Data Viz 101: The Essential Guide to Data Visualisation

Hello everyone! I will share data storytelling with Tableu topic at brightonSEO . Therefore, I want to share one of my favourite topics in detail as a key component of the powerful data stories. Here's the data visualisation in all aspects!

While I took a course a few years ago from CXL Academy, I met this topic. Then, I figured out how to use Looker Studio by myself and started to create dashboards. However, that was not enough. I would say that data visualisation plays a crucial role when we’re presenting data. In my master's programme, I took classes about all kinds of data stuff and Tableau. After I read books about this concept, I got an AHA moment many times.

I realised everyone has to take a look back at their data graphs because of the importance of clear delivery. Not only I am passionate about this topic, but also I wanted to prepare you to explore the foundations of my MeasureFest talk and I prepared this data viz 101 guide for you. Let’s deep dive into the data visualisation and effective dashboard creation!

Importance of Data Visualisation

Data visualisation plays an important role in helping people make sense of complex data in terms of delivering cohesive stories. It enables us to communicate insights and trends more effectively, allowing us to spot patterns, identify outliers, and draw conclusions that might not be immediately apparent from raw data. By presenting data visually, we can quickly and easily grasp large amounts of information, gain insights, and make informed decisions. Data visualisation can also make data more accessible and engaging to a wider audience, from C-levels to executives, by using compelling visual representations of data that are easy to understand. Therefore, effective data visualisation is essential for businesses, researchers, and individuals to communicate their data-driven insights and findings with clarity and impact.

Data Ink Ratio Concept

First of all, I would like to mention our brain’s science of communication. Unfortunately, our brains are not good at comparing areas. That’s why we have to maximise the data-pixel ratio. The data pixel ratio is calculated by dividing the data information pixel (pixels that represent actual data or information) by all non-background pixels ( every pixel that is not simply the background colour).

Edward Tufte introduced the Data-Ink ratio in his book, The Visual Display of Quantitative Data. He is an expert whose work has significantly contributed to the design of effective data presentations. You can get more info here .

A graph with too many unnecessary and distracting components is deemed to have low data-ink.

When a graph uses the fewest visuals to represent data, it is said to have a high data-ink. That’s why we need to maximise the data-pixel ratio.

An example is presented below, in which all distracting components have been removed so that the visualisation is clear and the attention is only on the data.

One of the most crucial aspects is our horizontal and vertical lines. We have to bold our line chart to parse from background lines.

As Cole Nussbaumer Knaflic said that "clutter is our enemy".

Smart Colour Selection

Sometimes colours can be unnecessary :)

• When we’re creating graphs, we can use double colours such as high contrast and low contrast colour.

• Purple and grey can be used for vital parts. After creating a graph, you can colour the parts.
• You have to be cognizant of colour blindness. You can avoid red and greens due to their connotations.

Types of Data Visualisation Charts

Pie Charts

Pie charts are most effective when only two items are included, such as the percentage of new users and remarketing customers.

Here are some cons of pie chart visualisation:

• Some pie chart areas can be displayed very small. It’s hard to read for interpretation.
• The total ratio sometimes isn’t completely 100%.
• Similar areas sometimes cannot be displayed equally while we’re comparing.
• Pie charts are usually not a good way to present our data.

Horizontal Bar Charts

• It’s an easy way to interpret complex data.

• Ideal for comparing categories.

• Sometimes vertical bar charts can be tricky. (be careful when you’re using)

• Horizontal bar charts support easy comparisons of multiple metrics.

• Ensuring the width of the bar is greater than the space between the bars.

• Avoid multiple series in a single bar chart.

*A vertical bar chart makes it difficult to distinguish between declining and growing values.

Line Charts for Time Series

• It’s best for showing trends over time and cronical stories.
• Avoid using a line chart to use categorical data. You can use a bar chart for categorization as I mentioned above.
• Don’t show more than 5 lines on a single chart.
• Avoid labelling every data point, you can label if there are specific points.

Sparklines and Small Multiples

You can show limited details with sparklines. It can be used for periods and trends presentation.

Text as a Visualisation

Primary text can be 16 punto while secondary text 14 punto as bold. I recommend using this small text to show percentages, it is a handy type of visualisation.

Heatmaps

Heatmaps contain two dimensions and one metric area. Heatmaps can reinforce the detail in the numbers. Please, remember the warning about red-green blindness. Blue to orange or yellow can be used for colouring. Heatmaps can work on maps but we have to be careful when we’re setting.

Scatterplots

Zero-based axes and scatterplots can be tricky :) It shows us the relationship between the two metrics. If you are using standard charts you can gain cognitive load. The non-standard chart should only be used when they’re needed.

Here's a more comprehensive template to explore what kind of chart you need:

Key Takeaways:

• Maximise Data Pixel Ratio
• Use proper colours
• Select graphs wisely according to your aims
• Use storytelling

Dashboard Considerations

When creating your dashboard, you should have a strong knowledge of your data and your KPIs.

Some tips for effective dashboards:

• Simplicity and clarity are essential.
• KPIs should be centred.
• Reports can be delivered as scheduled and on demand.
• It contains a source for all the answers.

BI Tools for Data Visualisation

Excel: It is a basic tool that you can use for auto tables and graphs. Probably, you are using Excel every day :)

Looker Studio: It is Google’s data visualisation tool where you can integrate many data sources such as Google Analytics, GSC, and Google Ads. There are many templates that you can use but if you ask me, I will recommend beginning with a plain template and examining what you want exactly to see, as always I do.

Tableau: Tableau allows you to conduct faster analyses and create interactive dashboards. You can connect many types of data sources. In Tableau Public , you can find various templates for everything. I think Tableau is my favourite tool :) I will write a special episode for Tableau in detail, so stay tuned.

PowerBI: It is a Microsoft data visualisation tool. It provides interactive visualisations and business intelligence capabilities for users to create their reports and dashboards like Tableau.

For now, I'll see you in the next episode... oops, I mean I hope to see you at my #Measurefast talk next time! Let's meet at 11:10 AM on 4th October . I'll tell you all about data storytelling and Tableau.

Continue exploring All About Digital Marketing & Data Analytics and don't forget to subscribe to stay updated on the latest information. Cheers!