memory
1) Iconic Memory (Pre-cognition) -?captures everything you see, it passes information to the short-term memory. Also called the visual sensory register.
2) Short-term Memory- The tiny little subsets that the iconic memory grasps as important are sent here. It has a small capacity, as described by George Miller, called "Seven Plus or Minus Two". Which indicated that our short-term memory can hold seven plus or minus two pieces of information.
3) Long-term memory-?thoughts and things that the short-term memory store for longer are sent to the long-term memory.
Since short-term memory has a limited capacity it is important to present data that requires less cognitive load for the viewers.
- The Gestalt Theory/Psychology- Humans perceive patterns or configurations, not merely individual components and this relates so well with the Millers' low- the limited capacity of the brain to look for patterns. The Gestalt principles include- proximity, similarity, closure, connectedness, and continuity.
- Data Visualisation - The importance of axes- Avoid using multiple axes in one chart-
1) Instead try creating two charts, to decrease the cognitive load. Easily be misinterpreted. Use multiple charts with one axes instead.
2) As a general rule, makes sure to use 0 based y-axes.
- Data Visualisation- Horizontal bar charts for comparing categories
1) Horizontal bar charts are best for comparing metric different categories. Verticle bar charts are tricky, as it does not allow longer text to be able to get wrapped properly. It just ends up looking like hanging texts at the bottom of the graph. However, horizontal bar?charts support more natural reading and we can easily read the text from top to bottom.
2) As we are already used to reading text from top to bottom this decreases cognitive load. if working with multiple metrics encourages easy comparison of multiple metrics. (insert image)
- Ensure the width of the bar (or set of bars) is greater than the space between the bars.
- Avoid stacked bar instead use different bar charts to showcase the data
- Avoid multiple series
- Use a bright color to add accent, the base level chart does not need color so use grey.
- *If you are working with bar charts, trying to minimize the mental stress it causes by dividing into different bar charts.
Data Visualisation - Line charts
- Line charts are best for showing trends over time. The slope of the line between two points shows the pattern of change. Avoid using line charts with categorical data. Categorical data might be put together in order of greatest to least, but they are not connected.
- It is important to avoid showing more than 2-3 lines on a line chart, as it's hard to read.
- Avoid using a legend in a line chart and directly label each series if possible.
- Avoid labeling every data point, not adding too many numbers, if you wish to talk about the number use a table or an interactive form of display. Adding dots on the line charts is another way of saying let me add more to make it look better.
- **Use the dot selectively on the line chart just to highlight the points you want to, instead of turning on things in mass so that is on the line chart.
- Line charts do help decrease the cognitive load and if used wisely you can be in really good shape.
- Data Visualisation- Sparklines and small multiples
What are sparklines and how and when are they most effective?
- Its little sparklines are added together with data without axes. It's a compact way to show intra-period trends for a metric. But it does not show the actual scale or values. Sparklines are not perfect but they provide a lot of data in a compact form.
- Sparklines are available in every platform and small multiples help quickly see which trend is upwards or downwards. Small multiples shine when you do them across dimensions. Small multiples can be used for any type of chart (they do not have a whole lot of precision or detail and are foreign to the brain) so we need to use them carefully.
- Data Visualisation - Text as a visualization
Text is the original data visualization, stand-alone text can be a very powerful visualization. Numbers are a "visualization of data." Hence, the bigger/larger the font, the more of a?"visualization"?the numbers become.
- Use significant font size differences to differentiate between a key metric and context metrics. It is easy to just present the information in one bold sentence which has not a lot of distraction.
- The best tip so far is to not use pure black on your text as it looks harsher. Back it off to 95%.
- Use shades of dark grey instead.
Data Visualisation- Heatmaps
- Heatmaps are graphical representations of a metric across two dimensions (even though they may vary quite a bit) using color or shading. They can be used to improve the readability of a table and geographical plot.
Data visualization - Speciality charts and chart elements
- Standard charts = less/lowest cognitive load
- Non- standard chart types, increases cognitive load but we do it with purpose
Data storytelling matters
- Before you start telling your story, you first need to consider whom you’ll be telling it to. Think about what makes them tick, where their interests lie, and how best to connect with them. To win an audience over, you have to understand where they’re coming from, and then connect with them on an emotional and personal level.
- When considering your audience remember that different team members have different objectives and different points of view. Effective data storytelling should speak to these differences.
- First things first, to tell a clear and compelling story, you need to separate the signal from the noise, and choose the right data for your needs:
- Identify the subsets of data that represent the specific points you want to convey.
- Remove any extraneous data that isn’t imperative to your story – showing too much information makes it hard for readers to spot the insights you want them to see.
- Use metrics and naming conventions that your audience will recognize – things like capital expenditure, change in sales, or time to hire.
Draw attention to key 3 information
- To be an effective data storyteller, you need to direct your audience’s attention to the most important points within your data. If your graphs are dense and don’t emphasize what matters, your audience will likely have trouble grasping the point of your story. They may even come to the wrong conclusions.
- In short, you can’t expect your audience to know where to look or what to think — you need to show them. Fortunately, a few simple design tricks can help you draw attention to where it’s needed most